research in the schools - fall 2012
DESCRIPTION
Fall 2012 edition of MSERA's Research in the Schools journalTRANSCRIPT
Anthony J. Onwuegbuzie and John R. Slate, Editors
Volume 19, Number 2 Fall 2012 Editorial: Evidence-based Guidelines for Avoiding Reference List Errors in Manuscripts Submitted to Journals for Review for Publication: A Replication Case Study of Educational Researcher ............................................................................................................................... i - xvi
Anthony J. Onwuegbuzie, Eunjin Hwang, Julie P. Combs, and John R. Slate Foreign Language Immersion Programs and School Policy: Conflicting Agendas ................. 1-16
Heather K. Olson Beal, Michelle Haj-Broussard, and Nicole Boudreaux Factors Predicting Pre-service Teachers’ Adoption of Web 2.0 Technologies ....................... 17-29
Jongpil Cheon, Fanni Coward, Jaeki Song, and Sunho Lim The Revised School Culture Elements Questionnaire: Gender and Grade Level Invariant? .............................. 30-44
Thomas A. DeVaney, Nan B. Adams, Flo Hill-Winstead, and Mitzi P. Trahan Examining Variations in Programmatic Delivery on Teaching Candidates’ Sense of Efficacy .................................................................................................................................................. 45-61
S. Michael Putman Research Teaching Pedagogy: Lessons Learned From an Arabic Language Intervention Pilot Study ........................................................................................................................................ 62-74
Thomas W. Christ and Hosam Elmetaher Unpacking the Effects: Identifying School and Teacher Factors and Their Influence on Teachers’ Intentions to Stay or Leave the Profession ............................................................................. 75-89
Amy L. Sedivy-Benton and Carrie J. Boden-McGill
RESEARCH IN THE
SCHOOLS
A nationally refereed journal sponsored by the Mid-South Educational Research Association and
Sam Houston State University
MSERA
Research in the Schools Information for Authors
Statement of Purpose RESEARCH IN THE SCHOOLS (ISSN 1085-5300) is a nationally/internationally refereed journal co-sponsored by the Mid-
South Educational Research Association and Sam Houston State University. The journal publishes original contributions
in the following areas: (1) Research in Practice--empirical studies focusing on the results of applied educational research
including cross-cultural studies, (2) Topical Articles--scholarly reviews of research, perspectives on the use of research
findings, theoretical articles, and related articles, (3) Methods and Techniques--descriptions of technology applications in
the classroom, descriptions of innovative teaching strategies in research/measurement/statistics, evaluations of teaching
methods, and similar articles of interest to instructors of research-oriented courses, (4) Assessment--empirical studies of
norm-referenced, criterion-referenced, and informal tests in the areas of cognitive ability, academic achievement,
personality, vocational interests, neuropsychological functioning, and the like, (5) Educational Policy, Reform, and
Accountability–empirical studies or conceptual pieces focusing on educational policy, school and educator accountability,
politics of education, or educational responses to reform mandates, and (6) Other topics of interest to educational
researchers. RESEARCH IN THE SCHOOLS is devoted to research conducted in any educational setting from a conventional
elementary school or high school to a training program conducted within an industry. Likewise, there are no age
restrictions on the sample, since the educational settings may include preschools, continuing education classes for adults,
or adaptive skills courses in nursing homes. Studies conducted in settings such as clinics, hospitals, or prisons are
ordinarily inappropriate for RESEARCH IN THE SCHOOLS unless they involve an educational program within such a setting.
One goal of RESEARCH IN THE SCHOOLS is to provide a training ground for graduate students to learn effective reviewing
techniques. Consequently, the journal utilizes a Graduate Student Editorial Board composed mostly of students in
educational psychology and educational research. Members of this Editorial Board, each sponsored by a professor,
provide supplementary reviews for a selection of submitted articles, and receive both direct and indirect feedback of the
quality of these reviews.
Submission of Manuscripts Manuscripts should be submitted electronically to Anthony J. Onwuegbuzie, Co-Editor, RESEARCH IN THE
SCHOOLS. Please direct questions via email to [email protected]. All manuscripts will undergo a rigorous
internal review process to ensure that the manuscript is within the scope of RITS, is of sufficient quality, and conforms to
sixth edition APA style guidelines. Only manuscripts that satisfy all of the following requirements will be sent out for
external review:
Manuscripts that have been published elsewhere in part or in whole will not be considered for publication in
RITS. We will not consider any manuscripts that have been published in such arenas as conference
proceedings, yearbooks, online sources, journals, and book chapters. Only original unpublished works will be
considered for review in RITS. Manuscripts of which only the abstract has been published previously may be
submitted for review.
Length of the manuscripts, including references and tables, should ordinarily range from about 10 to 40 typed,
double-spaced, 8-1/2 X 11-inch pages, using 12-point type.
Abstracts are limited to 120 words.
The title page should provide title, names, and affiliation of all authors, and contact information of one
corresponding author.
Tables should be created using the autotable format (using the "Table" menu in Word) such that no column
lines (i.e., vertical lines) are used in the tables; and only row lines (i.e., horizontal lines) are used at the top,
bottom, and heading of the table (cf. APA, 2010, pp. 127-150).
Tables and figures should not be embedded within the body of the manuscript; rather, all tables should appear
(one table on each page) after the reference list, followed by all figures (one figure per page) (cf. APA, 2010,
pp. 127-150).
Brief reports of research are not encouraged.
All manuscripts submitted for review should be original material, not published in part or in whole in any other
publication. Moreover, no manuscripts submitted for review should be under review at any other journal.
Authors are encouraged to keep an electronic copy of the manuscript to guard against loss. The editorial team will "blind"
all manuscripts before sending them for external review. Review time is 3-4 months.
Copyright and Permissions Authors are granted permission to reproduce their own articles for personal use. Others must request permission to
reproduce tables, figures, or more than 500 words of text from the editors. Copyright © 2012 by the Mid-South
Educational Research Association.
RESEARCH IN THE SCHOOLS
Volume 19, Number 2 Fall 2012
EDITORS
Anthony J. Onwuegbuzie and John R. Slate
Sam Houston State University
COPY EDITOR Gail H. Hughes
University of Arkansas, Little Rock
EDITORS EMERITI
James E. McLean
The University of Alabama
Alan S. Kaufman
Yale University School of Medicine
PRODUCTION EDITOR
Eunjin Hwang
Sam Houston State University
EDITORIAL BOARD
Amy B. Dellinger, D.A.T.A., LLC
Angela Gibson, American Public University System
Bethany Bell, University of South Carolina
Bruce Thompson, Texas A & M University and Baylor
College of Medicine (Houston)
Cecil R. Reynolds, Texas A & M University
Charles M. Achilles, Seton Hall University and Eastern
Michigan University
Charles McLafferty, Purpose Research, LLC
Cindy Benge, Sam Houston State University
Claire Meljac, Hospital Sainte-Anne (France)
Daniel Krenzer, Mesa County Valley School District
David Segal, University of Central Florida
David T. Morse, Mississippi State University
Debbie Hahs-Vaughn, University of Central Florida
Dianne Swain, The University of West Alabama
Dimiter Dimitrov, George Mason University
James M. Ernest, University of Alabama at Birmingham
Jane Nell Luster, Louisiana State University
Janet C. Richards, University of South Florida
Jennifer Styron, University of South Alabama
Jennifer Purcell, Touro University
Jerrid Freeman, University of Arkansas
Jim Flaitz, University of Louisiana
Julie P. Combs, Sam Houston State University
Jwa K. Kim, Middle Tennessee State University
Kathleen Campbell, Southeastern Louisiana University
Kathleen M. T. Collins, University of Arkansas at
Fayetteville
Keith Hyatt, Western Washington University
Kevin Kieffer, Saint Leo University
Linda Zientek, Sam Houston State University
Lynn Loftin, Southern University
Margaret L. Rice, The University of Alabama
Mark Baron, University of South Dakota
Michelle Brown, Texas A&M, Kingsville
Michael D. Richardson, Southeastern Louisiana
University
Nancy L. Leech, University of Colorado Denver
Nataliya V. Ivankova, University of Alabama at
Birmingham
Paula Stanley, Radford University
Qaisar Sultana, Eastern Kentucky University
Randy Parker, Louisiana Tech University
R Burke Johnson, University of South Alabama
Robert Kennedy, University of Arkansas for Medical
Sciences
Ronald Styron, The University of South Alabama
Russell Mays, Georgia Southern University
Sandra Acosta, Texas A&M University
Shujie Liu, Northeast Normal University, China
Susan Troncoso Skidmore, Sam Houston State University
Thomas W. Christ, University of Bridgeport-Carlson
Tracy Goodson-Espy, Appalachian State University
William A. Spencer, Auburn University
GRADUATE STUDENT EDITORIAL BOARD Brandi Van Horn, University of Denver
Christina Hagerty, Sam Houston State University
Christy Wilmore, Sam Houston State University
Heather Elliot, University of Michigan
Megan Jones, Sam Houston State University
Robert Horton, Sam Houston State University
Copyright 2012 by the RESEARCH IN THE SCHOOLS
Mid-South Educational Research Association 2012, Vol.19, No. 2, i-xvi
Fall 2012 i RESEARCH IN THE SCHOOLS
Editorial: Evidence-based Guidelines for Avoiding Reference List Errors in Manuscripts
Submitted to Journals for Review for Publication: A Replication Case Study of Educational
Researcher
Anthony J. Onwuegbuzie, Eunjin Hwang, Julie P. Combs, and John R. Slate
Sam Houston State University
In a previous editorial, Onwuegbuzie,
Hwang, Frels, and Slate (2011) reported the
findings of Onwuegbuzie and Hwang (2012) (see
also, Onwuegbuzie & Hwang, 2013), who
examined the frequency and characteristics of
violations to the American Psychological
Association (APA) style guide (APA, 2010) that
occurred in the reference lists of manuscripts
(hereafter referred to as reference list errors)
initially submitted (i.e., unpublished manuscripts)
to Research in the Schools (RITS), a
nationally/internationally refereed journal, as well
as the relationships between reference list errors
and selected manuscript variables (e.g., number of
authors, editor decision [e.g., reject, accept]).
Onwuegbuzie and Hwang (2012, 2013) used mixed
research techniques to examine 131 manuscripts
submitted to RITS over a 6-year period (i.e., 2004-
2010). These researchers identified a total of 466
unique reference list errors that were identified
across these 131 manuscripts, with the prevalence
of each of these reference list errors ranging from 1
(0.75%) to 102 (76.7%). Also, Onwuegbuzie and
Hwang (2012, 2013) documented an average of 12
reference list errors per manuscript (M = 12.83, SD
= 7.25).
Further, Onwuegbuzie and Hwang (2012,
2013) conducted a qualitative analysis (i.e.,
constant comparison analysis; Glaser, 1965; Glaser
& Strauss, 1967) of the identified 466 reference list
errors that yielded the following 14 reference list
error themes: (a) General errors; (b) Reference
heading, (c) Names of authors, (d) Publication
year/date, (e) Title of work, (f) Publisher
information, (g) Source of journal/periodical, (h)
Source of authored book, (i) Source of edited book,
(j) Source of website, (k) Source of paper
presentation, (l) Source of dissertation/thesis, (m)
Correspondence concerning this article should be
addressed to Anthony J. Onwuegbuzie, Department
of Educational Leadership and Counseling, Box
2119, Sam Houston State University, Huntsville,
Texas 77341-2119, or E-Mail:
Onwuegbuzie, Hwang, Frels, and Slate (2011) reported the findings of Onwuegbuzie and Hwang
(2012, 2013), who examined the frequency and characteristics of violations to the American
Psychological Association (APA) style guide (APA, 2010) that occurred in the reference lists of 131
manuscripts submitted to Research in the Schools over a 6-year period. Findings revealed that authors
committed more than 12 reference list errors per manuscript, on average (M = 12.83, SD = 7.25).
Further, a total of 466 unique reference list errors were identified, which yielded 14 reference list
error themes. However, it could be questioned whether the same reference list error rates would be
observed among manuscripts submitted to Tier I journals. Consequently, in the present editorial, we
replicate and extend their work by using mixed analysis techniques to examine the reference list error
rate of 83 manuscripts submitted to the highest ranked educational journal, Educational Researcher,
over a 3.5-year period. Findings revealed that Educational Researcher authors committed more than
14 reference list errors per manuscript, on average (M = 14.25, SD = 8.05). Also, a total of 324
unique reference list errors were identified, which yielded the same aforementioned 14 reference list
error themes. A multiple regression analysis revealed that every author of a manuscript was
associated with an increase of 2.40 reference list errors, on average. These findings support the
hypothesis that journals with the highest impact factors also have manuscripts submitted to their
journals that have high rates of reference list errors. Implications are discussed.
EDITORIAL: EVIDENCE-BASED GUIDELINES FOR AVOIDING REFERENCE LIST ERRORS IN
MANUSCRIPTS SUBMITTED TO JOURNALS FOR REVIEW FOR PUBLICATION: A REPLICATION
CASE STUDY OF EDUCATIONAL RESEARCHER
Fall 2012 ii RESEARCH IN THE SCHOOLS
Source of newspaper article, and (n) Source of
government document. An exploratory factor
analysis of these 14 reference list error themes led
to the identification of four meta-themes that
contained between two and five reference list error
themes.
Onwuegbuzie and Hwang (2012, 2013) also
conducted a latent class analysis of the six
reference list error themes that were committed by
the majority (i.e., > 50%) of the 131 authors, which
revealed two distinct clusters of manuscripts, with
one cluster (comprising 57.1% of the manuscripts)
being relatively high with respect to all six
reference list error themes, and the other cluster
(comprising 42.9% of manuscripts) being high on
two reference list error themes and low on the
remaining four reference list error themes.
Finally, Onwuegbuzie and Hwang (2012,
2013) conducted three sets of quantitative analyses,
namely, a correlation analysis, independent
samples t test, and canonical correlation analysis.
Specifically, with respect to the correlation
analysis, a series of nonparametric (i.e., Spearman)
correlations, after applying the Bonferroni
adjustment to control for the inflation of Type I
error, revealed that the number of reference list
errors was statistically significantly and positively
related to (a) the number of citation errors (rs[131]
= .39, p < .001), suggesting a moderate-to–large
relationship (Cohen, 1988); and (b) the length of
manuscript (rs[131] = .23, p < .001), suggesting a
small-to-moderate relationship (Cohen, 1988). The
independent samples t test revealed that
manuscripts that were not accepted by the editor
(M = 13.70, SD = 7.43)—that is, they were either
rejected or received a revise-and-resubmit
decision—contained statistically significantly more
reference list errors than did manuscripts that were
accepted (M = 9.44, SD = 4.95), with a large effect
size of 0.83 (Cohen, 1988).
A canonical correlation analysis revealed a
multivariate relationship between the 14 reference
list error themes and selected demographic
variables. Specifically, this multivariate
relationship was mainly characterized by the
relationship between reference list errors associated
with publisher information, source of
dissertation/thesis, and source of edited book on the
one side, and number of authors and length of
manuscript on the other side.
However, although some of the authors who
publish articles in RITS are among the most
prolific, it is likely that a much greater proportion
of prolific authors submit their manuscripts to
journals with the highest impact factors. And,
assuming that prolific authors tend to be the most
experienced authors, it is reasonable to hypothesize
that journals with the highest impact factors would
have manuscripts submitted to their journals that
have a significantly lower rate of reference list
errors. Yet, to date, this hypothesis has not been
tested empirically. Consequently, in this editorial,
we replicate and extend the works of Onwuegbuzie
and Hwang (2012, 2013) and Onwuegbuzie,
Hwang, et al. (2011) by examining the frequency
and characteristics of reference list errors among
manuscripts submitted to a top Tier I educational
journal.
Sources of Evidence
In order to examine the frequency and
characteristics of reference list errors among
manuscripts submitted initially to a top Tier I
educational journal, we conducted a mixed research
study of 83 manuscripts submitted to Educational
Researcher over a period of 3.5 years. The journal
Educational Researcher was selected because not
only does it represent the premier flagship journal
of the American Educational Research Association
but also it represents the educational journal with
the highest impact factor—specifically, its impact
factor of 3.774 makes it the highest ranking among
177 journals representing education and education
research. Two of the authors of this editorial were
part of the editor team (i.e., editor and associate
editor) of Educational Researcher (2006-2010) that
secured this extremely high impact factor. As such,
they had complete access to every manuscript
submitted to Educational Researcher during this
period. The 83 manuscripts selected for study
represented those manuscripts that were submitted
for the first time to the Research News and
Comment section of Educational Researcher—one
of two sections at that time (with the other section
being called Features that was co-edited by Drs.
Patricia B. Elmore and Gregory Camilli). Further,
these 83 manuscripts represented those manuscripts
that had not undergone what is referred to as a desk
reject or internal rejection during the internal
review process (i.e., before the manuscript is sent
out for external review, the manuscript was deemed
inappropriate for Educational Researcher because
it had a focus or content that was outside the scope
of the journal [e.g., the topic did not pertain to an
educational issue]; did not follow adequately the
stipulated format for manuscripts [e.g., the
manuscript resembled more of a traditional
empirical report rather than an essay]; or the
manuscript was written in a style that was not
adequately consistent with APA [e.g., the
manuscript followed Chicago Manual of style;
Chicago Manual, 2003; no discernible style]). In
other words, each of the 83 manuscripts had
satisfied the criteria for being sent out for external
review. These 83 manuscripts represented 49.11%
ANTHONY J. ONWUEGBUZIE, EUNJIN HWANG, JULIE P. COMBS, AND JOHN R. SLATE
Fall 2012 iii RESEARCH IN THE SCHOOLS
of all manuscripts submitted to the Research News
and Comment section of Educational Researcher
over this period, which rendered our findings
generalizable to the population of manuscripts
submitted to Educational Researcher—at least over
this period of time.
The two editors of the Research News and
Comment section of Educational Researcher
meticulously documented every APA error—
including reference list errors—committed by these
83 sets of authors over the 3.5-year period. Each
manuscript took up to 8 hours to identify all the
APA errors—representing as much as 664 hours of
coding. Alongside collecting information about
each APA error, these editors collected an array of
information corresponding to each of these
manuscripts, including the following: the length of
the manuscript (i.e., number of pages, number of
words), the length of the reference list (i.e., total
number of references), topic of the manuscript, and
the number of authors per manuscript. As such, the
data set created by these editors was extremely
rich, representing a data set that only journal
editors have the opportunity to develop.
Methodology
Dialectic pluralism was the philosophical
lens that drove our mixed research study. As
conceptualized by Johnson (2012), dialectic
pluralism represents a philosophical stance
wherein multiple epistemological perspectives are
combined by the researcher(s) within the same
investigation. With respect to the data analysis
step of our mixed research study, we utilized a
sequential mixed analysis (Onwuegbuzie &
Combs, 2010). Specifically, because our present
study represented a replication of the studies of
Onwuegbuzie and Hwang (2012, 2013) and
Onwuegbuzie, Hwang, et al. (2011), we used a
five-stage sequential mixed analysis procedure.
Each of these stages is described below.
Stage 1 Analysis
The first stage of the sequential mixed
analysis involved conducting a classical content
analysis (Berelson, 1952; see also Leech &
Onwuegbuzie, 2007, 2008, 2011) of the 83
manuscripts to determine the number of unique
reference list errors. In addition, the total number
of reference list errors per manuscript was
identified. Thus, the classical content analysis led
to the determination of prevalence rates.
Stage 2 Analysis
Determination of the reference list errors led
to the second stage. This stage involved conducting
a constant comparison analysis (Glaser, 1965;
Glaser & Strauss, 1967) of the reference list errors
to determine the number of themes that could be
extracted from them. These themes were extracted
iteratively (Constas, 1992). That is, these themes
were extracted using both a priori coding and a
posteriori coding. The a priori coding involved the
potential use of the aforementioned 14 themes that
stemmed from the study of Onwuegbuzie and
Hwang (2012, 2013). In addition, the researchers
allowed for the possibility of other themes
emerging.
Stage 3 Analysis
The third stage of the sequential mixed
analysis involved quantitizing the reference list
error themes (Sandelowski, Voils, & Knafl, 2009;
Tashakkori & Teddlie, 1998). This quantitizing
involved converting the themes that were extracted
in the second stage (i.e., qualitative data) to
numerical codes (i.e., quantitative data) for
statistical analyses. Specifically, each reference list
error theme was converted to a quantitative code by
assigning a “1” if the manuscript contained one or
more reference list error errors that were classified
under that theme and a “0” if the manuscript did
not contain any reference list error errors that were
classified under that theme (Onwuegbuzie, 2003;
Onwuegbuzie & Teddlie, 2003)—leading to the
creation of what Onwuegbuzie (2003) referred to as
an “inter-respondent matrix” (i.e., manuscript x
reference list error theme matrix) that comprised a
combination of 0s and 1s (p. 396). This inter-
respondent matrix was used to conduct a principal
component analysis (cf. Field, 2009) to determine
the underlying structure of the reference list error
themes by transforming it to a matrix of bivariate
associations that represented tetrachoric correlation
coefficients to take into account the fact that the
reference list error themes had been quantitized to
dichotomous data (i.e., “0” vs. “1”). As noted by
Onwuegbuzie et al. (2007), tetrachoric correlation
coefficients are appropriate to use when examining
the association between two (artificial)
dichotomous variables. In addition, a varimax (i.e.,
orthogonal) rotation was employed (Field, 2009),
using the following three procedures to determine
an appropriate number of factors to retain:
eigenvalue-greater-than-one rule (i.e., K1; Kaiser,
1958), scree test (representing a plot of eigenvalues
against the factors in descending order; Cattell,
1966; Zwick & Velicer, 1986), and a parallel
analysis (involving extracting eigenvalues from
random data sets that parallel the actual data set
with respect to the sample size and number of
variables; Thompson, 2004; Zwick & Velicer,
1982, 1986). These extracted factors yielded meta-
themes (Onwuegbuzie, 2003) such that each meta-
theme contained one or more of the reference list
error themes that emerged in the second stage of
the mixed analysis. As described by Onwuegbuzie
(2003), the proportion of variance explained by
each factor after rotation (i.e., the trace) served as
EDITORIAL: EVIDENCE-BASED GUIDELINES FOR AVOIDING REFERENCE LIST ERRORS IN
MANUSCRIPTS SUBMITTED TO JOURNALS FOR REVIEW FOR PUBLICATION: A REPLICATION
CASE STUDY OF EDUCATIONAL RESEARCHER
Fall 2012 iv RESEARCH IN THE SCHOOLS
an effect size index for each meta-theme. Using
Constas’s (1992) typology, by determining the
hierarchical relationship among the reference list
themes and identifying the meta-themes, the
verification component of categorization was
technical, empirical, and rational.
Stage 4 Analysis
The fourth stage of the sequential mixed
analysis process involved conducting a latent class
analysis to ascertain the number of clusters or
groups (i.e., latent classes) underlying the reference
list error themes. The latent class analysis was
conducted under the assumption that the 83
manuscripts could be classified into a small number
of distinct clusters known as latent classes based on
their profiles of reference list error themes, such
that each manuscript belonged to only one cluster.
This latent class analysis represented qualitizing of
the data, which involves converting numeric data
into (qualitative) narrative profiles (Tashakkori &
Teddlie, 1998).
Stage 5 Analysis
The fifth stage of the sequential mixed
analysis involved using the inter-respondent matrix
to examine the relationship between the reference
list error themes and selected variables.
Specifically, we conducted the following analyses:
(a) a series of correlation analyses to investigate the
relationships between the total number of reference
list errors and the number of citation errors, number
of authors, and length of manuscript; and (b) a
multiple linear regression analysis to identify an
optimal combination of independent variables (i.e.,
number of references in the reference list, number
of manuscript pages, number of manuscript words,
number of authors) that predicted the number of
reference list errors. Further, we were interested in
conducting a canonical correlation analysis to
examine the multivariate relationship between the
reference list error themes and selected
demographic variables (i.e., gender of the lead
author, number of authors, length of manuscript,
and size of institution of the lead author). However,
because of the relatively small case-to-variable
ratio (i.e., 83 cases to 14 reference list error themes
and 4 demographic variables), this analysis lacked
sufficient statistical power. Also of interest was to
examine whether the number of reference list errors
predicted the decision made by the editor regarding
the suitability of the manuscript for publication.
However, because the Educational Researcher
editors accepted only one of these 83 manuscripts
the first time, it was not possible to examine the
relationship between the number of reference list
errors and the decision made by the editors.
Results
Stage 1 Findings
The classical content analysis (Berelson,
1952) revealed a total of 1,183 reference list errors
across the 83 manuscripts, which represented more
than 14 reference list errors per manuscript, on
average (M = 14.25, SD = 8.05). The number of
unique reference list errors per manuscript ranged
from 3 to 35, with 83.1% of manuscripts containing
more than five unique reference list errors, 60.2%
of manuscripts containing more than 10 unique
reference list errors, and 24.1% of manuscripts
containing more than 20 unique reference list
errors. The classical content analysis also led to the
identification of a total of 324 unique reference list
errors that were identified across these 83
manuscripts. Additionally, this analysis revealed
that the prevalence of each of these reference list
errors ranged from 0 to 58.
Because of the number of unique reference
list errors identified (i.e., n = 324), we decided to
deem an error as being significantly common when
it occurred a minimum of five occasions. The cut-
point of five was used because it represented an
endorsement rate of 6%, which translated to a
moderate effect size, using Cohen’s (1988, pp. 180-
183) non-linear arcsine transformation criteria.
Interestingly, a total of 60 (18.5%) reference list
errors yielded moderate effect sizes (i.e.,
endorsement rates of five or greater). Table 1
presents the 50 most prevalent reference list errors
documented by Onwuegbuzie and Hwang (2012,
2013) and the ranks of these frequencies, as well as
the frequencies and ranks pertaining to these
reference list errors stemming from the present
study. As such, Table 1 provides a direct
comparison of the frequencies between
Onwuegbuzie and Hwang’s (2012, 2013)
investigation and the present study. It can be seen
from this table that the same two most common
reference list errors were the two most frequent
errors in both studies. In fact, three of the top four
reference list errors yielded identical ranks in both
studies, namely: Serial (issue) numbers presented
when the page numbers in each volume are
continuous; Comma not presented to separate two
authors; and Space not presented between initials
of each author.
ANTHONY J. ONWUEGBUZIE, EUNJIN HWANG, JULIE P. COMBS, AND JOHN R. SLATE
Fall 2012 v RESEARCH IN THE SCHOOLS
Table 1
Stage 1 Findings: The 50 Most Prevalent Reference List Errors in Present Study and Onwuegbuzie and
Hwang’s (2012, 2013) Study and the Corresponding Rank in Present Study
Reference List Error Frequency1
Rank2 Frequency
3 Rank
4
Serial (issue) numbers presented when the
page numbers in each volume are continuous
102
1
58
1
Comma not presented to separate two authors
56 2 48 2
Superscripts inappropriately used when
providing edition number
53
3
33
6
Space not presented between initials of each
author
49
4
38
4
Period not presented after the author’s name
(when the author does not represent a person
but an organization) and before the publication
year
37
5
19
13
Website inappropriately underlined 34
6 23 9
Month not given for a paper presentation
31 7 10 30
"Publications” or “Publications Inc"
inappropriately presented when listing the
publisher
30
8
13
20
Reference list not double spaced 28
9 36 5
Citations not presented in alphabetical order
27 10 11 28
Title of journal article inappropriately
capitalized
27
10
39
3
Comma not presented after retrieval year of
Internal source
25
12
21
12
Volume number not italicized 24
13 26 7
"Inc" inappropriately presented when listing
the publisher
21
14
13
20
Title of book inappropriately capitalized
20 15 15 16
"&" not used to separate the last two authors
18 16 9 34
Reference heading is bolded 18
16 25 8
Retrieval date not provided for web-based
citations
18
16
13
20
First letter of the second-part of the title not
capitalized
18
16
14
19
Title of edited books inappropriately
capitalized
17
20
9
34
EDITORIAL: EVIDENCE-BASED GUIDELINES FOR AVOIDING REFERENCE LIST ERRORS IN
MANUSCRIPTS SUBMITTED TO JOURNALS FOR REVIEW FOR PUBLICATION: A REPLICATION
CASE STUDY OF EDUCATIONAL RESEARCHER
Fall 2012 vi RESEARCH IN THE SCHOOLS
Reference List Error Frequency1
Rank2 Frequency
3 Rank
4
Title of journal not italicized 16
21 3 89
Space not presented to separate initials of each
editor of an edited book
15
22
15
16
"And" instead of “&” to separate the last two
authors
14
23
13
20
City, state, and/or publisher not always
provided
14
23
11
28
Title of book not always italicized 14
23 5 55
Period inappropriately appears after the
numbers of ERIC
14
23
1
236
Page number of book chapters not presented
after the title of the book
12
27
23
9
Space not presented between pp and the page
number
12
27
4
69
Page number of journal articles not presented
12 27 7 37
Initials of all authors not presented 11
30 3 89
State pertaining to the publisher not
abbreviated
11
30
2
124
Reference heading represented by all
uppercase text
11
30
10
30
Serial number not presented when
discontinuous when the page numbers in each
volume are not continuous
11
30
6
38
Volume number of journal article not provided
11 30 12 25
Comma not presented to separate the last two
authors of a reference
(when references have more than two authors)
10
35
10
30
Period not presented after an author's initial
10 35 18 15
Volume number of journal (periodicals) not
italicized
10
35
22
11
Abbreviation (of authors) inappropriately
included
9 38 1 236
Citations not presented in chronological order
9
38
8
36
Title of paper presentation not italicized
9
38
0
323
Period inappropriately presented at the end of
the reference (e.g., when the reference ends
with a website address)
9
38
0
323
ANTHONY J. ONWUEGBUZIE, EUNJIN HWANG, JULIE P. COMBS, AND JOHN R. SLATE
Fall 2012 vii RESEARCH IN THE SCHOOLS
Reference List Error Frequency1
Rank2 Frequency
3 Rank
4
Title of edited book not italicized 9
38 2 124
Title of journal article inappropriately
italicized
9
38
1
236
Period not presented at the end of reference
8 44 19 13
Reference list does not begin on a separate
page
8
44
12
25
Comma inappropriately appears between
initials of some authors
8
44
15
16
State of publisher not provided 8
44 12 25
"And" instead of “&” used to separate the
editors of edited books
8
44
10
29
Space inappropriately appears between six
numbers of an ERIC
8
44
1
236
Space inappropriately appears between volume
number and series number of a periodical
8
44
13
20
1 Represents data from Onwuegbuzie and Hwang’s (2012, 2013) study; Frequencies between 8 and 21 represent
moderate effect sizes; frequencies greater than 22 represent large effect sizes, using Cohen’s (1988, pp. 180-
183) non-linear arcsine transformation criteria. 2 Represents data from Onwuegbuzie and Hwang’s (2012, 2013) study
3 Represents data from the present study; Frequencies between 5 and 13 represent moderate effect sizes;
frequencies greater than 13 represent large effect sizes, using Cohen’s (1988, pp. 180-183) non-linear arcsine
transformation criteria. 4 Represents data from the present study
Stage 2 Findings
A constant comparison analysis of these 324
reference list errors yielded the following 14
reference list error themes: (a) General errors; (b)
Reference heading, (c) Names of authors, (d)
Publication year/date, (e) Title of work, (f)
Publisher information, (g) Source of
journal/periodical, (h) Source of authored book, (i)
Source of edited book, (j) Source of website, (k)
Source of paper presentation, (l) Source of
dissertation/thesis, (m) Source of newspaper article,
and (n) Source of government document. These
reference list error themes were identical to the
ones identified by Onwuegbuzie and Hwang (2012,
2013). Table 2 presents descriptive statistics
regarding the number of reference list errors for
each of the 14 citation error themes. It can be seen
from this table that reference list errors associated
with the Source of journal/periodical represented
the most prevalent errors, followed by reference list
errors associated with Names of authors.
Stage 3 Findings
With regard to the number of factors
underlying the 14 emergent reference list error
themes extracted in Stage 2, the Kaiser-Meyer-
Olkin (KMO) measure of sampling adequacy was
greater than .5 (i.e., KMO = .51) and Bartlett’s test
of sphericity was statistically significant (Χ2[91] =
128.45, df = 91, p = .006), thereby justifying
conducting a principal component analysis. Both
the eigenvalue-greater-than-one rule (i.e., K1;
Kaiser, 1958) and scree test indicated that four
factors (i.e., meta-themes) should be retained. A
parallel analysis verified the K1 and scree test
(Zwick & Velicer, 1982, 1986) for the current data
of 83 manuscripts and 14 variables (i.e., reference
list error themes). Specifically, a series of (i.e., n =
1,000) random data matrices of size 83 x 14 was
generated, and eigenvalues were computed for the
correlation matrices for the original data and for
each of the 1,000 random data sets. Next, the
eigenvalues derived from the actual data were
compared to the eigenvalues derived from the
EDITORIAL: EVIDENCE-BASED GUIDELINES FOR AVOIDING REFERENCE LIST ERRORS IN
MANUSCRIPTS SUBMITTED TO JOURNALS FOR REVIEW FOR PUBLICATION: A REPLICATION
CASE STUDY OF EDUCATIONAL RESEARCHER
Fall 2012 viii RESEARCH IN THE SCHOOLS
random data for the purpose of identifying the
number of components that accounted for more
variance than did the components obtained from
random data. Consistent with the K1 and scree test,
the parallel analysis suggested retaining four
factors.
Table 3 presents this four-factor principal
components solution. Using a cutoff correlation of
0.3 recommended by Lambert and Durand (1975)
as an acceptable lower bound for pattern/structure
coefficients, it can be seen from Table 3 that three
reference list error themes had pattern/structure
coefficients with large effect sizes on the first
factor: (a) Source of edited book, (b) Names of
authors, and (c) Source of authored book. It should
be noted that in addition to having a
pattern/structure coefficient with a large effect size
on Factor 1, Source of authored book also had a
significant but smaller pattern/structure coefficient
on Factor 4 (i.e., cross-loading). Further, the
following four reference list error themes had
pattern/structure coefficients with large effect sizes
on the second factor: (a) Source of website, (b)
Publication year/date, (c) Source of newspaper
article, and (d) Title of work. Interestingly, in
addition to having a pattern/structure coefficient
with a large effect size on Factor 2, Source of
newspaper article also had a significant but smaller
pattern/structure coefficient on Factor 4, and Title
of work also had a significant but smaller
pattern/structure coefficient on Factor 1 (i.e., cross-
loadings). Furthermore, the following four
reference list error themes had pattern/structure
coefficients with large effect sizes on the third
factor: (a) Source of dissertation/thesis, (b) Source
of government document, (c) Source of paper
presentation, and (d) Reference heading. In
addition to having a pattern/structure coefficient
with a large effect size on Factor 3, Source of paper
presentation also had a significant but smaller
pattern/structure coefficient on Factor 2 (i.e., cross-
loading). Finally, three reference list error themes
emerged that had pattern/structure coefficients with
large effect sizes on the fourth factor: (a) Publisher
information, (b) General errors, and (b) Source of
journal/periodical. In addition to having a
pattern/structure coefficient with a large effect size
on Factor 4, Reference heading also had a
significant but smaller pattern/structure coefficient
on Factor 3 (i.e., cross-loading).
Table 2
Stage 2 Findings: Prevalence Rates of Themes Emerging from Reference List Errors for Manuscripts Submitted
to Educational Researcher
Reference List Error Theme
Total Number of
Unique Reference
List Errors Contained
in Theme
Total Number
of Reference
List Errors
Contained in
Theme
Average incidence of
reference list errors per
manuscript (%)
Source of journal/periodical 65 233 96.4
Names of authors 39 209 83.1
General errors 19 92 68.7
Source of edited book 43 149 63.9
Source of authored book 27 85 62.7
Reference heading 13 78 62.7
Title of work 19 81 59.0
Source of website 23 86 51.8
Publisher information 24 69 48.2
Publication year/date 22 42 41.0
Source of paper presentation 22 48 27.7
Source of government document 5 5 6.0
Source of newspaper article 1 4 4.8
Source of dissertation/thesis 2 2 2.4
ANTHONY J. ONWUEGBUZIE, EUNJIN HWANG, JULIE P. COMBS, AND JOHN R. SLATE
Fall 2012 ix RESEARCH IN THE SCHOOLS
Table 3
Stage 3 Findings: Summary of Themes and Factor Pattern/Structure Coefficients from Principal Component
Analysis (Varimax): Four-Factor Solution
Factor Coefficients1
Theme 1 2 3 4 Communality
Coefficient
Source of edited book .77 .05 .02 -.10 .61
Names of authors .68 .10 .10 -.20 .52
Source of authored book .64 -.04 .02 .43 .60
Source of website -.09 .63 -.09 .09 .42
Publication year/date .17 .62 .12 .19 .46
Source of newspaper article -.01 .61 -.12 -.37 .52
Title of work .40 .54 .20 .21 .54
Source of dissertation/thesis .07 .01 .73 .01 .54
Source of government document .06 -.11 .72 -.10 .54
Source of paper presentation -.02 -.31 .55 .28 .48
Reference heading -.03 .06 -.39 .34 .27
Publisher information .18 -.10 -.11 .75 .61
General errors -.20 .17 -.09 .40 .24
Source of journal/periodical -.11 .16 .14 .38 .20
Trace 1.76 1.63 1.63 1.52 6.54
% variance explained 12.57 11.65 11.65 10.82 46.70
1Coefficients in bold represent pattern/structure coefficients with the largest effect size within each theme using
a cut-off value of 0.3 recommended by Lambert and Durand (1975).
The first meta-theme (i.e., Factor 1) was
labeled Author and Book and explained 12.57% of
the total variance; the second meta-theme (i.e.,
Factor 2) was labeled: Website, Year, Newspaper,
and Title and explained 11.65% of the total
variance; the third meta-theme (i.e., Factor 3) was
labeled: Dissertation/Thesis, Government, Paper
Presentation, and Heading and explained 11.65 of
the total variance; and the fourth meta-theme (i.e.,
Factor 4) was labeled Publisher Information,
Miscellaneous Errors, and Journal and explained
10.82% of the total variance. These four meta-
themes combined accounted for 46.70% of the total
variance, which using the findings of Henson,
Capraro, and Capraro (2004) and Henson and
Roberts (2006), represents a large effect size.
The manifest effect size—(i.e., actual
reference list error rate per meta-theme) associated
with the four meta-themes was as follows: (a)
Author and Book (91.6%), (b) Website, Year,
Newspaper, and Title (79.5%), (c)
Dissertation/Thesis, Government, Paper
Presentation, and Heading (74.7%), and (d)
Publisher Information, Miscellaneous Errors, and
Journal (98.8%). Figure 1 displays the thematic
structure (i.e., relationships among the reference
list error themes and the reference list error meta-
themes), including the manifest effect sizes and
latent effect sizes. This figure represents a
crossover visual representation (Onwuegbuzie &
Dickinson, 2008), which depicts the integration of
both quantitative and qualitative findings within the
same display.
Stage 4 Findings
A latent class analysis was conducted to
determine the smallest number of clusters (i.e.,
latent classes) that explains all the relationships
among select reference list error themes under the
assumption that manuscripts could be classified
into a small number of distinct clusters known as
latent classes depending on their profiles of the
select reference list errors, such that each
manuscript belonged to only one cluster. We
decided to conduct the latent class analysis on the
six most common error themes because for all of
these themes, at least 60% of the authors made
reference list errors that were classified under these
themes—namely, Source of journal/periodical,
Names of authors, General errors, Source of edited
EDITORIAL: EVIDENCE-BASED GUIDELINES FOR AVOIDING REFERENCE LIST ERRORS IN
MANUSCRIPTS SUBMITTED TO JOURNALS FOR REVIEW FOR PUBLICATION: A REPLICATION
CASE STUDY OF EDUCATIONAL RESEARCHER
Fall 2012 x RESEARCH IN THE SCHOOLS
book, Source of authored book, and Reference
heading, respectively.
Our latent class analysis of the six reference
list error themes revealed a two-cluster solution (L2
= 39.07, df = 50, p = .87, Bootstrap p = .13). Figure
2 displays these two distinct groups of manuscripts.
It can be seen from Figure 2 that Cluster 1
(comprising 66.0% of manuscripts) was relatively
high with respect to all six reference list error
themes, whereas Cluster 2 (comprising 34.0% of
manuscripts) was high on General errors, Names of
authors, Reference heading, and Source of
journal/periodical, but relatively low on Source of
authored book and Source of edited book.
Figure Caption
Figure 1. Stage 3 Findings: Thematic structure pertaining to reference list error themes and meta-themes.
Author and Book
Latent Effect Size = 12.57%
Manifest Effect Size = 91.6%
Website, Year, Newspaper, and
Title
Latent Effect Size = 11.65%
Manifest Effect Size = 79.5%
Source of
website
Title of work
Names of
authors
Source of edited
book
Source of authored
book
Source of
newspaper
article
Publication
year/date
Dissertation/Thesis, Government,
Paper Presentation, and
Heading
Latent Effect Size = 11.65%
Manifest Effect Size = 74.7%
Government and Miscellaneous Errors
Latent Effect Size = 10.82%
Manifest Effect Size = 98.8%
Source of
dissertation/
thesis
Source of
paper
presentation
Source of
government
document
Reference
heading
Publisher
information
Source of
journal/periodical
General errors
ANTHONY J. ONWUEGBUZIE, EUNJIN HWANG, JULIE P. COMBS, AND JOHN R. SLATE
Fall 2012 xi RESEARCH IN THE SCHOOLS
Figure 2. Stage 4 Findings: Profiles of the manuscripts with respect to the reference list error themes.
Stage 5 Findings
Correlation analysis. A series (i.e., n = 6) of
correlations was conducted to assess the
relationship between the number of reference list
errors and an array of reference list error variables,
namely, the total number of reference list errors,
and the following five components identified by
Onwuegbuzie, Frels, and Slate (2010): Not in
Reference List, Not Consistent with Reference List,
Not in Text, Incomplete or Incorrect Citation, and
Incomplete or Incorrect Reference. According to
Onwuegbuzie and Daniel (2002), variables for
which either the standardized skewness coefficient
(i.e., skewness coefficient divided by its standard
error) or the standardized kurtosis coefficient (i.e.,
kurtosis coefficient divided by its standard error),
or both, are outside the 3 range suggest extreme
departure from normality. An examination of the
skewness and kurtosis coefficients pertaining to the
six reference list error variables revealed a serious
departure from normality for all six reference list
error variables, with all these variables indicating
distributions that were positively skewed and
leptokurtic (i.e., overly peaked shape). Thus, a
nonparametric correlation, namely, Spearman’s
rank, was used to examine these six sets of
relationships. After applying the Bonferroni
adjustment to control for the inflation of Type I
error (i.e., adjusted α = .05/6 = .00833), the
Spearman’s rank correlations revealed that the
number of reference list errors was statistically
significantly and positively related to five of the six
reference list error variables, namely: total number
of reference list errors (rs[83] = .59, p < .001), Not
in Reference List (rs[83] = .39, p < .001), Not
Consistent with Reference List (rs[83] = .48, p <
.001), Not in Text (rs[83] = .42, p < .001), and
Incomplete or Incorrect Reference (rs[83] = .38, p <
.001). All these relationships were large (Cohen,
1988). The only statistically non-significant
Ge
ne
ra
l
0-1
Me
an
Na
me
_A
uth
or
0-1
Me
an
Re
f_h
ea
din
g
0-1
Me
an
So
urce
_A
uth
ore
dB
oo
k
0-1
Me
an
So
urce
_E
dite
dB
oo
k
0-1
Me
an
So
urce
_Jo
urn
al
0-1
Me
an
1 .0
0.8
0.6
0.4
0.2
0.0
Cluster1
Cluster2
EDITORIAL: EVIDENCE-BASED GUIDELINES FOR AVOIDING REFERENCE LIST ERRORS IN
MANUSCRIPTS SUBMITTED TO JOURNALS FOR REVIEW FOR PUBLICATION: A REPLICATION
CASE STUDY OF EDUCATIONAL RESEARCHER
Fall 2012 xii RESEARCH IN THE SCHOOLS
relationship was that between the number of
reference list errors and reference list errors
pertaining to Incomplete or Incorrect Citation
(rs[83] = .13, p = .266).
Further, a series (i.e., n = 4) of nonparametric
(i.e., Spearman) correlations, after applying the
Bonferroni adjustment to control for the inflation of
Type I error, revealed that the number of reference
list errors was statistically significantly related to
the number of references in the reference list
(rs[83] = .27, p = .016). Using Cohen’s (1988)
criteria, this relationship was moderate. However,
no statistically significant relationship was
observed between the number of reference list
errors and the number of manuscript pages (rs[83]
= .17, p = .12), the number of manuscript words
(rs[83] = .21, p = .06), and the number of authors
(rs[83] = .17, p = .19).
Multiple regression analysis. An all
possible subsets (APS) multiple linear regression
(Onwuegbuzie & Daniel, 2003; Thompson 1995)
was used to identify an optimal combination of the
four independent variables (i.e., number of
references in the reference list, number of
manuscript pages, number of manuscript words,
number of authors) that predicted the number of
reference list errors. This analysis, which has been
advocated by many statisticians (e.g.,
Onwuegbuzie & Daniel, 2003; Thompson 1995),
involved examining all possible models containing
some or all of the four independent variables.
When using this analytical technique, separate
regression analyses are conducted for the one
possible set of four independent variables, all four
possible trios of independent variables, all six
possible pairs of independent variables, and all four
independent variables singly—yielding the fitting
of 15 multiple regression models. These 15 models
then were compared to identify the best subset of
independent variables using the following two
criteria: (a) the maximum proportion of variance
explained (R2) and (b) Mallow’s Cp (Myers, 1986;
Sen & Srivastava, 1990). The APS multiple
regression analysis revealed that a model
containing one variable provided the best fit to
these data. In fact, the four-variable model (i.e.,
adding the remaining three independent variables)
only increased the proportion of variance explained
by 3.5%. In addition, Mallow’s Cp was closer in
value to the number of regressor variables (Myers,
1986; Sen & Srivastava, 1990) with the one-
variable solution than with any other variable
solution.
The selected model indicated that the
following variable contributed significantly (F[1,
82] = 4.53, p < .05) to the prediction of the number
of reference list errors: the number of authors. This
variable explained 6.8% of the variation in the
number of reference list errors (Adjusted R2 =
5.3%). Using Cohen’s (1988) criteria for assessing
the predictive power of a set of independent
variables in a multiple regression model, the
proportion of variance explained indicates a small
effect size, because it lay between 2% and 12.99%.
With respect to the assumptions for the
selected one-variable linear regression model, the
Durbin-Watson coefficient of 1.90 was sufficiently
close to 2.00 to suggest that for any two
observations, the residual terms were uncorrelated
(i.e., lack of autocorrelation), which was a desirable
outcome. In addition, an examination of the
standardized residuals pertaining to each of the
participants revealed that no manuscript had a
standardized residual that exceeded 2.00. Thus, in
summary, the selected final regression model
suggested that the manuscripts with the most
citation errors tended to have the highest number of
authors. The regression equation was as follows:
Number of Reference List Errors = 10.69 + 2.40 *
Number of Authors
This equation indicated that among manuscripts
submitted to Educational Researcher, every
additional author of a manuscript was associated
with an increase of 2.40 reference list errors, on
average.
Discussion of Findings
Our current editorial provides further
compelling evidence that APA-related reference
list errors are very common among manuscripts
initially submitted to journals for consideration for
publication. More specifically, the present findings
indicate that not only do reference list errors
permeate manuscripts that are submitted to RITS
but also they similarly pervade manuscripts that are
submitted to the foremost journal in education,
namely, Educational Researcher. And, our
experience as editors of Educational Researcher
led us to conclude that a high proportion of authors
who submit manuscripts to Educational Researcher
are among the most prolific of authors in the world.
Thus, our present editorial has led us to conclude
that reference list errors are not only committed by
beginning authors, but also they are being
committed by prolific authors—thereby supporting
the hypothesis that journals with the highest impact
factors also have manuscripts submitted to their
journals that have high rates of reference list errors.
In fact, the characteristics of reference list errors
for both the Educational Researcher manuscripts
and RITS manuscripts were very similar, including
the distributions of the 14 reference list error
themes. Other similarities in the findings pertaining
ANTHONY J. ONWUEGBUZIE, EUNJIN HWANG, JULIE P. COMBS, AND JOHN R. SLATE
Fall 2012 xiii RESEARCH IN THE SCHOOLS
to the RITS authors and Educational Researcher
authors include the mean number of reference list
errors per manuscript (M = 12.83 vs. M = 14.25,
respectively) and the frequency rate of the
reference list errors. With respect to the latter, from
Table 1, the relationship between the frequencies of
the 50 themes relating to the two studies was
statistically significant and large for both the raw
frequencies (r = .80, p < .001) and the frequency
ranks (rs = .56, p < .001), with three of the top four
reference list errors yielding identical ranks in both
studies.
As was the case for the RITS authors, not a
single author had an error-free reference list, with
the smallest number of unique reference list errors
being three. Also, it should be noted that the mean
number of reference list errors of 14.25 among
Educational Researcher authors likely represents a
lower bound when one takes into account that these
manuscripts were submitted before the writers of
sixth edition of APA required authors to include
digital object identifiers (DOIs) whenever they are
available (cf. section 6.31). As explained by the
writers of the sixth edition of [APA] Publication
Manual, DOI numbers represent unique numbers
assigned by the publisher for electronic referencing
of published journal articles and other documents.
Accordingly, in reference lists, authors should
place the DOI at the end of the reference. Thus,
under the sixth edition of APA, failure to include
available DOI numbers represents a reference list
error—specifically, a reference list error pertaining
to Source of journal/periodical. Moreover, when
we take into account the fact that the 83
Educational Researcher manuscripts represented
those manuscripts that were sent out for external
review, it is reasonable to conclude that this 14.25
mean number of reference list errors is even more
of a lower bound. Further, this mean number of
reference list errors is even more of a lower bound
when we take into account the fact that we only
assessed reference list errors with respect to
violations to APA, and did not assess reference list
errors by comparing each reference contained in
the reference list to the original work, as has many
researchers across numerous fields and disciplines
(e.g., business, economics, social work,
psychology, medicine, library information science;
cf. Adhikari & Bhandari, 2011; de Lacey, Record,
& Wade, 1985; Doms, 1988; Eichorn & Yankauer,
1987; Faunce & Job, 2001; Gatten, 2010; Gosling,
Cameron, & Gibbons, 2004; Hernon & Metoyer-
Duran, 1992; Holt, Siebers, Suder, Loan, & Jeffery,
2000; Kristof, 1997; Ngan Kee, Roach, & Lau,
1997; Nishina, Asano, Mikawa, Maekawa, &
Obara, 1995; O’Connor, 2002; O'Connor &
Kristof, 2001; Roach, Lau, & Ngan Kee, 1997;
Siebers, 2000; Siebers & Holt, 2000; Spivey &
Wilks, 2004; White, 1987)—leading to reference
list errors that have ranged from 4.1% to 66.7%
(Onwuegbuzie, 2012).
The statistically significant and moderate
relationship between the number of reference list
errors and the number of references in the reference
list, although not surprising, suggests that authors
who write manuscripts that contain many
citations—and hence contain longer reference
lists—should be especially careful when compiling
their reference list. An extremely disturbing finding
is that manuscripts submitted to Educational
Researcher that involve more co-authors tend to
exhibit more reference list error themes. This
finding, which replicates the finding of
Onwuegbuzie, Hwang, et al. (2011) and echoes the
positive relationship between the number of
authors and the number of citation errors
documented by Onwuegbuzie, Frels, et al. (2010)
and Onwuegbuzie, Combs, Frels, and Slate
(2011)—again suggests that the bystander effect
likely prevails (Darley & Latané, 1968; Hudson &
Bruckman, 2004; Levine & Thompson, 2004). That
is, when manuscripts are written by multiple
authors, some, if not all, authors assume that one or
more of their other co-authors will/have checked
the in-text citations and the reference list carefully
and/or they believe that one or more of their other
co-authors are more qualified to check the in-text
citations and the reference list and thus their
contribution in this area is not needed
(Onwuegbuzie, Frels, et al., 2010). Thus, our series
of editorials on citation errors and reference list
errors clearly indicate that for manuscripts that
involve multiple authors, all co-authors should be
aware of the potential pitfalls stemming from the
bystander effect and attempt to maximize the
communication channels among all members of the
authorship team.
Conclusions
Our series of articles and editorials in the area
of APA errors of omission and commission make it
clear that the vast majority of authors who submit
manuscripts to journals commit an unnecessarily
large number of APA errors at all components of
an article, including the abstract (Hahs-Vaughn,
Onwuegbuzie, Slate, & Frels, 2009) and body of
the manuscript (Onwuegbuzie & Combs, 2009;
Onwuegbuzie, Combs, Slate, & Frels, 2010). Thus,
in these works, we provided tools and strategies
that we hope will help authors minimize these APA
errors in the future. However, our most recent
articles and editorials on citation errors
(Onwuegbuzie, Combs, et al., 2011; Onwuegbuzie,
Frels, et al., 2010) and reference list errors
(Onwuegbuzie & Hwang, 2012, 2013;
Onwuegbuzie, Hwang, et al., 2011), including the
present editorial, have demonstrated that authors
also commit an unnecessarily large number of APA
EDITORIAL: EVIDENCE-BASED GUIDELINES FOR AVOIDING REFERENCE LIST ERRORS IN
MANUSCRIPTS SUBMITTED TO JOURNALS FOR REVIEW FOR PUBLICATION: A REPLICATION
CASE STUDY OF EDUCATIONAL RESEARCHER
Fall 2012 xiv RESEARCH IN THE SCHOOLS
errors in their reference lists. And, just as authors
who commit in-text APA errors are significantly
more likely to have their manuscripts rejected for
publication by editors, so too are authors who
commit reference list errors significantly more
likely to have their manuscripts rejected
(Onwuegbuzie, Hwang, et al., 2011). Yet, avoiding
reference list errors is not just merely a quality
issue, it is, even more importantly, also an ethical
one, wherein all authors should make every attempt
to document all their sources accurately and fully.
Various professionals at the college level (e.g.,
instructors, mentors, advisors, thesis/dissertation
committee members, chairs/supervisors) and
beyond (e.g., journal editors, publishers, and
writers of future editions of the APA Publication
Manuals) can play an important role in promoting
what Onwuegbuzie, Hwang, et al. (2011) referred
to as a “culture of reference lists that are minimally
error free” (p. xiv). Onwuegbuzie, Combs, et al.
(2010), Onwuegbuzie, Hwang, et al. (2011), and
Onwuegbuzie, Frels, et al. (2010) have discussed
several ways in which these personnel can assist, as
well as have provided tools and strategies (e.g.,
using Table 1 and Table 2 presented in the current
editorial as starting points by focusing on these
most common types of reference list errors and
reference list error themes, respectively) to help
students, researchers, and experienced scholars
prevent making reference list errors. Whatever
strategies are used, we hope that the efficacy of
these strategies is monitored, documented, and
disseminated by those persons using them—as was
undertaken in the aforementioned series of articles
and editorials—so that we can all learn from their
findings.
References
Adhikari, P., & Bhandari, S. (2011). Accuracy of
references in Internet Journal of Medical
Update. Internet Journal of Medical
Update, 6(2), 47-49. Retrieved from
http://www.akspublication.com/ijmu
American Psychological Association. (2010).
Publication manual of the American
Psychological Association (6th ed.).
Washington, DC: Author.
Berelson, B. (1952). Content analysis in
communicative research. New York, NY:
Free Press.
Cattell, R. B. (1966). The scree test for the number
of factors. Multivariate Behavioral Research,
1, 245-276. doi:10.1207/s15327906mbr01
02_10
Chicago Manual of Style. (15th ed.). (2003).
Chicago, IL: The University of Chicago
Press.
Cohen, J. (1988). Statistical power analysis for the
behavioral sciences (2nd ed.). Hillsdale, NJ:
Lawrence Erlbaum.
Constas, M. A. (1992). Qualitative data analysis as
a public event: The documentation of
category development procedures. American
Educational Research Journal, 29, 253-266.
doi:10.3102/00028312029002253
Darley, J. M., & Latané, B. (1968). Bystander
intervention in emergencies: Diffusion of
responsibility. Journal of Personality and
Social Psychology, 8, 377-383. Retrieved
from
http://www.wadsworth.com/psychology_d/te
mplates/student_resources/0155060678_rathu
s/ps/ps19.html
de Lacey, G., Record, C., & Wade, J. (1985). How
accurate are quotations and references
in medical journals? British Medical
Journal (Clinical Research ed.), 291(6499),
884-886. doi:10.1136/bmj.291.6499.884
Doms, C. A. (1988). A survey of reference
accuracy in five national dental journals.
Journal of Dental Research, 68, 442-444.
Eichorn, P., & Yankauer, A. (1987). Do authors
check their references? A survey of accuracy
of references in three public health journals.
American Journal of Public Health, 77,
1011-1012. doi:10.2105/AJPH.77.8.1011
Faunce, G. J., & Job, R. F. S. (2001). The accuracy
of reference lists in five experimental
psychology journals. American Psychologist,
56, 829-830.
doi:10.1037//0003066X.56.10.829
Field, A. (2009). Discovering statistics using SPSS
(3rd ed.). London, England: Sage.
Gatten, R. (2010). A case study in reference list
accuracy. New Library World, 111(1), 16-25.
doi:10.1108/03074801011015658
Glaser, B. G. (1965). The constant comparative
method of qualitative analysis. Social
Problems, 12, 436-445.
doi:10.1525/sp.1965.12.4.03a00070
Glaser, B. G., & Strauss, A. L. (1967). The
discovery of grounded theory: Strategies for
qualitative research. Chicago, IL: Aldine.
Gosling, C. M., Cameron, M., & Gibbons, P. F.
(2004). Referencing and quotation accuracy
in four manual therapy journals. Manual
Therapy, 9(1), 36-40.
doi:10.1016/S1356689X(03)00056-0
Hahs-Vaughn, D. L., Onwuegbuzie, A. J., Slate, J.
R., & Frels, R. K. (2009). Editorial: Bridging
research-to-practice: Enhancing knowledge
through abstracts. Research in the Schools,
16(2), xxxvii-xlv. Retrieved from
http://www.msera.org/download/RITS_16_2
_Abstracts.pdf
ANTHONY J. ONWUEGBUZIE, EUNJIN HWANG, JULIE P. COMBS, AND JOHN R. SLATE
Fall 2012 xv RESEARCH IN THE SCHOOLS
Henson, R. K., Capraro, R. M., & Capraro, M. M.
(2004). Reporting practice and use of
exploratory factor analysis in educational
research journals: Errors and explanation.
Research in the Schools, 11(2), 61-72.
Henson, R. K., & Roberts, J. K. (2006). Use of
exploratory factor analysis in published
research. Educational and Psychological
Measurement, 66, 393-416.
doi:10.1177/0013164405282485
Hernon, P., & Metoyer-Duran, C. (1992).
Literature reviews and inaccurate
referencing: An exploratory study of
academic librarians. College & Research
Libraries, 53, 499-512.
Holt, S., Siebers, R., Suder, A., Loan, R., &
Jeffery, O. (2000). The accuracy of
references in Australian and New Zealand
Medical Journals. New Zealand Medical
Journal, 113, 416-417.
Hudson, J. M., & Bruckman, A. S. (2004). The
bystander effect: A lens for understanding
patterns of participation. Journal of the
Learning Sciences, 13, 165-195.
doi:10.1207/s15327809jls1302_2
Johnson, R. B. (2012). Dialectical pluralism and
mixed research. American Behavioral
Scientist, 56, 751-754.
doi:10.1177/0002764212442494
Kaiser, H. F. (1958). The varimax criterion for
analytic rotation in factor analysis.
Psychometrika, 23, 187-200. doi:10.1007/
BF02289233
Kristof, C. (1997). Accuracy of references in five
entomology journals. American
Entomology, 43, 246-251.
Lambert, Z. V., & Durand, R. M. (1975). Some
precautions in using canonical analysis.
Journal of Market Research, 12, 468-475.
doi:10.2307/3151100
Leech, N. L., & Onwuegbuzie, A. J. (2007). An
array of qualitative analysis tools: A call for
data analysis triangulation. School
Psychology Quarterly, 22, 557-584.
doi:10.1037/10453830.22.4.557
Leech, N. L., & Onwuegbuzie, A. J. (2008).
Qualitative data analysis: A compendium of
techniques for school psychology research
and beyond. School Psychology Quarterly,
23, 587-604. doi:10.1037/1045-
3830.23.4.587
Leech, N. L., & Onwuegbuzie, A. J. (2011).
Beyond constant comparison qualitative data
analysis: Using NVivo. School Psychology
Quarterly, 26, 70-84. doi:10.1037/a0022711
Levine, M., & Thompson, K. (2004). Identity,
place, and bystander intervention: Social
categories and helping after natural disasters.
Journal of Social Psychology, 144, 229-245.
doi:10.3200/SOCP.144.3.229-245
Myers, R. H. (1986). Classical and modern
regression with applications. Boston, MA:
Duxbury Press.
Ngan Kee, W. D., Roach, V. J., & Lau, T. K.
(1997). How accurate are references in the
Australian and New Zealand Journal of
Surgery? Australian and New Zealand
Journal of Surgery, 67, 417-419.
doi:10.1111/j.1445-2197.1997.tb02005.x
Nishina, K., Asano, M., Mikawa, K., Maekawa, N.,
& Obara, H. (1995). The accuracy of
reference lists in Acta Anaesthesiologica
Scandinavica. Acta
Anaesthesiologica Scandinavica, 39, 577-
578. doi:10.1111/j.1399-
6576.1995.tb04129.x
O’Connor, A. E. (2002). A review of the accuracy
of references in the journal. Emergency
Medicine. Emergency Medicine, 14, 139-141.
doi:10.1046/j.1442-2026.2002.00307.x
O'Connor, L. G., & Kristof, C. (2001). Verify your
citations: Accuracy of reference citations in
twelve business and economics journals.
Journal of Business & Finance
Librarianship, 6(4), 23-40.
doi:10.1300/J109v06n04_03
Onwuegbuzie, A. J. (2003). Effect sizes in
qualitative research: A prolegomenon.
Quality & Quantity: International Journal of
Methodology, 37, 393-409.
doi:10.1023/A:1027379223537
Onwuegbuzie, A. J. (2012). Meta-analysis of
studies of the accuracy of reference lists in
published articles. Unpublished manuscript,
Sam Houston State University, Huntsville,
TX.
Onwuegbuzie, A. J., & Combs, J. P. (2009).
Writing with discipline: A call for avoiding
APA style guide errors in manuscript
preparation. School Leadership Review, 4,
116-149.
Onwuegbuzie, A. J., & Combs, J. P. (2010).
Emergent data analysis techniques in mixed
methods research: A synthesis. In A.
Tashakkori & C. Teddlie (Eds.), Handbook of
mixed methods in social and behavioral
research (2nd ed., pp. 397-430). Thousand
Oaks, CA: Sage.
Onwuegbuzie, A. J., Combs, J. P., Frels, R. K., &
Slate, J. R. (2011). Editorial: Citation errors
revisited: The case for Educational
Researcher. Research in the Schools, 18(1),
i-xxxv.
Onwuegbuzie, A. J., Combs, J. P., Slate, J. R., &
Frels, R. K. (2010). Editorial: Evidence-
based guidelines for avoiding the most
common APA errors in journal article
submissions. Research in the Schools, 16(2),
ix-xxxvi. Retrieved from
EDITORIAL: EVIDENCE-BASED GUIDELINES FOR AVOIDING REFERENCE LIST ERRORS IN
MANUSCRIPTS SUBMITTED TO JOURNALS FOR REVIEW FOR PUBLICATION: A REPLICATION
CASE STUDY OF EDUCATIONAL RESEARCHER
Fall 2012 xvi RESEARCH IN THE SCHOOLS
http://msera.org/download/RITS_16_2_APA
Errors6th.pdf
Onwuegbuzie, A. J., & Daniel, L. G.
(2002). Uses and misuses of the
correlation coefficient. Research in
the Schools, 9(1), 73-90.
Onwuegbuzie, A. J., & Daniel, L. G. (2003,
February 12). Typology of analytical and
interpretational errors in quantitative and
qualitative educational research. Current
Issues in Education, 6(2). Retrieved from
http://cie.ed.asu.edu/volume6/number2/
Onwuegbuzie, A. J., & Dickinson, W. B. (2008).
Mixed methods analysis and information
visualization: Graphical display for effective
communication of research results. The
Qualitative Report, 13, 204-225. Retrieved
from http://www.nova.edu/ssss/QR/QR13-
2/onwuegbuzie.pdf
Onwuegbuzie, A. J., Frels, R. K., & Slate, J. R.
(2010). Editorial: Evidence-based guidelines
for avoiding the most prevalent and serious
APA error in journal article submissions—
The citation error. Research in the Schools,
17(2), i-xxiv. Retrieved from
http://www.msera.org/download/RITS_17_2
_Citations.pdf
Onwuegbuzie, A. J., & Hwang, E. (2012).
Reference list errors in manuscripts
submitted to a journal for review for
publication. Unpublished manuscript, Sam
Houston State University, Huntsville, TX.
Onwuegbuzie, A. J., & Hwang, E. (2013).
Reference list errors in manuscripts
submitted to a journal for review for
publication. International Journal of
Education, 5(1), 1-14.
doi:10.5296/ije.v5i2.2191. Retrieved from
http://www.macrothink.org/journal/index.php
/ije/article/view/2191
Onwuegbuzie, A. J., Hwang, E., Frels, R. K., &
Slate, J. R. (2011). Editorial: Evidence-based
guidelines for avoiding reference list errors in
manuscripts submitted to journals for review
for publication. Research in the Schools,
18(2), i-xli. Retrieved from
http://msera.org/rits.htm
Onwuegbuzie, A. J., & Teddlie, C. (2003). A
framework for analyzing data in mixed
methods research. In A. Tashakkori & C.
Teddlie (Eds.), Handbook of mixed methods
in social and behavioral research (pp. 351-
383). Thousand Oaks, CA: Sage.
Onwuegbuzie, A. J., Witcher, A. E., Collins, K. M.,
Filer, J. D., Wiedmaier, C. D., & Moore, C.
W. (2007). Students’ perceptions of
characteristics of effective college teachers:
A validity study of a teaching evaluation
form using a mixed-methods analysis.
American Educational Research Journal, 44,
113-160. doi 10.3102/0002831206298169
Roach, V. J., Lau, T. K., & Ngan Kee, W. D.
(1997). The quality of citations in major
international obstetrics and gynecology
journals. American Journal of Obstetrics and
Gynecology, 177, 973-975.
doi:10.1016/S0002-9378(97)70303-X
Sandelowski, M., Voils, C. I., & Knafl, G. (2009).
On quantitizing. Journal of Mixed Methods
Research, 3, 208-222.
doi:10.1177/1558689809334210
Sen, A. K., & Srivastava, M. (1990). Regression
analysis: Theory, methods, and applications.
New York, NY: Springer-Verlag.
Siebers, R. (2000). The accuracy of references of
three allergy journals. Journal of Allergy and
Clinical Immunology, 105, 837-838.
doi:10.1067/mai.2000.104935
Siebers, R., & Holt, S. (2000). Accuracy of
references in five leading medical journals.
Lancet, 356 (9239), 1445.
doi:10.1016/S0140-6736(05)74090-3
Spivey, C. A., & Wilks, S. E. (2004). Reference list
accuracy in social work journals. Research
on Social Work Practice, 14, 281-286.
doi:10.1177/1049731503262131
Tashakkori, A., & Teddlie, C. (1998). Mixed
methodology: Combining qualitative and
quantitative approaches. Applied Social
Research Methods Series (Vol. 46).
Thousand Oaks, CA: Sage.
Thompson, B. (1995). Stepwise regression and
stepwise discriminant analysis need not apply
here: A guidelines editorial. Educational and
Psychological Measurement, 55, 525-534.
doi:10.1177/0013164495055004001
Thompson, B. (2004). Exploratory and
confirmatory factor analysis: Understanding
concepts and applications. Washington, DC:
American Psychological Association.
White, A. (1987). Reference list inaccuracies: A
four-decade comparison. Journal of
Counseling & Development, 66, 195-196.
doi:10.1002/j.1556-6676.1987.tb00846.x
Zwick, W. R., & Velicer, W. F. (1982). Factors
influencing four rules for determining the
number of components to retain. Multivariate
Behavioral Research, 17, 253-269.
doi:10.1207/s15327906mbr1702_5
Zwick, W. R., & Velicer, W. F. (1986).
Comparison of five rules for determining the
number of components to retain.
Psychological Bulletin, 99, 432-442.
doi:10.1037//0033-2909.99.3.432
Copyright 2012 by the RESEARCH IN THE SCHOOLS
Mid-South Educational Research Association 2012, Vol.19, No. 2, 1-16
Fall 2012 1 RESEARCH IN THE SCHOOLS
Foreign Language Immersion Programs and School Policy: Conflicting Agendas
Heather K. Olson Beal
Stephen F. Austin State University
Michelle Haj-Broussard
McNeese State University
Nicole Boudreaux
Lafayette Parish School System
School districts and policymakers across the
United States are experimenting with various
reforms in an attempt to improve academic
achievement, close the achievement gap, satisfy
ongoing court-ordered desegregation mandates, and
enhance parent-school-community connections
(Aberger, Brown, Mantil, & Perkins, 2009; Caldas
& Bankston, 2005; Epstein et al., 2008; Rothstein,
2004). These reforms include curricular changes,
such as schools with a unified curricular theme, and
structural changes, such as single-sex classes, small
learning communities (SLCs), and early college
high schools (Conchas & Rodríguez, 2008; Hayes,
Pahlke, & Bigler, 2011; Hoffman & Vargas, 2005).
Early foreign language education models are
another type of promising educational reform.
Although research consistently shows that foreign
language study improves academic achievement
(Cobb, Vega, & Kronauge, 2009; Genesee, 2007;
Gómez, Freeman, & Freeman, 2005; Shneyderman
& Abella, 2009; Taylor & Lafayette, 2010), the
percentage of public elementary schools offering
foreign language instruction decreased from 24%
in 1997 to 15% in 2008 (Center for Applied
Linguistics, 2009a).
Despite the overall decrease in early foreign
language education programs, immersion programs
have grown significantly in the United States since
the first one began in 1971 to 448 programs in 38
states and Washington D.C. (Center for Applied
Linguistics, 2011). Twenty-two foreign
languages are represented, with Spanish (45%) and
French (21%) making up the majority of available
programs. There are three types of immersion
programs: total immersion (i.e., at least 90% of
instruction is delivered in the target language),
partial immersion (i.e., approximately 50% of
instruction is delivered in the target language), or
two-way or duali immersion (i.e., equal emphasis
is placed on English and a second language, with
content taught in both languages). The overarching
goal of immersion education is for students to
develop second language proficiency while
mastering subject content, which they learn through
instruction in the second language. In the United
States, immersion students usually are monolingual
English-speakers who participate in an immersion
program in order to acquire a second language.
Immersion programs vary with respect to the
Correspondence concerning this article should be
addressed to Heather K. Olson Beal, Stephen F.
Austin State University, Secondary Education and
Educational Leadership Department, SFA Station
13018 Nacogdoches, TX 75962-3018
Email: [email protected]
In this position article, we explore what happens when school district policies regarding
desegregation, accountability, and foreign language immersion education collide. Specifically, we
contrast 2 immersion programs that experienced distinct outcomes as a result of the conflicting
agendas underlying these 3 policies. One program, originally created to satisfy court-ordered
desegregation objectives, grew and continues to thrive, whereas the other program was damaged and
eventually destroyed due to conflicts that emerged from the collision of the diverse objectives of these
educational policies and programs. We highlight significant implications for practice that emerge
from our analysis of conflicting policy agendas.
HEATHER K. OLSON BEAL, MICHELLE HAJ-BROUSSARD, AND NICOLE BOUDREAUX
Fall 2012 2 RESEARCH IN THE SCHOOLS
objectives of the program, the percentage of
instruction in the target language, the
characteristics of participating students, the
teachers’ primary language, the age at which
second language instruction is initiated, and the
language used to teach basic subjects. However,
they all share three objectives: (a) linguistic:
developing second language proficiency (Collier &
Thomas, 2004; Genesee, 2004; Hammerly, 1987),
(b) academic: increasing content area academic
achievement (Cobb et al., 2009; Genesee, 2007;
Gómez et al., 2005; Shneyderman & Abella, 2009),
and (c) socio-cultural: promoting positive attitudes
towards self and others (Caldas & Boudreaux,
1999; Haj-Broussard, 2003; Ingram & O'Neill,
1999).
Immersion programs also have been used to
achieve other goals unrelated to second language
acquisition. This article focuses on two
educational policies that have been used in
conjunction with immersion programs in
Louisiana: policies designed to promote school- or
district-wide desegregation and policies designed
to raise standardized test scores (Olson Beal, 2008;
Olson Beal, Haj-Broussard, & Boudreaux, 2007).
When immersion programs are co-opted for
policy-related purposes, the objectives of
immersion education can become compromised.
What happens, for instance, when racial quotas
imposed for desegregation purposes become more
important than the immersion curriculum? What
happens when standardized test scores take
priority over target language proficiency? The
results of this complex relationship between
immersion education and educational policy
ranges from diminishing the linguistic proficiency
objective to dissolving immersion programs
altogether when they fail to achieve mandated
racial quotas or predetermined test score targets.
Despite a significant body of literature regarding
the impact of immersion programs on student
achievement (Genesee, Lindholm-Leary,
Saunders, & Christian, 2006; Gómez et al., 2005;
Shneyderman & Abella, 2009), one area that
remains largely unexamined in the literature is the
consequences of using immersion education to
address other policy objectives. Thus, this
position article explores the complexities that arise
when the goals of immersion education become
entangled with desegregation and standardized
testing policy objectives by contrasting two
immersion contexts in Louisiana. One context—
South Boulevard Foreign Language Academic
Immersion Magnetii (South Boulevard) in Baton
Rouge—was created explicitly to achieve
desegregation goals. The other context—an
immersion program at School Xiii—ultimately was
destroyed because compliance with desegregation
mandates and accountability measures took
precedence over maintaining the integrity of the
immersion program.
In the next section, we briefly review the
bodies of literature pertinent to this position article
and then describe the history and research upon
which Louisiana immersion programs are created.
We present a brief portrait of each school and then
explore what happened in two school contexts
when immersion education objectives were co-
opted to achieve other educational objectives. We
conclude with some policy implications.
Review of Relevant Literature
This section provides a brief overview of three
bodies of literature relevant to this position article:
the primary objectives of immersion education, the
primary objectives of school desegregation
policies, and the effects of accountability policy on
language education. These bodies of literature are
relevant to this position article because both
programs highlighted here have been shaped
largely by the convergence of immersion
objectives, school desegregation objectives, and
accountability policies.
Objectives of Immersion Education
Relative to the linguistic objective of
immersion education, results are mixed regarding
immersion students’ target language proficiency.
Although some research has focused on immersion
students’ near-native comprehension skills, high
levels of fluency, and confidence in speaking the
second language (Collier & Thomas, 2004; Swain,
2000), other research has focused on immersion
students’ less-than-native-like grammatical and
lexical accuracy (Hammerly, 1987; Lyster, 1998)
or their less-than-native-like productive language
skills (Genesee, 2004). In his seminal critique of
immersion education, Hammerly (1987) argued
that although immersion programs have been
culturally and politically successful, they fail
linguistically, resulting in students whose target
language proficiency is poor and underdeveloped.
Thus, one strand of immersion pedagogy research
has focused on identifying and implementing
strategies that promote greater accuracy while
simultaneously maintaining the emphasis on
communication.
Relative to the academic objective of
immersion education, much academic research
shows that immersion students experience no
negative effects on either their native language
proficiency or their achievement in other academic
subjects (Caldas & Boudreaux, 1999; Gilbert,
2001; Shneyderman & Abella, 2009). In fact,
findings from many studies show that immersion
students outperform their non-immersion peers
(Genesee, 2007; Howard, Christian, & Genesee,
2003; Howard, Sugarman, & Christian, 2003; Kirk-
FOREIGN LANGUAGE IMMERSION PROGRAMS AND SCHOOL POLICY: CONFLICTING AGENDAS
Fall 2012 3 RESEARCH IN THE SCHOOLS
Senesac, 2002; Thomas & Collier, 2002). Caldas
and Boudreaux (1999) found that Louisiana
immersion students outperformed non-immersion
students on both the English Language Arts (ELA)
and the mathematics sections of the Louisiana
Educational Assessment Program (LEAP) iv —a
particularly important finding considering that the
mathematics instruction was conducted entirely in
French. Haj-Broussard (2003) concluded that the
French immersion education context appeared to
bridge the achievement gap between White non-
immersion and Blackv immersion students.
Regarding the sociocultural objective of
immersion education, foreign language education
advocates assert that language study should
broaden students’ horizons by promoting
appreciation, tolerance, and respect for other
cultures and peoples (National Standards in
Foreign Language Education Project, 1999).
Despite the difficulty in measuring attitudes and
beliefs, some researchers have studied the impact
of immersion programs on this important issue.
Several early studies of the relationship between
foreign language education and attitudes revealed
that immersion students made more favorable
assessments of the “Other” than did non-immersion
students (Cziko, Lambert, & Gutter, 1979; Lambert
& Tucker, 1972). Numerous other researchers
since similarly have reported that second language
education created an openness to cultural and
linguistic differences (Kirk-Senesac, 2002;
Lindholm-Leary & Borsato, 2006; Walsh, 2005).
Haj-Broussard (2003) also studied the affective
dimension of immersion education and found that
Black elementary French immersion students had a
higher collective self-esteem, particularly in
regards to how they viewed their schools, than did
Black non-immersion participants.
Objectives of School Desegregation Early school desegregation research (Coleman
et al., 1966; St. John, 1970) involved determining
the academic and social effects of school
desegregation. In terms of the academic effects,
early research indicated that Black students’
achievement increased as the proportion of White
students in a classroom or school increased
(Coleman et al., 1966; Mahard & Crain, 1983)—
findings that became a catalyst for the
implementation of busing policies to achieve racial
balance in public schools (Armor, 1972; Bradley &
Bradley, 1977; Gerard & Miller, 1975).
Additional research since then consistently has
indicated a positive relationship between
attendance at socioeconomically and racially
diverse schools and higher test scores, grades, and
high school and college completion rates (Condron,
2009; Goldsmith, 2011; Hanushek, Kain, & Rivkin,
2009; Harris, 2006; Lleras, 2008; Mickelson, 2003;
Southworth, 2010).
Numerous researchers have sought to
understand the social effects of desegregation
(Allport, 1954; Braddock II & Gonzales, 2010;
Carter, 2010; Feddes, Noack, & Rutland, 2009;
Schofield, 1991; Tatum, 2003). In a seminal
study, Allport (1954) posited that intergroup
contact would lead to reduced intergroup prejudice
if it occurred in a positive context, which he
defined as one characterized by equal status among
the groups, common goals, lack of competition
among the groups, and institutional sanction for the
contact. More recently, Carter (2010) studied the
role of racially integrated schools in the
development of “cultural flexibility” and found that
mixed-race schools promote social cohesion.
Feddes et al. (2009) and Aboud, Mendelson, and
Purdy (2003) similarly found that cross-ethnic
friendships in ethnically heterogeneous schools led
to positive out-group evaluations among majority
status children (but not minority status children).
Some studies indicate that early school racial
segregation can lead to continued segregation as
adults (Braddock II & Eitle, 2004; Braddock II &
Gonzales, 2010). Other researchers have focused
on the organizational characteristics of schools
(e.g., tracking, transportation methods,
extracurricular offerings) that promote (or inhibit)
“a sense of belonging” among students (Holland,
2012; Meier, 2002; Osterman, 2000).
Accountability Policy and Second Language
Education In addition to the academic and social
objectives of immersion education and
desegregation policies, schools also have to
contend with increased accountability measures
brought on by No Child Left Behind (NCLB) (U.S.
Department of Education, 2001). Many educators
and researchers argue that NCLB has not achieved
its primary goal of reducing the achievement gap
but rather has distorted how education is delivered
in the United States by narrowing curriculum to
include only what is tested (Boykin & Noguera,
2011; Popham, 2004; Rothstein, 2008). Another
significant body of researchers has sought to
identify the effects of increased accountability
measures (particularly high stakes testing) on the
educational experiences of English Language
Learners (Abedi, 2004; Menken, 2009; Solórzano,
2008; Wiley & Wright, 2004) and on minority
youth (McNeil, 2000; Sloan, 2007; Wiley &
Wright, 2004).
A small but growing body of research has
explored the effects of NCLB on second language
instruction and has indicated that despite the fact
that NCLB established foreign languages as a core
curricular content area, implementation of the law
has negatively affected second language programs.
According to Rosenbusch and Jensen (2005), for
instance, foreign language principals and teachers
HEATHER K. OLSON BEAL, MICHELLE HAJ-BROUSSARD, AND NICOLE BOUDREAUX
Fall 2012 4 RESEARCH IN THE SCHOOLS
identified numerous negative effects of NCLB.
More than one fifth of 165 respondents reported
that after passage of NCLB, their school or school
district had eliminated one or more grade levels in
their foreign language program; almost one fourth
of the respondents reported that foreign language
teaching positions had been eliminated and more
than one fifth of respondents reported that one or
more languages had been eliminated. When
asked to identify the cause of these cuts to their
foreign language programs, 43% of respondents
identified insufficient funds, 18% identified
administrator support, and 14% identified state
testing. A 2008 national survey similarly revealed
that more than one third of principals, foreign
language teachers, and department chairs from
more than 5,000 public and private schools stated
that NCLB’s focus on mathematics and reading test
scores had drawn attention and resources away
from foreign language programs (Center for
Applied Linguistics, 2009b).
Understanding these bodies of literature is
important when considering the academic and
social objectives of immersion programs that have
been established in areas (such as Louisiana) that
use immersion education in conjunction with
district desegregation policies and objectives.
Whereas other research focuses on only one or
perhaps two of these policies (Collier & Thomas,
2004; Holland, 2012; Lleras, 2008; Menken, 2009),
this article connects all three policies—immersion
education, school desegregation, and increased
accountability measures—and identifies some
recommendations for practice.
Immersion Education in Louisiana
Louisiana is an apt context through which to
explore this issue for two primary reasons. First,
Louisiana has been grappling with ongoing school
desegregation litigation since the first
desegregation case was filed by the National
Association for the Advancement of Colored
People (NAACP) in New Orleans in 1952 (Muller,
1976). Parish school districts across the state
continue to grapple with mandates to desegregate
their student populations (Bankston & Caldas,
2002). Second, Louisiana has the fourth highest
number of immersion programs in the United
States despite being only the 22nd most populous
state (Center for Applied Linguistics, 2011).
Louisiana has a rich cultural heritage, which
includes the original Native American inhabitants
and the descendants of enslaved persons from
Africa as well as French, Spanish, German, and
Acadian settlers. Louisiana culture has been
largely influenced by its French and Spanish
history. As evidence of that French legacy,
Caldas and Boudreaux (1999) asserted that as much
as 50% of the population of some parishes in
Acadiana still spoke French in the 1990s, although
many of these residents were more than 50 years of
age. Louisiana has renewed its commitment to the
revival of the French language by creating
numerous immersion programs throughout the
state.
In 1984, the Louisiana Board of Elementary
and Secondary Education (BESE) mandated that a
foreign language be taught to all academically able
students in fourth through eighth grades.vi The
result of this mandate has been the initiation and
growth of FLES (Foreign Language in Elementary
School) and immersion programs throughout the
state. As of the 2010-2011 school year, there
were 31,876 fourth- through eighth-grade students
enrolled in French, Spanish, and Latin FLES
programs and an additional 15,441 pre-K–3
students in French and Spanish FLES programs.
In Grades pre-K–8, 3,416 students participated in
French immersion programs, 1,061 in Spanish
immersion programs, and 22 in a new Chinese
immersion program, for a total of 4,499 students.
Thirty-one percent of eligible pre-K-8 Louisiana
public schools offer a foreign language. Both
programs described in this position article are
partial immersion programs, which means that
students spend approximately 60% of the
instructional day immersed in the second language
(in this case, either French or Spanish). Table 1
provides some basic information about both school
sites.
Recommendations in this position article are
based upon several data sources—namely, two
extensive case studies: a case study of South
Boulevard, a K-5 Spanish and French partial
immersion program in Baton Rouge, Louisiana
(Olson Beal, 2008); and a case study of School X, a
French partial immersion program in south central
Louisiana (Haj-Broussard, 2003). Thus, findings
from this article do not constitute original empirical
research; rather, they are the result of our
experience conducting research in two study sites,
as well as our combined lived experiences as
immersion teachers, researchers, administrators,
and parents (Patton, 1990). All three authors have
connections to the school settings that provided us
with unique vantage points from which to explore
these issues. One author was an immersion
teacher and is the current president of the Louisiana
Consortium of Immersion Schools, one was a
parent of two immersion students, and one is a
district immersion program coordinator in
Louisiana. All three authors have conducted
research on immersion programs in Louisiana. A
brief summary of the research design of the two
original studies follows.
Data and recommendations regarding South
Boulevard Elementary are based on an
ethnographic case study involving document
analysis, interviews, and participant observation.
FOREIGN LANGUAGE IMMERSION PROGRAMS AND SCHOOL POLICY: CONFLICTING AGENDAS
Fall 2012 5 RESEARCH IN THE SCHOOLS
In-depth interviews were conducted with 53
students, parents, school faculty, district
administrators, and school board members. Using
purposeful sampling, participants were selected
who represented diverse backgrounds and
perspectives. All interviews were recorded and
transcribed. On-site participant observation
(including classes, recess, lunch, PTO activities and
meetings, and school board meetings) was
conducted for 1 school year.
Table 1.
Immersion Contexts in Study
Data and recommendations regarding School X
stem from a mixed methodology study that examined
academic achievement and experiences of Louisiana
fourth-grade students and teachers in both the regular
education and the French immersion contexts. The
quantitative phase compared these students' LEAP
test scores using an analysis of covariance,
controlling for the Iowa Tests of Basic Skills (ITBS)
scores.vii The qualitative phase was a cross-case
comparison of four classrooms—an extreme class
(90% of the school population in poverty) and a
typical class (Black students of average academic
achievement) in each context using observation,
interviews, and questionnaires. School X’s
participating class—a fourth-grade class of students
in their fifth year of the French immersion
program—was the extreme French immersion case
study from the larger study. All 14 students in the
class were Black and 93% received free/reduced
lunch. Information on the demise of School X,
which occurred after the above study, was drawn
from authors’ lived experiences as officers in the
Louisiana Consortium of Immersion Schools and
from board minutes from the school district
meetings.
In both studies, all interviewees signed Informed
Consent forms and were promised confidentiality.
Thus, pseudonyms are used throughout this article to
protect the identities of study participants. In terms
of fieldwork, detailed field notes were kept along
with a contact summary sheet for each site visit
(Miles & Huberman, 1984). All the data (i.e.,
interview transcripts, field notes) were broken down
into units of meaning (Lincoln & Guba, 1985) and
then the data were organized into discrete categories
that represented events, ideas, or themes that
emerged from the data via an open coding process
(Strauss & Corbin, 1998). The categories then were
compared with one another via the method of
constant comparison (Glaser & Strauss, 1967). A
number of strategies were implemented in both
studies to enhance the credibility, transferability,
dependability, and confirmability of the findings,
including prolonged engagement at the study site
(i.e., 1 school year at South Boulevard and 4 months
at School X), data triangulation through the use of
multiple sources (i.e., follow-up interviews at both
sites, questionnaires at School X, observation,
archival research), peer debriefing, member checks,
and thick description (Geertz, 1973; Lincoln & Guba,
1985; Patton, 2002).
School Location Type of
Program
Admissions
Criteria
Target Student
Demographic
Quotas
Actual Student
Demographics
at time of study
Original
Research
Time
frame
Current
Status
South
Boulevard
Foreign
Language
Academic
Immersion
Magnet
Baton
Rouge
K-5 partial
immersion
in either
French or
Spanish
*Academic
screening
*must be
on-grade
level
1996-2007:
race-based
quota (60%
Black, 40%
non-Black)
2008-present:
family income
quota (55%
full-pay/45%
free or reduced
lunch)
2007: 58%
Black, 42%
non-Black
59%
free/reduced
lunch
08/2006–
08/2007
Program
still
thriving
School X
Acadiana
area
K-5
partial
immersion
in French
*No criteria
*Parents
must
request the
program
More than
90% free and
reduced lunch
or 90% Black
100% Black
93%
free/reduced
lunch
08/2001-
08/2002
Program
closed
in 2003
HEATHER K. OLSON BEAL, MICHELLE HAJ-BROUSSARD, AND NICOLE BOUDREAUX
Fall 2012 6 RESEARCH IN THE SCHOOLS
We conclude this section with a final word
regarding transferability or applicability (or the more
traditional term, generalizability). Although the
two cases examined in this article included a
relatively small number of participants, thereby
potentially limiting the generalizability of the
findings, the authors were able to probe deeply into
the stakeholders’ lived experiences in those contexts.
This thick description allows for a high level of
transferability and applicability. The potential
conflicts that arise when educational reforms have
multiple and sometimes competing objectives are not
unique to Louisiana. Therefore, the positions taken
in this article might be applicable to other
geographical areas and educational contexts and
likely would be of interest to multiple groups who
have an interest in the success of public education:
parents, teachers, administrators, politicians,
policymakers, and concerned citizens and community
members.
South Boulevard: Created for Desegregation
Purposes
The immersion program at South Boulevard, an
urban school located in a historically Black
neighborhood in downtown Baton Rouge, was
originally created in 1996 by a court-approved
Consent Decree (U. S. District Court Middle District
of Louisiana, 1996) to the desegregation litigation
that was originally filed in 1965 with the express
purpose of creating a racially desegregated student
population. Findings in the following section—in
which we explore the three primary objectives of
immersion education and how each was impacted by
the imposition of one or more additional educational
policies—are based on data derived from interviews
and participation observation from the original study
(Olson Beal, 2008).
The development of students’ linguistic skills is
a primary objective of immersion education
(Johnstone, 2002). Participant observation in
immersion classes confirmed that students made
numerous grammatical and syntax errors in speech—
particularly when they branched out to conversation
topics outside the school setting—yet, these mistakes
did not impede communication. Although no
official tests had been administered, South Boulevard
immersion teachers developed their own end-of-
school oral proficiency interview that teachers
conducted to assess students’ target language oral
proficiency. Students were interviewed in their
second language by a teacher (other than their regular
classroom teacher) who assessed their speech on a
scale from 1 (needs work) to 4 (very good) according
to fluency, comprehension, and vocabulary.
Because of confidentiality rules and norms, we did
not have access to these test scores. However,
during an interview, the head immersion teacher
recalled staff members’ reactions after administering
the test for the first time: “We realized that the
children could understand everything we were
saying, but they could only give us a little back. Or
they couldn’t explain themselves.” Staff members
used results from the yearly oral proficiency testing
to identify weaknesses and create campus-wide goals
to strengthen the students’ target language
proficiency. Because reading and writing in the
target language are not a focus of the curriculum,
students’ skills in these two areas are not as strong.
In general, the head immersion teacher reported that
South Boulevard students are “very good orally.
They need to work on their grammar skills.
Especially on written work.”
South Boulevard students spoke their target
language with confidence and ease. One Spanish
immersion teacher originally from Venezuela
concurred, noting the following in an interview:
“These kids, they have incredible Spanish. The way
they talk is incredible. Their pronunciation,
especially. They speak really good Spanish.”
They are willing to take risks with the language,
often speaking “Franglais” or “Spanglish”—as do
their native-speaking teachers. For example, during
fieldwork, a fourth-grade teacher asked her class to
explain what a “recurso natural” (natural resource)
was. A boy immediately raised his hand and
offered the following impromptu explanation: “Un
recurso natural es una cosa que una persona no build;
es de nature” (“A natural resource is something that
a person doesn’t build; it’s from nature”). He
neither stumbled nor hesitated. The teacher
enthusiastically accepted his response and continued
with the lesson. In sum, South Boulevard students’
target language skills are something of a mixed bag.
They make mistakes in oral and written
communication; yet, they also understand the target
language, speak it fluently, and are understood by
each other, by their teachers, and by native speakers
outside the South Boulevard community.
Increasing student achievement in academic
content areas is another objective of immersion
education. Although there are numerous ways to
judge a school’s success, because standardized test
scores are currently the most widely accepted
measure used to evaluate students and schools, they
warrant attention here. Findings from the larger
case study (Olson Beal, 2008) revealed a marked
emphasis on standardized testing throughout the
school year despite the principal’s concern expressed
during interviews that there was too much emphasis
on standardized testing. The LEAP is administered
in the fourth grade—the fifth year of students’
immersion education. Participant observation at the
school revealed a noticeable increase in the quantity
of English used in fourth-grade classrooms by both
teachers and students compared to the lower grades.
During interviews, teachers noted that this shift in
emphasis was due to pressure from both the principal
and the parents to prepare students for this test.
FOREIGN LANGUAGE IMMERSION PROGRAMS AND SCHOOL POLICY: CONFLICTING AGENDAS
Fall 2012 7 RESEARCH IN THE SCHOOLS
Parents of fourth graders are required to attend
an informational meeting about the LEAP test every
fall. During the meeting, the principal explains the
importance of the test—both to individual students
(who must receive a score of Basic or above on the
mathematics and ELA portions of the test in order to
be promoted to fifth grade) and to the school
(because test scores comprise 90% of the school
performance score and monetary awards are given
for performance). The fourth-grade teachers
described the kinds of questions students can expect
and explained how student responses are graded.
Parents also learned that students would receive a
LEAP packet every Monday that needed to be
completed and brought back to school every Friday.
All these LEAP practice questions were written in
English. Thus, although immersion students in
other grades completed homework exercises in the
second language, the fourth graders spent significant
amounts of time completing homework in English.
The teachers spent part of each day, also in English,
reviewing the practice questions. One Spanish
teacher even explained that they “had to tell the
students some of the words in English, because that’s
the way they will see it on the test.” Thus, in this
case, the linguistic proficiency objective was
compromised—although perhaps only temporarily—
in an attempt to reach the academic objective of
increased test scores.
Considering that students received 60% of their
instruction in a second language and that the tests
were in English, one might expect immersion
students to underperform compared to similar peers
on the standardized tests. Instead, analysis of
School Report Cards from 1998-2010 viii revealed
that South Boulevard fourth graders had consistently
scored better on the LEAP than did other students in
East Baton Rouge Parish (EBRP) and in the state of
Louisiana. Figures 1 and 2 belowix illustrate trends
in ELA and mathematics test scores at South
Boulevard compared to the district and the state.
These test scores make South Boulevard fifth out of
54 elementary schools in reading and fourth in
mathematics. Thus, the program is considered
successful because of its students’ standardized test
scores.
The promotion of positive attitudes towards
other cultures, languages, and peoples is a third
objective of immersion education. Although the
South Boulevard immersion curriculum did not
explicitly include multicultural education, teachers in
the sample believed that participation in the
immersion program made students more accepting
and more aware of cultural differences. The head
immersion teacher explained during interviews that
the immersion curriculum exposed students to
diversity to a greater degree than did a traditional
curriculum:
I do think it makes them more culturally aware.
We have Vietnamese children here. We have
Hispanic children here. We have East Indian
children here. So we have a mixed culture of
students and parent population that they get to be
exposed to and see that we’re all people. I
really think they get a much better sense of that.
Madame Rivet, a native Francophone teacher from
Belgium, noted that being in the immersion program
“opens the mind of the children that there is not just
their country, that not everyone is the same.”
Seňore Gonzalez, a Spanish immersion teacher
originally from Colombia, similarly commented that
the immersion curriculum, by definition, promotes
respect, explaining: “This is the idea. You are in
immersion, and immersion means that you need to
know other cultures. You have to respect.”
Cross-racial friendships were commonplace at
South Boulevard. Several parents in the larger
study (Olson Beal, 2008) commented that they liked
the diversity in which their children were immersed
at South Boulevard and identified diversity as a
primary factor that motivated them to choose the
school. David, for example, explained that he did
not think race impacts social relationships at South
Boulevard and described his son’s three best friends
via the following quotation: “Michael is Asian,
James is White, and Anthony is Black.” Ms.
Brown, a veteran South Boulevard ELA teacher,
explained that “for years there have been friendships
across racial lines.” The majority of on-site
observations confirmed positive, cross-racial social
interactions during recess and in classrooms.
Figure 1. Percentage Proficient (Basic or Above),
fourth-grade English Language Arts (ELA) LEAP
scores.
0
20
40
60
80
100
120
1999
2001
2003
2005
2007
2009
South
Boulevard
District
State
HEATHER K. OLSON BEAL, MICHELLE HAJ-BROUSSARD, AND NICOLE BOUDREAUX
Fall 2012 8 RESEARCH IN THE SCHOOLS
Figure 2. Percentage Proficient (Basic or Above),
fourth-grade Mathematics LEAP scores.
Furthermore, the immersion curriculum creates
a unique subculture that adds to the cultures and
identities the students bring with them to school and
creates a culture of integration that nurtures positive
social relationships. The immersion curriculum
causes students to expand upon or to multiply their
identities to include an identity of themselves as
French or Spanish speakers. They either adopt a
new name in Spanish or French or their teachers
pronounce their English names with Spanish or
French pronunciation. Students refer to their
teachers with Spanish and French titles. They refer
to each other as “Spanish kids” and “French kids”
rather than referring to themselves and others as
White, Black, Mexican, or Asian. This new role
allows students to see themselves as members of a
unique community of learners in which all students
are equal. The immersion curriculum is like a third
or an in-between space that belongs equally to all the
students because it is new and unrelated to students’
race, language, family socioeconomic status, or
gender.
In addition to these three objectives, the
immersion program at South Boulevard has another
unique objective that perhaps trumps the other three
in terms of continued administrative support of the
program: the achievement of specific racial and
socioeconomic status quotas. Between 1996 and
2006, the targeted racial quota for South Boulevard
was 55% Black students and 45% non-Black
students. Between 1996 and 2002, non-Black
student enrollment followed the same downward
trend that EBRP as a whole experienced because
many non-Black parents chose to send their children
to private schools and schools in outlying parishes
(Bankston & Caldas, 2002). Beginning in 2002,
however, non-Black enrollment began to increase.
Figure 3 compares student enrollment at South
Boulevard since the immersion program began in
1996 to EBRP as a whole.x
Figure 3. Percentage of non-Black enrollment at
South Boulevard and in the EBRP district, 1995-
2007.
In 2007, the parish moved from race-based
quotas to a family income quota (55% full-pay lunch,
45% free and/or reduced lunch). If and when the
school receives more applications than it has
available seats (which occurs every year), all student
names go into a lottery that admits students
according to the family income quota and an
academic screening that ensures that they are at least
on-grade level. The inclusion of a minimum
academic standard for admission is unusual for
immersion programs, which generally take all
interested students. In fact, at the time of the study,
South Boulevard’s immersion program was one of a
few in Louisiana to include a minimum academic
standard for admission. However, the inclusion of
the academic screening was not a district or a school-
based strategy; rather, it was implemented in
conjunction with ongoing parish desegregation
litigation. As part of the 4-year term of the Final
Settlement Agreement (U. S. District Court Middle
District of Louisiana, 2003), the EBRP school board
agreed to operate an academic theme strand of
dedicated magnet schools, which included South
Boulevard. Addition of the academic label meant
that all students would be screened to ensure that
they were at least on grade level and would have to
maintain a minimum 2.5 GPA to remain in the
program. One participant (a former school board
president) in the larger study explained that this
decision was made to appease the plaintiffs of the
original desegregation lawsuit, who wanted to see
students tested to get into magnet programs. Thus,
the immersion admission policy was dictated—at
least in part—by ongoing school desegregation
policy.
0
20
40
60
80
100
1999
2001
2003
2005
2007
2009
South
Boulevard
District
State
0
10
20
30
40
50
60
70
80
90
100
1995 1999 2002 2004 2007
Perc
en
tag
e o
f n
on
-bla
ck e
nro
llm
en
t
Year
South Boulevard
FOREIGN LANGUAGE IMMERSION PROGRAMS AND SCHOOL POLICY: CONFLICTING AGENDAS
Fall 2012 9 RESEARCH IN THE SCHOOLS
The success of the immersion magnet program at
South Boulevard has been measured principally
according to the degree to which it has achieved the
objectives of school desegregation: creating a more
racially desegregated student body and increasing
student achievement. South Boulevard has been
largely successful at both creating an integrated
student population and increasing student
achievement. The degree to which South Boulevard
has been successful at achieving one of the primary
goals of immersion education—second language
proficiency—has not been considered by district
administrators. During interviews, EBRP district
administrators were asked whether they considered
South Boulevard to be a successful school and, if so,
in what ways. All of the administrators stated that
South Boulevard was a successful school for a
variety of reasons, including “a positive trend in
improvement in student achievement by standard
measures,” “meets the needs of its clients,” and “the
parents are happy; the kids are happy.” None of the
administrators in the sample mentioned any aspect of
second language acquisition in their explanation of
why South Boulevard was a successful school. The
primary issues they mentioned were the racial
composition of the school, students’ test scores, and
parental satisfaction. Thus, desegregation
objectives take precedence over immersion
objectives.
School X: Destroyed by Desegregation and
Accountability Policies In contrast to South Boulevard, which was
created because of school desegregation objectives,
desegregation and accountability policies ultimately
destroyed School X’s immersion program. The
immersion program at School X was negatively
affected by conflicts among immersion education
objectives, testing policies brought about by
increased accountability measures, and school
desegregation policies. As mentioned earlier, a
primary objective of immersion education is the
development of students’ linguistic proficiency. In
the case of School X, the second language
development objective was compromised by policies
imposed by the school principal (who had no
background or training in immersion education) in
the hopes of increasing student test scores. For
instance, the principal required all teachers to
compartmentalize, with different teachers selecting
certain content areas to teach and students moving
from teacher to teacher rather than remaining with a
single teacher who would teach all the subject areas.
She believed that this compartmentalization would
ultimately lead to higher student achievement by
reducing the amount of planning time for teachers
and allowing them to focus on teaching a single
subject area.
This compartmentalization meant that
immersion students had one teacher for ELA, another
for science and mathematics, another for social
studies, and a fourth for French Language Arts—a
context that made it difficult for the immersion
teachers to create a cohesive immersion experience
for their students. The students started each day in
English, stayed in the same class for a French lesson,
and then moved down the hall to their French
mathematics class. For their French social studies
lesson, they went downstairs, outside and across
campus. Finally, for their French science lesson,
they once more went across campus and upstairs
until they were back in the same room as their
mathematics lesson. Having three separate teachers
teach them in French created a disarticulated system.
Rather than meeting the goal of the school, which
was to strengthen subject knowledge, the constant
movement resulted in the loss of valuable
instructional time and an inability to create a
cohesive French environment necessary for the
development of second language proficiency. Thus,
an instructional policy meant to strengthen test scores
actually weakened French immersion students’
content knowledge as well as their second language
proficiency.
Another school policy that weakened students’
French language skills was the principal’s insistence
that because the LEAP test was in English, the
students needed to have anything they did not
understand translated into English. This practice
kept the students from having to work in and
negotiate with French—an important process of
second language acquisition. The students simply
pulled out what the researcher referred to as their
LEAP Trump Card and told the teacher that s/he
must translate for them. The following is a
quotation from the field notes of an interaction
between a fourth-grade immersion student and her
French immersion social studies teacher, M. Kaiga:
Rashona says that she doesn't understand and
that the test will be in English. Kaiga says the
test will be "en françaisvendredi"(in French
Friday). Rashona says no "the LEAP test
will be in English" and she doesn't
understand. (FN, A3, January 24, 2002)
Due to the disarticulated schedule and the LEAP
trump card, the students rarely spoke in French in the
fourth grade at School X. Thus, in order to keep
class going as much as possible in French, the
teachers began to employ what the researcher
described as a language safety net—jumping in
whenever students stumbled with the language to
provide the language for them. Again, this practice
kept students from working and negotiating with the
language. The following is another quotation from
the field notes of an interaction between two fourth-
grade immersion students and M. Kaiga:
Rashona asks what something is. Brianca starts
to explain in French, but Madame Maurice takes
over and answers Rashona's question. Brianca
HEATHER K. OLSON BEAL, MICHELLE HAJ-BROUSSARD, AND NICOLE BOUDREAUX
Fall 2012 10 RESEARCH IN THE SCHOOLS
keeps explaining only quietly (to herself in
French) (FN, A4, January 28, 2002).
These issues stemmed from the principal’s and
parents’ inability to understand the immersion
paradox—that more time in the second language
leads to higher academic achievement in the core
content areas (Johnstone, 2002).
These differences in scheduling and instruction
had ramifications in terms of the students’ academic
achievement. Considering the immersion students’
strong performance in 2001 (with a typical
immersion setting in which students had a single
teacher for mathematics, science and social studies
who spoke in the target language the entire time),
there were some surprising findings in terms of
students’ academic achievement in 2002 (see Table
2). The highest achieving boy whose mathematics
ITBS score was 200 (above the national norm) in
2001 failed the mathematics portion of the LEAP test
in 2002. So although his first 4 years of immersion
(pre-LEAP test pressures and with a single
immersion teacher each year rather than numerous
teachers with compartmentalized subjects) were very
successful, the experiences in his fourth-grade year
negatively affected his test scores.
A descriptive examination of the data shows a
decline in academic achievement throughout the
class when comparing third-grade ITBS scores to
fourth-grade LEAP scores. Table 2 includes
students’ scores on both measures, labeling those
scores as At or Below level, and then marking with
an asterisk when those scores were deemed
unacceptable by the state rating system. Whereas
more than 70% of the students were at or above the
national average of 185 on the ITBS in the third
grade, only approximately 40% were at the basic
level after fourth grade. Furthermore, one half of
the students who failed the state accountability test in
the fourth grade were at or above the national
average in the third grade.
Table 2
School X ITBS and LEAP Achievement Scores
Student number
Mathematics
ITBS scores (185
= national norm)
On/above or
below the
national average
Mathematics
LEAP scores
On or below level
(315 is on
level/basic)
Gain
(+),
loss
(-) or
=
1 202 On 304 Below -
2 187 On 295 Below -
3 203 On 326 On =
4 209 On 356 On =
5 219 On 365 On =
6 188 On 323 On =
7 201 On 323 On =
10 170 Below 289 Below =
11 186 On 241 Below* -
12 200 On 267 Below* -
13 179 Below 283 Below =
14 155 Below 256 Below* =
In addition to these structural and pedagogical
challenges, the principal also struggled adequately
to train and to mentor the immersion teaching
corps. Although the three immersion teachers at
School X were native French speakers, they were
new to immersion education, to Louisiana, and to
the United States, and, thus, had professional
development needs that were unmet. For
example, during fieldwork, the French Language
Arts teacher, who also taught daily 30-minute
French classes to the non-immersion students,
informed the researcher that she did not even
realize that the immersion class was taught using
immersion methods; she informed the researcher
she thought they were the smart fourth-grade class.
Thus, the French immersion teachers—although
fluent French speakers and experienced
educators—needed professional development in the
grammar of U.S. schooling (Tyack & Cuban,
1995), as well as training in immersion pedagogy.
In the midst of School X’s teachers’ efforts to
increase student achievement and promote second
language proficiency, School X came under
scrutiny because its student population was racially
identifiable. xi The school board—in order to
comply with a federal desegregation order—
FOREIGN LANGUAGE IMMERSION PROGRAMS AND SCHOOL POLICY: CONFLICTING AGENDAS
Fall 2012 11 RESEARCH IN THE SCHOOLS
decided to reassign teachers to racially balance the
district’s teaching corps across its campuses.
Thus, the strongest immersion teachers from
School X—three African immersion teachers who
had developed bonds with the Black students at
School X—were transferred to a majority White
campus. In the following quotation, one of the
African immersion teachers described the
similarities they had discovered between
themselves and their Black students:
When I first came here they told me, OK,
your school is the majority Black students.
So, this was a big surprise for me. First, I
was expecting to have White children in my
class . . . I discovered that we have some
things in common that are related to culture . .
. and also some people speak French. A kind
of French, which is different from what I
studied at school, but it's French. I can talk
with them and communicate and this is also
amazing … for us the United States there is
no other language—just English. (INT, Kaiga,
January 15, 2002)
Unfortunately, the ethnic and cultural similarities
that M. Kaiga identified were considered a negative
under the desegregation order, and the teachers
were not given a chance to capitalize on those
similarities. Instead, several weaker White
teachers were transferred into School X—a
majority Black campus. One of the White
teachers who was transferred to School X had been
asked to leave a neighboring parish school the
previous year because she was unsuccessful at
teaching at the majority Black campus. Once this
unsuccessful teacher had joined School X, the
attrition rate in the immersion classroom
accelerated.
The immersion program—already weakened
by scheduling constraints and pedagogical
challenges that emerged due to the principal’s lack
of understanding of immersion education—
suffered further because of staffing decisions made
to satisfy a school desegregation mandate. In this
case, the pressures of accountability and the
implementation of the district’s school
desegregation policies resulted in the destruction of
one of the last majority Black French immersion
programs in the state.
Notes Toward a Conclusion
In this article, we contrasted two immersion
programs that have experienced unique outcomes
as a result of educational policies with conflicting
agendas. In the case of South Boulevard, school
desegregation policies largely have dictated the
particulars of the immersion program from its
inception. South Boulevard’s immersion program
is evaluated primarily according to two criteria:
(a) its success at creating a student body that meets
first the racial and now socioeconomic targets
dictated by federal school desegregation policy and
(b) its students’ test scores. South Boulevard’s
students’ Spanish and French language proficiency
is not a criterion by which the school is evaluated
either by the school principal or district
administration, who are largely unaware of the
unique programmatic and pedagogical concerns of
immersion education. Thus, a desire to increase
student test scores often dictates scheduling
concerns and other pedagogical issues, such as
homework and the use (or lack of use) of the
second language in the classroom. The need to
remain in compliance with desegregation mandates
also dictates admission policies. Despite these
potentially troubling issues, South Boulevard’s
immersion program is thriving. Its students are
achieving success on their state’s standardized test,
they have a racially and socioeconomically diverse
student population and teaching staff, and they
continue to make strides in terms of second
language proficiency.
In contrast, the stakeholders' reactions to the
challenges brought on by conflicting policies
ultimately led to the demise of the immersion
program at School X. School and district
administrators—unaware of the unique needs of
immersion education—implemented scheduling
and staffing decisions that compromised the needs
of the immersion program. Efforts to increase
students’ test scores conflicted with students’
second language needs and their second language
development was ultimately derailed. Students
suffered further when teachers were reassigned to
satisfy district desegregation mandates. Thus, the
immersion program was systematically weakened
due to the pressures of standardized tests and
misguided school administration that did not
understand immersion and, eventually, the program
was closed.
What lessons did we learn from this analysis of
the complex interactions among competing
educational policies at these two schools? First,
as immersion educators, we must increase our
awareness and knowledge of educational policies
that affect our programs. We learned that these
immersion programs—like other programs with
specialized curricula—do not exist in isolation;
rather, they have their own unique objectives that
must be understood in conjunction with non-
immersion objectives. In order for these
immersion programs to thrive, their stakeholders
(including administrators, teachers, parents, and
even students) must become adept at negotiating in
order to protect the interests of their program. If
we do not advocate for maintaining the integrity of
the immersion programs with which we work,
immersion objectives might be compromised and
the programs themselves might be eliminated so
that other programs or interests might be served.
HEATHER K. OLSON BEAL, MICHELLE HAJ-BROUSSARD, AND NICOLE BOUDREAUX
Fall 2012 12 RESEARCH IN THE SCHOOLS
Second, the experiences of the immersion
programs at South Boulevard and School X
highlight a need for multiple layers of professional
development. Our school- and district-level
administrators need professional development that
will further their knowledge of immersion
pedagogy and allow them to make decisions that
will enable the immersion programs under their
purview to be successful. School- and district-
level administrators in these two cases also need
professional development to help them understand
the socio-cultural issues that many international
teachers face when they come to Louisiana to teach
in our public schools. Lastly, immersion teachers
at these schools need professional development to
deepen their understanding of immersion
pedagogy, allowing them to become teacher-
leaders on their campuses who are able to advocate
for their students’ academic and linguistic needs
and contribute in positive ways to the dialogue and
negotiation necessary for unique educational
programs with sometimes conflicting objectives to
thrive.
Lastly, in this era of increasing competition
among traditional public, charter, and private
schools, all stakeholders in these programs need to
understand and be explicit about the goals of their
programs. School boards, administrators,
teachers, and parents might have different primary
objectives. Immersion programs—particularly in
the case of Louisiana—often are used to entice
White and higher income parents to choose public
over private education. Thus, one group of
stakeholders might want to create a program that
will be the best carrot to attract parents to public
schools with unique educational programs.
Another might want to create a program that will
help its students achieve the highest test scores
possible. Yet another might want a program that
will enable students to develop high levels of
second language proficiency and intercultural
competence. These goals might conflict with
each other, which might lead to misunderstandings
and disappointment across all parties. Thus,
transparency and diplomacy are key to creating
programs that can meet the diverse needs of all
invested stakeholders.
The lead editors for this article were Anthony J.
Onwuegbuzie and John R. Slate.
References
Abedi, J. (2004). The No Child Left Behind Act
and English language learners:
Assessment and accountability issues.
Educational Researcher, 33(1), 4-14.
doi:10.3102/0013189X033001004
Aberger, S., Brown, B., Mantil, A., & Perkins, A.
(2009). Closing the student achievement
gap: The overlooked strategy of
socioeconomic integration. Paper
presented to Prof. Jal Mahta, Harvard
Graduate School of Education. Retrieved
from
http://a100educationalpolicy.pbworks.com
/f/Closing+the+Achievement+Gap+-
+Socioeconomic+Integration.pdf
doi:10.1080/01650250244000164
Aboud, F. E., Mendelson, M. J., & Purdy, K. T.
(2003). Cross-race peer relations and
friendship quality. International Journal
of Behavioral Development, 27, 165-173.
doi: 10.1080/01650250244000164
Allport, G. W. (1954). The nature of prejudice.
Cambridge, MA: Addison-Wesley.
Armor, D. J. (1972). The evidence on busing.
Public Interest, 28, 90-126. Retrieved
from
http://www.nationalaffairs.com/public_int
erest/issues/summer-1972
Bankston, C. L., & Caldas, S. L. (2002). A troubled
dream: The promise and failure of school
desegregation in Louisiana. Nashville,
TN: Vanderbilt University Press.
Boykin, A. W., & Noguera, P. (2011). Creating the
opportunity to learn: Moving from
research to practice to close the
achievement gap. Alexandria, VA:
Association for Supervision and
Curriculum Development.
Braddock II, J. H., & Eitle, T. (2004). The effects
of school desegregation. In J. Banks & C.
M. Banks (Eds.), Handbook of research
on multicultural education (2nd ed., pp.
828-843). San Francisco, CA: Jossey-
Bass.
Braddock II, J. H., & Gonzales, A. D. C. (2010).
Social isolation and social cohesion: The
effects of K–12 neighborhood and school
segregation on intergroup orientations.
Teachers College Record, 112, 1631-
1653.
Bradley, L. A., & Bradley, G. W. (1977). The
academic achievement of black students in
desegregated schools: A critical review.
Review of Educational Research, 47, 399-
449. doi:10.3102/00346543047003399
Caldas, S. J., & Bankston, C. L. (2005). Forced to
fail: The paradox of school desegregation.
Westport, CT: Praeger.
Caldas, S. J., & Boudreaux, N. (1999). Poverty,
race, and foreign language immersion:
Predictors of academic achievement.
Learning Languages, 5(1), 4-15.
Carter, P. L. (2010). Race and cultural flexibility
among students in different multiracial
FOREIGN LANGUAGE IMMERSION PROGRAMS AND SCHOOL POLICY: CONFLICTING AGENDAS
Fall 2012 13 RESEARCH IN THE SCHOOLS
schools. Teachers College Record, 112,
1529-1574.
Center for Applied Linguistics. (2009a). Directory
of two-way bilingual immersion programs
in U.S. schools. Center for Applied
Linguistics. Retrieved from
http://www.cal.org/twi/directory/index.ht
ml
Center for Applied Linguistics. (2009b). World
language teaching in U.S. schools:
Preliminary results from the National K-
12 Foreign Language Survey. Retrieved
from
http://www.cal.org/projects/flsurvey.html
Center for Applied Linguistics. (2011). Directory
of partial immersion programs in U.S.
schools. Washington, DC: Center for
Applied Linguistics. Retrieved from
http://www.cal.org/resources/immersion/
Cobb, B., Vega, D., & Kronauge, C. (2009).
Effects of an elementary dual language
immersion school on junior high school
achievement. In D. L. Hough (Ed.),
Middle grades research: Exemplary
studies linking theory to practice (pp. 1-
20). Charlotte, NC: Information Age
Publishing.
Coleman, J. S., Campbell, E., Hobson, C.,
McPartland, J., Mood, A., Weinfeld, F., et
al. (1966). Equality of educational
opportunity. Washington, DC: U.S. Office
of Education.
doi:10.1080/0020486680060504
Collier, V. P., & Thomas, W. P. (2004). The
astounding effectiveness of dual language
education for all. NABE Journal of
Research and Practice, 2(1), 1-19.
Conchas, G. Q., & Rodríguez, L. F. (2008). Small
schools and urban youth: Using the power
of school culture to engage students.
Thousand Oaks, CA: Corwin Press.
Condron, D. (2009). Social class, school and non-
school environments, and Black/White
inequalities in children's learning.
American Sociological Review, 74, 683-
708. doi:10.1177/000312240907400501
Cziko, G. A., Lambert, W. E., & Gutter, R. (1979).
French immersion programs and students'
social attitudes: A multidimensional
investigation. Working Papers on
Bilingualism, 19, 13-28.
Epstein, J., Sanders, M. G., Sheldon, S., Simon, B.
S., Salinas, K. C., Jansorn, N. R., et al.
(2008). School, family, and community
partnerships: Your handbook for action
(3rd ed.). Thousand Oaks, CA: Corwin
Press.
Feddes, A. R., Noack, P., & Rutland, A. (2009).
Direct and extended friendship effects on
minority and majority children’s
interethnic attitudes: A longitudinal study.
Child Development, 80(2), 377-390.
doi:10.1111/j.1467-8624.2009.01266.x
Geertz, C. (1973). The interpretation of cultures.
New York, NY: Basic Books.
Genesee, F. (2004). What do we know about
bilingual education for majority language
students? In T. K. Bhatia & W. C. Ritchie
(Eds.), Handbook of Bilingualism and
Multiculturalism (pp. 547-576). Malden,
MA: Blackwell.
Genesee, F. (2007). French immersion and at-risk
students: A review of research evidence.
The Canadian Modern Language Review,
63, 655-688. doi:10.3138/cmlr.63.5.655
Genesee, F., Lindholm-Leary, K., Saunders, W.
M., & Christian, D. (Eds.). (2006).
Educating English language learners: A
synthesis of research evidence. New York,
NY: Cambridge University Press.
doi:10.1017/CBO9780511499913
Gerard, H. B., & Miller, N. (Eds.). (1975). School
desegregation: A long-term study. New
York, NY: Plenum Press.
Gilbert, S. M. (2001). The impact of two-way dual-
language programs on fourth-grade
students: Academic skills in reading and
math, language development, and self-
concept development. (Unpublished
doctoral dissertation). New Mexico State
University, Las Cruces, NM.
Glaser, B., & Strauss, A. (1967). Discovery of
grounded theory. Chicago, IL: Aldine.
Goldsmith, P. R. (2011). Coleman revisited: School
segregation, peers, and frog ponds.
American Educational Research Journal,
48, 508-535.
doi:10.3102/0002831210392019
Gómez, L., Freeman, Y., & Freeman, D. (2005).
Dual language education: A promising 50-
50 dual language model. Bilingual
Research Journal, 29, 145-164.
doi:10.1080/15235882.2005.10162828
Green v. School Board of New Kent County,
Virginia, 391 U.S. 430 C.F.R. (1968).
Haj-Broussard, M. (2003). Language, identity and
the achievement gap: Comparing
experiences of African-American students
in a French immersion and a regular
education context. (Unpublished doctoral
dissertation). Louisiana State University,
Baton Rouge, LA.
Hammerly, H. (1987). The immersion approach:
Litmus test of second language acquisition
through classroom communication.
Modern Language Journal, 71, 395-401.
doi:10.1111/j.1540-4781.1987.tb00378.x
Hanushek, E. A., Kain, J. F., & Rivkin, S. G.
(2009). New evidence about Brown v.
Board of Education: The complex effects
HEATHER K. OLSON BEAL, MICHELLE HAJ-BROUSSARD, AND NICOLE BOUDREAUX
Fall 2012 14 RESEARCH IN THE SCHOOLS
of school racial composition on
achievement. Journal of Labor
Economics, 27, 349-383.
doi:10.1086/600386
Harris, D. N. (2006). Lost learning, forgotten
promises: A national analysis of school
racial segregation, student achievement
and controlled choice plans. Washington,
DC: Center for American Progress.
Retrieved from
http://www.americanprogress.org/issues/2
006/11/pdf/lostlearning.pdf
Hayes, A. R., Pahlke, E. E., & Bigler, R. S. (2011).
The efficacy of single-sex education:
Testing for selection and peer quality
effects. Sex Roles, 65(9-10), 693-703.
doi:10.1007/s11199-010-9903-2
Hoffman, N., & Vargas, J. (2005). Early College
High School Initiative: Integrating grades
9 through 19: The Bill and Melinda Gates
Foundation. Retrieved from
http://eric.ed.gov/PDFS/ED486161.pdf
Holland, M. M. (2012). Only here for the day : The
social integration of minority students at a
majority white high school. Sociology of
Education, 85, 101-120.
doi:10.1177/0038040712440789
Howard, E. R., Christian, D., & Genesee, F.
(2003). The development of bilingualism
and biliteracy from grade 3 to 5: A
summary of findings from the
CAL/CREDE study of two way immersion
education. Washington, DC: Center for
Applied Linguistics. Retrieved from
http://www.cal.org/crede/pdfs/rr13.pdf.
Howard, E. R., Sugarman, J., & Christian, D.
(2003). Trends in two-way immersion
education: A review of the research.
Washington, DC: Center for Applied
Linguistics. Retrieved from
http://www.csos.jhu.edu/crespar/techRepo
rts/Report63.pdf
Ingram, D. E., & O'Neill, S. (1999). Cross-cultural
attitudes as a goal of language teaching in
the global context (No. ED430398).
Washington, DC: ERIC. Retrieved from
http://eric.ed.gov/PDFS/ED430398.pdf
Johnstone, R. (2002). Immersion in a second or
additional language at school: A review of
the international research: Retrieved from
the Scottish CILT website:
http://www.scilt.stir.ac.uk
Kirk-Senesac, B. V. (2002). Two-way bilingual
immersion: A portrait of quality
schooling. Bilingual Research Journal,
26(1), 85-102.
Lambert, W. E., & Tucker, R. (1972). Bilingual
education of children: The St. Lambert
experiment. Rowley, MA: Newbury
House.
Lincoln, Y., & Guba, E. G. (1985). Naturalistic
inquiry. Beverly Hills, CA: Sage.
doi:10.1016/0147-1767(85)90062-8
Lindholm-Leary, K. J., & Borsato, G. (2006).
Academic achievement. In F. Genesee, K.
Lindholm-Leary, W. Saunders & D.
Christian (Eds.), Educating English
language learners: A synthesis of research
evidence (pp. 176-222). New York, NY:
Cambridge University Press.
doi:10.1017/CBO9780511499913.008
Lleras, C. (2008). Race, racial concentration, and
the dynamics of educational inequality
across urban and suburban schools.
American Educational Research Journal,
45, 886-912.
doi:10.3102/0002831208316323
Lyster, R. (1998). Negotiation of form, recasts, and
explicit correction in relation to error
types and learner repair in immersion
classrooms. Language Learning, 48, 183-
218. doi:10.1111/1467-9922.00039
Mahard, R. E., & Crain, R. L. (1983). Research on
minority achievement in desegregated
schools. In C. H. Rossell & W. D. Hawley
(Eds.), The consequences of school
desegregation (pp. 103-125). Philadelphia,
PA: Temple University Press.
McNeil, L. M. (2000). Contradictions of school
reform: Educational costs of standardized
testing. New York, NY: Routledge.
Meier, D. (2002). In schools we trust: Creating
communities of learning in an era of
testing and standardization. Boston, MA:
Beacon Press.
Menken, K. (2009). No Child Left Behind and its
effect on language policy. Annual Review
of Applied Linguistics, 29, 103-117.
doi:10.1017/S0267190509090096
Mickelson, R. A. (2003). The academic
consequences of desegregation and
segregation: Evidence from the Charlotte-
Mecklenburg schools. North Carolina
Law Review Association, 81, 1513-1562.
Miles, M. B., & Huberman, A. M. (1984). Drawing
valid meaning from qualitative data:
Toward a shared craft. Educational
Researcher, 13(5), 20-30.
doi:10.3102/0013189X013005020
Muller, M. L. (1976). New Orleans public school
desegregation. Louisiana History: The
Journal of the Louisiana Historical
Association, 17(1), 69-88. Retrieved
from
http://cls.louisiana.edu/lha/journal/index.s
html
National Standards in Foreign Language Education
Project. (1999). Standards for foreign
language learning in the 21st century.
Yonkers, NY: Author.
FOREIGN LANGUAGE IMMERSION PROGRAMS AND SCHOOL POLICY: CONFLICTING AGENDAS
Fall 2012 15 RESEARCH IN THE SCHOOLS
Olson Beal, H. K. (2008). Speaking the language of
desegregation: A case study of South
Boulevard Foreign Language Academic
Immersion Magnet. (Unpublished doctoral
dissertation). Louisiana State University,
Baton Rouge, LA.
Olson Beal, H. K., Haj-Broussard, M., &
Boudreaux, N. (2007, November). Lived
experiences: Perspectives on immersion
and desegregation in the South. Paper
presented at the annual meeting of the
American Council of Teachers of Foreign
Languages.
Osterman, K. F. (2000). Students' need for
belonging in the school community.
Review of Educational Research, 70, 323-
367. doi:10.3102/00346543070003323
Patton, M. Q. (1990). Qualitative evaluation and
research methods (2nd ed.). Newbury
Park, CA: Sage.
Patton, M. Q. (2002). Qualitative research and
evaluation methods (3rd ed.). London,
England: Sage.
Popham, W. J. (2004). America's "failing" schools:
How parents and teachers can cope with
No Child Left Behind. New York, NY:
Routledge Falmer.
Rosenbusch, M. H., & Jensen, J. (2005). Status of
foreign language programs in the
NECTFL states. NECTFL Review, 56, 26-
37. Retrieved from
http://nectfl.net/review.html
Rossell, C., & Clark, R. C. (1990). The carrot or
the stick for school desegregation policy:
Magnet schools or forced busing.
Philadelphia, PA: Temple University
Press.
Rothstein, R. (2004). Class and schools: Using
social, economic, and educational reform
to close the black-white achievement gap.
New York, NY: Teachers College Press.
Rothstein, R. (2008). Grading education: Getting
accountability right. New York, NY:
Teachers College Press.
Schofield, J. W. (1991). School desegregation and
intergroup relations: A review of the
literature. Review of Research in
Education, 17, 335-409.
doi:10.3102/0091732X017001335
Shneyderman, A., & Abella, R. (2009). The effects
of the extended foreign language
programs on Spanish-language
proficiency and academic achievement in
English. Bilingual Research Journal, 32,
241–259.
doi:10.1080/15235880903372829
Sloan, K. (2007). High-stakes accountability,
minority youth, and ethnography:
Assessing the multiple effects.
Anthropology and Education Quarterly,
38(1), 24-41.
doi:10.1525/aeq.2007.38.1.24
Solórzano, R. W. (2008). High states testing:
Issues, implications, and remedies for
English language learners. Review of
Educational Research, 78, 260-329.
doi:10.3102/0034654308317845
Southworth, S. (2010). Examining the effects of
school composition on North Carolina
student achievement over time. Education
Policy Analysis Archives, 18(29), 1-45.
Retrieved from
http://epaa.asu.edu/ojs/article/view/848/87
2
St. John, N. H. (1970). Desegregation and minority
group performance. Review of
Educational Research, 40(1), 111-133.
doi:10.3102/00346543040001111
Strauss, A., & Corbin, J. (1998). Basics of
qualitative research. Thousand Oaks, CA:
Sage.
Swain, M. (2000). French immersion research in
Canada: Recent contributions to SLA and
applied linguistics. Annual Review of
Applied Linguistics, 20, 199-212.
doi:10.1017/S0267190500200123
Tatum, B. D. (2003). "Why are all the Black kids
sitting together in the cafeteria?" and
other conversations about race. New
York, NY: Basic Books.
Taylor, C., & Lafayette, R. (2010). Academic
achievement through FLES: A case for
promoting greater access to foreign
language study among young learners. The
Modern Language Journal, 94(1), 22-42.
doi:10.1111/j.1540-4781.2009.00981.x
Thomas, W. P., & Collier, V. P. (2002). A national
study of school effectiveness for language
minority students' long-term academic
achievement: Center for Research on
Education, Diversity and Excellence,
University of California-Santa Cruz.
Tyack, D., & Cuban, L. (1995). Tinkering towards
utopia: A century of public school reform.
Cambridge, MA: Harvard University
Press.
U.S. Department of Education. (2001). No Child
Left Behind Act of 2001 (Vol. Pub. L.
107-110, § 115, Stat. 1425 (2002).
U. S. District Court Middle District of Louisiana.
(1996). Consent Decree (No. Civil Action
No. 56-1662-A). Baton Rouge, LA.
U. S. District Court Middle District of Louisiana.
(2003). Final Settlement Agreement (No.
Civil Action No. 56-1662-D-M3). Baton
Rouge, LA.
Walsh, N. (2005). Collaborer pour s’ouvrir aux
autres: L’éveil aux langues dans une
classe d’immersion française
(Unpublished master’s thesis). Simon
HEATHER K. OLSON BEAL, MICHELLE HAJ-BROUSSARD, AND NICOLE BOUDREAUX
Fall 2012 16 RESEARCH IN THE SCHOOLS
Fraser University, British Columbia,
Canada.
Wiley, T. G., & Wright, W. E. (2004). Against the
undertow: Language-minority education
policy and politics in the "age of
accountability." Educational Policy,
18(1), 142-168.
doi:10.1177/0895904803260030
i Dual immersion or two-way immersion programs
are ideally composed of a balance of native
English-speakers and native speakers of the non-
English language, whereas total and partial
immersion programs are typically composed of
native English-speakers. ii South Boulevard, the actual name of the school,
was used in the larger study because a history of
the school from 1956-2007 was a component of the
study. Furthermore, at the time of the study,
South Boulevard was the only public immersion
program in Louisiana to offer both Spanish and
French immersion tracks. Thus, the school would
have been easily identifiable. However,
pseudonyms are used throughout this article to
protect the anonymity of the participants. iii School X is the pseudonym that was used in the
larger study to protect the anonymity of the school
and the study participants. Pseudonyms are used
throughout this article to protect the anonymity of
the participants. ivThe LEAP is a criterion-referenced (CRT) test
that measures how well a student has mastered the
state content standards. The LEAP is
administered in grades 4 and 8. v We use the term “Black” rather than “African-
American” throughout because it was used by
study participants and in demographic records
throughout the era of court-ordered desegregation. vi Details regarding this mandate can be found in
Bulletin 741: Louisiana Handbook for School
Administrators, which can be accessed at the
following URL:
http://doe.louisiana.gov/bese/policies.html vii This was possible because strong and
statistically significant correlations were found
between the covariates (ITBS scores) and the
dependent variables (the LEAP scores) in and
across the mathematics and language subject tests,
with correlation coefficients ranging from .58 to
.74. viii 1998-1999 was the first school year during
which the LEAP for Mathematics and English
Language Arts was administered to all Louisiana
public school students in Grades 4 and 8. ix Data obtained from annual School Report Cards
issued by the Louisiana State Department of
Education and are available online at
www.louisianaschools.net from 1996 to the
present.
x Data obtained from Louisiana School Directories
and Louisiana State Department of Education
Annual Financial and Statistical Reports. xi The phrase “racially identifiable” came from a
statement from the Commission on Civil Rights
that was cited in the Supreme Court Green vs.
County School Board of New Kent County (1968)
decision, which stated that New Kent County’s
freedom of choice plan was inadequately
addressing lingering racial segregation in the
district. Later, Rossell and Clark (1990) defined it
as a school with a racial composition that deviated
20 percentage points or more from the district
racial composition.
Copyright 2012 by the RESEARCH IN THE SCHOOLS
Mid-South Educational Research Association 2012, Vol.19, No. 2, 17-29
Fall 2012 17 RESEARCH IN THE SCHOOLS
Factors Predicting Pre-service Teachers’ Adoption of Web 2.0 Technologies
Jongpil Cheon, Fanni Coward, Jaeki Song, and Sunho Lim
Texas Tech University
A decade ago, Prensky (2001), along with
others, called the new generation of students in K-12
classrooms digital natives because of the students’
familiarity with technologies such as computers,
video games, digital music players, and cell phones.
More than a decade later, young people have added
to their digital birthright the use of online spaces,
such as Google, YouTube, Pinterest, Facebook and
Twitter.
Web 2.0 is a term reflecting a family of online
environments in which users create, share and
combine information (O’Reilley, 2005). These new
environments afford opportunities for instructionally
driven, student-centered learning using social and
collaborative functionalities (Williams & Chinn,
2009). The adoption by school systems and teachers
of various web-enabled portable and hand-held
devices (e.g., laptops, iPods, and iPads) is helping
move K-12 teachers closer to a facilitative role (Li,
Liu, Li, Wang, & Chen, 2011; Russell, 2009). In
this context, however, teachers are the primary
curriculum designers and innovators, and their
readiness to apply new technologies represents a
significant milestone for use of these new
pedagogical approaches in the classroom.
As students think and process information in
ways that are different from the earlier generations
(Prensky, 2001), there is a corresponding pressure for
the role of teachers to shift away from being an
information provider toward becoming an
information facilitator. Recently, researchers have
begun to conceptualize the instructional use of
technology as an important dimension of teachers’
pedagogical content knowledge (Mishra & Koehler,
2006; Niess, 2005; Niess, Zee, & Gilow-Wiles,
2010). The factors that influence teachers’ adoption
of instructional technology is fairly well understood,
but much less is known about what influences the
instructional adoption of Web 2.0 technologies. The
current study employed the Theory of Planned
Behavior (TPB; Ajzen, 1991) as an analytical
framework to provide a more comprehensive
explanation of technology adoption behavior specific
to Web 2.0 technologies.
Technology Affordances and Teacher Beliefs
Web 2.0 Technologies
Web 2.0 refers to a read/write web in which
users actively participate in creating and sharing
information (e.g., Wikipedia, Facebook, Twitter,
blogs, or YouTube). Web 1.0, on the other hand,
refers to a read-only web in which users consume
available information (Murugesan, 2007). Web 2.0
Correspondence concerning this article should be
addressed to Fanni Coward, College of Education,
Texas Tech University Lubbock, TX 79409-1071,
USA. Email: [email protected]
Classrooms full of “digital natives” represent the norm in U. S. schools, but like their predecessors, they
mostly inhabit spaces characterized by a traditional view of teaching and learning. Understanding
contributors to this mismatch, and especially teachers’ role, is especially critical as Web 2.0 technologies
enable greater learner autonomy. This study used Theory of Planned Behavior (TPB) to predict pre-
service teachers’ (n = 172) intention to use new web-based technologies for instruction. The participants
were from 7 sections of a technology application course, and all of them were selected via a convenience
sampling. Self-reported data were analyzed using a structural equation modeling method, Partial Least
Squares (PLS). Results revealed that the attitude and perceived control components of the theory
predicted intention to use Web 2.0 technology to a statistically and practically significant degree, but
subjective norm did not. Antecedent predictors of the TPB constructs were all predictive. Discussion
focuses on the utility of TPB to examine how pre-service and in-service preparation might use the model to
create situations where technologies might be more widely applied in the classroom.
JONGPIL CHEON, FANNI COWARD, JAEKI SONG, AND SUNHO LIM
Fall 2012 18 RESEARCH IN THE SCHOOLS
technologies provide a variety of services for
collaboration, contribution, and communication that
enable users to produce and to share information with
others (Mills, 2007). These characteristics could
enhance teaching and learning by enabling various
student-centered learning activities requiring active
participation and collaboration (Cheon, Song, Jones,
& Nam, 2010; McLoughlin & Lee, 2008; Williams &
Chinn, 2009). Many researchers anticipate that use of
Web 2.0 resources has greater benefits including (a)
increasing students’ creativity when reasoning and
problem-solving, (b) enhancing interaction when
collaborating, (c) helping to sustain engagement, and
(d) developing communities for professional growth
(cf. International Society for Technology in
Education, 2008; McLoughlin & Lee, 2008;
Vaughan, Nickle, Silovs, & Zimmer, 2010). The
interactive affordances of Web 2.0 technologies
make them easily adopted tools in a student-centered
classroom (Grant & Mims, 2009).Learning that is
guided by Web 2.0 functions also is closely aligned
with a view of learning embraced by social
constructionism, which posits that learners construct
knowledge through social interaction. Researchers
are understandably eager to explore the potential of
Web 2.0 technologies to enhance student-centered
learning, and educators are eager to utilize the
technology to facilitate active participation.
Factors Related to Teachers’ Technology
Adoption Researchers have argued that educational
reforms often fail because teachers’ beliefs,
capabilities, and routine practices are not aligned
with the changes (Niederhauser & Stoddart, 2001).
The slow pace at which instructional technology, let
alone Web 2.0 resources, is adopted might be a case
in point. Thus, it is important to investigate how
technologies are being adopted and why the
technology adoption speed is slow. Currently,
there is a fair amount of consensus around factors
and barriers that directly or indirectly affect teachers’
adoption of instructional technology (e.g., Becker,
2000, 2001; Mumtaz, 2000; Sheingold & Hadley,
1990; Shiue, 2007). Factors that influence adoption
include: (a) individual perceptions (e.g., motivation,
commitment, or confidence); (b) extent of personal
experience; (c) technological issues (e.g., technical
supports, software quality of software and hardware);
and (d) environmental supports (e.g., access to
resources, incentives to change or school polices)
(Evans-Andris, 1995; Mumtaz, 2000). Among these
influences, positive attitudes towards technologies
seems to be the most widely recognized pre-
condition for teachers’ use of technologies for their
teaching (Shiue, 2007). Teachers with positive
attitudes are more apt to utilize technology for
instruction (e.g., Moseley & Higgins, 1999; Russell,
Bebell, O'Dwyer, & O'Connor, 2003; Smerdon et al.,
2000; Yildirim, 2000), and to have a higher
frequency of technical use in the classroom (e.g.,
Pedretti, Smith-Mayer, & Woodrow, 1999; Ravitz,
Wong, & Becker, 1999; Russell et al., 2003; van
Braak, Tondeur, & Valcke, 2004).
Technology and Teacher Preparation It is common for teacher educators to include
some degree of technology competency in their
teacher preparation programs (Shoffner, 2009).
Similar to the studies involving in-service teachers,
pre-service teachers’ self-efficacy or attitude toward
general technologies has been found to be a
significant predictor of technology use (e.g., Littrell,
Zagumny, & Zagummy, 2005; Wozney, Venkatesh,
& Abrami, 2006). Researchers have shown that
educational technology courses can improve pre-
service teachers’ computer self-efficacy (Abbitt &
Klett, 2007). Many researchers have pointed to the
need to integrate pedagogical methods and content
knowledge with technology skills (Abbitt, 2011;
Lambert & Cuper, 2009), but concerns often are
raised about the stand-alone nature of many
educational technology courses for pre-service
teachers. These courses often focus exclusively on
technical skill development in isolation instead of
fostering an understanding of technology as powerful
tools to increase student achievement and to improve
learning. The result of this divorce between technical
and learner-centered views is that integration of
technology in education is still lacking (Vockley,
2008).
At present, it is difficult to synthesize findings
across studies in a way that provides a systematic
picture of how the various beliefs and attitudes might
be related to intention and behavior. This lack of a
systematic view presents a particular challenge to
teacher education programs struggling to focus
technology preparation on improving student
achievement instead of building isolated skills. The
Theory of Planned Behavior offers a framework for
studying the relations among beliefs, attitudes, and
intentions to use Web 2.0.
Theory of Planned Behavior and Teaching
with Web 2.0
Theory and a Research Model The Theory of Planned Behavior (TPB) has been
applied in a wide variety of contexts (e.g.,
technology, health care, and political science) to
predict behavior (e.g., Cheon, Lee, Crooks, & Song,
2012; Conner & Armitage, 1998; Davis, 1989;
Sheppard, Hartwick, & Warshaw, 1988; Song &
Zahedi, 2005; Taylor & Todd, 1995). The theory
posits that individuals are inclined toward a behavior
based on three determinant intentions (Ajzen, 1991).
The first is an individual’s attitude toward the
behavior. Attitude refers to the degree to which a
person has a favorable or unfavorable feeling about
performing the particular behavior. The second is a
FACTORS PREDICTING PRE-SERVICE TEACHERS’ ADOPTION OF WEB 2.0 TECHNOLOGIES
Fall 2012 19 RESEARCH IN THE SCHOOLS
subjective norm. This norm reflects perceived
social pressure to perform or not to perform the
behavior. Finally, perceived behavioral control
reflects the individual’s belief about how easy it is to
perform the behavior. Based on TPB, we
formulated our research model (see Figure 1) to
predict pre-service teachers’ intention to adopt Web
2.0 technologies. For this study, we used teachers’
behavioral intention to use Web 2.0 technologies as
the final dependent variable based on
recommendations from Ajzen (1985, 1991). The
research model proposes that external beliefs
reinforce the three attitudinal constructs. These
attitudinal constructs, in turn, positively enhance
behavioral intention. The antecedent variables of
each attitudinal construct (shown in the left stem) are
described in Figure 1. Specific hypotheses follow
from the model.
Figure 1. Conceptual model for pre-service teachers’ adoption of Web 2.0 technologies
Attitudinal Constructs and Behavioral Intention In the TPB, attitude toward a behavior,
expectations based on a subjective norm, and
perceived control are all predictors of individuals’
intention to act. Researchers have repeatedly
demonstrated the connection between attitude and
intention (Davis, 1989; Notani, 1998; Taylor &
Todd, 1995). Subjective norm, too, has been linked
to intentions to act. For example, Ajzen (1991)
reported that 15 out of 19 TPB-based studies
showed that subjective norm had significant effects
on behavioral intention. Lastly, perceived
behavioral control also has been associated with
intention to act. Previous literature has suggested
that perceived behavioral control functions as a
crucial determinant of behavior when the behavior
cannot be performed without certain knowledge,
skills, resources, cooperation from others, or
opportunities (Ajzen, 1985; Hartwick & Barki,
1994; Sheppard et al., 1988).
Based on this prior work, we hypothesized:
H1: Pre-service teachers’ attitudes toward
Web 2.0 technologies is positively related to
their intentions to adopt Web 2.0 technologies;
H2: Pre-service teachers’ subjective norm
of Web 2.0 technologies is positively related to
their intentions to adopt Web 2.0 technologies;
H3: Pre-service teachers’ perceived
behavioral control of Web 2.0 technologies is
positively related to their intentions to adopt
Web 2.0 technologies.
Beliefs
Attitudinal Constructs
Behavioral Intention
Attitudinal Beliefs
Perceived Ease of Use H5
Perceived Usefulness H6
Normative Beliefs
Perceived Pressure
Attitude
Subjective
Norm Intention
H1
H2
H3
Perceived Behavioral
Control
H7
H8
H9
H10
Control Beliefs
Perceived General
Self-efficacy
Perceived Web 2.0 Specific
Self-efficacy
H4
JONGPIL CHEON, FANNI COWARD, JAEKI SONG, AND SUNHO LIM
Fall 2012 20 RESEARCH IN THE SCHOOLS
Antecedents of Attitude, Subjective Norm and
Perceived Control
Beliefs predict attitude. The antecedents of
the first attitudinal construct (i.e., attitude) are
attitudinal beliefs. Attitudes are developed from the
beliefs associated with attributes and outcomes
(Fishbein & Ajzen, 1975). For example, a person
will have a favorable attitude towards a new
technology when he/she believes that the
technology results in better job performance
(Taylor & Todd, 1995). In our research model,
variables for attitudinal beliefs were derived from
the Technology Acceptance Model (TAM; Davis,
1989). This model has been adapted by many
researchers to investigate the relation between
users’ perceptions and their actual use of a
technology system (Adams, Nelson, & Todd, 1992;
Benbasat & Barki, 2007; Segars & Grover, 1993).
Davis (1989) concluded that perceived ease of use
and perceived usefulness determine an individual's
intention to use a system.
Based on this work, we also included ease of
use and usefulness in our model and hypotheses.
H4: Pre-service teachers’ perceived ease
of use Web 2.0 technologies is positively
related to perceived usefulness of Web 2.0
technologies;
H5: Pre-service teachers’ perceived ease
of use of Web 2.0 technologies is positively
related to their attitudes toward Web 2.0
technologies;
H6: Pre-service teachers’ perceived
usefulness of Web 2.0 technologies is
positively related to their attitudes toward Web
2.0 technologies.
Normative beliefs toward subjective norm.
The representation of how others will view a
behavior reflects the individual’s subjective norm.
Attitudes reflective of the subjective norm account
for the social pressure an individual feels when
contemplating a particular behavior. Individuals are
dependent on context, and they are socially
constructed beings (Shah, 1998). The influence
from referent groups (e.g., friends, superiors,
family members, and neighbors) affects
individuals’ subjective norm (Taylor & Todd,
1995). For pre-service teachers, the pressure to
adopt new technologies could come from their
classes, peers, and instructors. Therefore, we
hypothesized:
H7: Pre-service teachers’ perceived
pressure of Web 2.0 technologies positively
influences their subjective norms with Web 2.0
technologies.
Control beliefs toward perceived behavioral
control. Perceived behavioral control is compatible
with the concept of self-efficacy. Self-efficacy
refers to individuals' beliefs about their ability and
motivation to perform specific tasks (Bandura,
1986, 1997). Previous researchers have
documented that higher levels of self-efficacy lead
to higher levels of behavioral intention and the
usage of information technology (Compeau &
Higgins, 1995; Gist, Schwoerer, & Rosen, 1989).
In the field of information technology,
researchers now distinguish between general
computer self-efficacy and task-specific self-
efficacy (Agarwal & Karahanna., 2000; Marakas,
Mun, & Johnson, 1998). General computer self-
efficacy is defined as an individual's judgment of
efficacy across multiple computer application
domains, whereas task-specific computer self-
efficacy is defined as perceptions of ability to
perform specific computer-related tasks. In the
context of Web 2.0 technologies, this study
employed these discrete concepts of self-efficacy
as control beliefs that influence perceived
behavioral control.
Based on these research findings, we
hypothesized:
H8: Pre-service teachers’ perceived
general self-efficacy of Web 2.0 technologies
is positively related to their perceived Web 2.0
specific self-efficacy.
H9: Pre-service teachers’ perceived
general self-efficacy is positively related to
their perceived behavioral control.
H10: Pre-service teachers’ perceived Web
2.0 specific self-efficacy is positively related to
their perceived behavioral control.
In summary, we hypothesized that TPB would
explain pre-service teachers’ intention to adopt
Web 2.0.
Method
Participants A total of 172 pre-service teachers at a large
university in the southwestern United States
participated in this study. The pre-service teachers
were required to take a technology application
course, and all of them in seven sections in the
course were selected as a convenience sample. The
participants were mostly in their junior or senior
years of post-secondary education. Their average
age was 23, and 82% were female. Their major
courses of study were varied and included early
childhood, mathematics education, history, language
arts, and special education. The Web 2.0
technologies that they had used mostly represented
social networking services (e.g., Facebook) and
video sharing service (e.g., YouTube), and 15
participants had used Google Docs. However, the
participants had not used any other Web 2.0
technologies presented in this study.
FACTORS PREDICTING PRE-SERVICE TEACHERS’ ADOPTION OF WEB 2.0 TECHNOLOGIES
Fall 2012 21 RESEARCH IN THE SCHOOLS
Procedures
The participants took a Web 2.0 special
training session, “Web 2.0 technologies for the
Classroom,” led by one of the authors. Four
categories were selected for the training among the
eight categories of Web 2.0 technologies (Franklin
& van Harmelen, 2007), because these services had
the potential to be a learning tool in the classroom
for pre-service teachers. In addition, they had more
collaborative functions than did other categories,
and they were relatively new to the participants.
One site for each category was selected based on
popularity and free functionality, and the last site
(Voicethread) was presented in addition. This site
does not belong to any Web 2.0 category, but it
provides options for group conversation that would
be useful for collaborative learning. The
participants had heard of Web 2.0, but none of
them knew these websites except Google Docs. A
total of five sites were introduced in the training
session as follows;
• Wikis – Pbworks (http://pbworks.com)
• Social bookmarking – Diigo
(http://diigo.com)
• Media-sharing services -Slideshare
(http://www.slideshare.net)
• Collaborative editing tools – Google Docs
(http://doc.google.com)
• Group conversation - Voicethread
(http://voicethread.com)
The training session consisted of three phases.
In the first phase, the trainer introduced the
dynamic nature of Web 2.0 technologies and the
potential benefits of using them for teaching and
learning. For the second phase, the five sites above
were introduced to the participants. The trainer
showed an introductory video clip for each site, and
basic functions were demonstrated. In the last
phase, the participants were asked to complete the
survey described below.
Instrument Development
We compiled a 27-item measure (three items
for nine constructs) from already-existing
questionnaire items found in published literature
(see Table 1). Participants responded using a 7-
point Likert-type scale, ranging from totally
disagree to totally agree. The first five constructs,
in the first column in Figure 1, are related to the
pre-service teachers' salient beliefs. The three
constructs in the second column are the attitudinal
constructs. The last construct is behavioral
intention to adopt Web 2.0 technologies.
To assess the score-validity of our instrument,
we administered a voluntary pilot test with 10
participants. Each of the 10 participants was asked
to (a) complete each item on the instrument to
indicate how they rated the importance of each
belief about Web 2.0 technologies; and (b) provide
comments on the wording, instruction, and clarity
of the items on the instrument. Based on feedback
from participants in the pilot test, we made minor
modifications to the survey items.
Data Analysis and Results
Partial Least Squares (PLS) method was used
to analyze the data. PLS is “a structured equation
modeling estimation technique which generates
estimation of item loadings (i.e., structure
coefficients) and path coefficients simultaneously”
(Gefen & Straub, 2005, p. 109). The PLS method is
component-based and formative analysis for
prediction rather than fit, whereas other structural
equation modeling (SEM) methods, such as
LISREL or AMOS, focus on testing the degree of
consistency between a model and data (Bontis,
Crossan, & Hulland, 2002; Chin, 1998). Further,
the PLS method can model latent constructs under
conditions of non-normality and small to medium
sample sizes (Chin, Marcolin, & Newsted, 1996,
Hulland, 1999). We completed data analysis with
SmartPLS in two steps: a measurement model
estimate, followed by a structural model evaluation
(Jarvenpaa, Shaw, & Staples, 2004).
Measurement Model Estimate
In order to estimate our measurement model,
we conducted a confirmatory factor analysis with
SmartPLS. Table 2 presents the relevant statistics.
For all constructs, the item score reliability and
convergent validity were evaluated by examining
the individual item loading (i.e., structure
coefficients), composite reliability (CR), and
average variance extracted (AVE). As shown in
Table 2, all structure coefficients on the respective
construct were higher than 0.7, with a range from
0.786 to 0.958 (Chin, 1998), and t values for the
items were statistically significant. The composite
reliability for each construct was greater than 0.7
(Nunnally & Bernstein, 1994). The values of
AVE were all above 0.50 (Fornell & Larcker,
1981). Thus, convergent validity was adequately
established.
The next step was to evaluate the discriminant
validity of the constructs. The square root of the
average variance extracted (AVE) for a given
construct was compared with the correlations
between the construct and other constructs (Fornell
& Larcker, 1981). If the square root of the AVE of
a construct is larger than its correlations with any
other construct, a construct is more closely related
with its indicators than with the other constructs
(Barclay, Thompson, & Higgins, 1995; Chin, 1998;
Fornell & Larcker, 1981; Hulland, 1999). Because
the square root of the AVE for each construct
exceeded the correlations between that construct
and other constructs, discriminant validity appeared
to be satisfactory for all constructs.
JONGPIL CHEON, FANNI COWARD, JAEKI SONG, AND SUNHO LIM
Fall 2012 22 RESEARCH IN THE SCHOOLS
Table 1
Survey Instrument and References
Constructs Items Survey items References
Perceived
Ease of Use
ease1 I believe that the services given by Web 2.0 are easy to use.
Davis, 1989;
Venkatesh & Davis, 2000 ease2
I believe that the services given by Web 2.0 are easy to get to
do what I want them to do.
ease3 I believe that the services given by Web 2.0 are easy to
operate.
Perceived
Usefulness
useful1 I believe that the services given by Web 2.0 allow me to get
my work done more quickly.
Davis, 1989;
Venkatesh & Davis, 2000 useful2
I believe that using the services given by Web 2.0 enhances
my effectiveness on my teaching.
useful3 I believe that the services given by Web 2.0 are useful for
my teaching.
Perceived
Pressure
pres1 I believe that the preparation courses for being a teacher
require me to adopt the services given by Web 2.0. Karahanna, Straub, &
Chervany, 1999;
Iacovou Benbasat, &
Dexter, 1995;
Premkumar & Roberts,
1999
pres2 Instructors of the courses I have taken think I should
integrate the services given by Web 2.0 into the classroom.
pres3 My peers in this course I am taking think I should use the
services given by Web 2.0 for teaching.
General
Self -Efficacy
eff1 I would be able to manage the services given by Web 2.0.
Compeau & Higgins,
1995 eff2
I am confident about managing the services given by Web
2.0.
eff3 Learning the services given by Web 2.0 is straightforward.
Web2.0
Specific Self-
Efficacy
web1 I would be able to enhance my teaching through student
collaboration with the services given by Web 2.0.
O’Reilley, 2005;
Mills, 2007 web2
I am confident that I can increase students’ participation with
the services given by Web 2.0.
web3 I would enrich communications with student and between
students with the services given by Web 2.0.
Attitude
att1 Adopting the services given by Web 2.0 would make me feel
good. Taylor & Todd, 1995;
Song & Zahedi, 2005
att2
If I adopt the services given by Web 2.0 in my future
teaching, I would feel that I am getting a good deal.
att3 Adopting the services given by Web 2.0 would be pleasant.
Subjective
Norm
snorm1 Other people have provided supporting evidence in favor of
adopting the services given by Web 2.0. Taylor & Todd, 1995;
Song & Zahedi, 2005
snorm2 Most people who are important to me think that it is fine to
adopt the services given by Web 2.0.
snorm3 Most people who are important to me would adopt the
service given by Web 2.0.
Perceived
Behavioral
Control
Becon1 The extent of knowledge that I feel I have in adopting Web
2.0 technologies is sufficient. Taylor & Todd, 1995;
Song & Zahedi, 2005
Becon2 The extent of control that I feel I have in making my
adopting decision is sufficient.
Becon3 The extent of self-confidence that I feel I have in making my
adopting decision is sufficient.
Behavioral
Intention
Beint1 I plan to use the services given by Web 2.0 sometime after
becoming a teacher. Song & Zahedi 2005;
Venkatesh, Brown,
Maruping, & Bala, 2008
Beint2 My use of the services given by Web 2.0 after becoming a
teacher would be probable.
Beint3 The likelihood that I would adopt the services given by Web
2.0 is high.
FACTORS PREDICTING PRE-SERVICE TEACHERS’ ADOPTION OF WEB 2.0 TECHNOLOGIES
Fall 2012 23 RESEARCH IN THE SCHOOLS
Table 2
Statistics of the Measurement Model
Construct Item Mean Standard
Deviation
Item
Loadinga
Standard
Error t value
Composite
Reliability AVE
Perceived
Ease of Use
ease1 5.895 1.054 0.935 0.024 15.709
.957 0.882 ease2 5.727 1.015 0.942 0.025 13.607
ease3 5.733 0.978 0.941 0.026 13.494
Perceived
Usefulness
useful1 5.407 1.159 0.786 0.034 9.236
.910 0.773 useful2 5.587 1.270 0.929 0.026 15.398
useful3 5.820 1.091 0.915 0.026 15.855
Perceived
Pressure
pres1 5.378 1.281 0.814 0.044 7.822
.896 0.742 pres2 5.250 1.415 0.885 0.037 10.430
pres3 5.128 1.300 0.884 0.049 8.792
General
Self -Efficacy
eff1 5.913 1.042 0.936 0.022 16.890
.938 0.835 eff2 5.785 1.068 0.915 0.023 15.448
eff3 5.756 1.159 0.890 0.032 11.738
Web2.0
Specific Self-
Efficacy
web1 5.581 0.973 0.881 0.022 15.753
.935 0.828 web2 5.738 0.983 0.916 0.019 19.430
web3 5.750 1.009 0.931 0.024 15.912
Attitude
att1 5.407 1.183 0.893 0.018 19.312
.928 0.811 att2 5.605 1.101 0.917 0.019 20.995
att3 5.587 1.059 0.892 0.018 20.483
Subjective
Norm
snorm1 5.233 1.151 0.791 0.038 9.349
.903 0.756 snorm2 4.727 1.294 0.923 0.025 15.569
snorm3 4.837 1.306 0.889 0.033 12.075
Behavioral
Control
Becon1 5.390 1.152 0.908 0.019 18.918
.937 0.833 Becon2 5.570 1.032 0.916 0.016 23.031
Becon3 5.698 0.974 0.914 0.019 20.446
Behavioral
Intention
Beint1 5.983 1.142 0.958 0.020 18.218
.960 0.889 Beint2 5.890 1.182 0.916 0.022 14.941
Beint3 5.849 1.160 0.954 0.022 16.725 a Structure coefficient
Structural Model Evaluation
The structural model analyzed the relations
among the various latent variables. Figure 2
presents the standardized path coefficients and the
explained construct variances. The results show
that nine of the 10 hypotheses were supported.
Hypotheses 1 and 3 were supported (H1,
coefficient of 0.50, t = 4.52, p < 0.01; H3,
coefficient of 0.37, t = 3.83, p < 0.01), whereas the
second hypothesis was not supported. In other
words, behavioral intention was positively related
to attitude and perceived behavioral control but not
to subjective norm, although the relation between
the two constructs was positive (t = 0.34).
All hypotheses regarding the relations between
the three attitudinal constructs and the antecedent
variables for each construct were supported. First,
both perceived ease of use (H5, coefficient of 0.25,
t = 2.97, p < 0.01) and perceived usefulness (H6,
coefficient of 0.56, t = 7.08, p < 0.01) were
statistically significantly related to attitude. Also,
perceived usefulness was related to a statistically
significant degree to perceived ease of use (H4,
coefficient of 0.61, t = 7.89). Second, the
assumption of the positive relation between
perceived pressure and subjective norm was met
(H7, coefficient of 0.62, t = 8.10, p < 0.01).
Last, we found evidence to support
Hypotheses 8 to 10. For example, perceived
general Web 2.0 self-efficacy was positively
related to perceived Web 2.0 specific self-efficacy
(H8, coefficient of 0.58, t = 8.34, p < 0.01). Both
self-efficacy perceptions were positively related to
perceived behavioral control (H9, coefficient of
0.19, t = 2.41, p < 0.05; H10, coefficient of 0.73, t
= 9.92, p < 0.01), although the relationship
involving Web 2.0 specific self-efficacy was much
greater than was the relationship involving general
self-efficacy.
JONGPIL CHEON, FANNI COWARD, JAEKI SONG, AND SUNHO LIM
Fall 2012 24 RESEARCH IN THE SCHOOLS
*: p < 0.05, **: p < 0.01, N.S.: Not significant
Figure 2. Results of the structural model with PLS.
Discussion
Our research model shows how pre-service
teachers’ perceptions are related to their behavioral
intention to adopt Web 2.0 technologies. We
believe that pre-service teachers who have strong
intention likely will adopt Web 2.0 technologies in
their future classrooms because the behavioral
intention is assumed to be the immediate
antecedent of actual behavior (Ajzen, 2002).
The results suggested that pre-service teachers’
attitudes and perceived behavioral control
positively influenced their intention to adopt Web
2.0 technologies, whereas the impact of their
subjective norm on behavioral intention was not as
strong. These findings have important theoretical
and practical implications, and suggest directions
for future research.
Support for the TPB and Implications for
Teacher Preparation
Usefulness of technology influences attitude.
The relations among beliefs (i.e., perceived ease of
use and usefulness), attitude and behavioral
intention confirmed the findings from Cheon et
al.’s (2010) study that employed the Expectancy-
Value Model and Technology Acceptance Model
(TAM). In line with that study, our findings
demonstrated that perceived ease of use and
usefulness are significant predictors of new
technology (i.e., Web 2.0 technologies) acceptance.
In other words, pre-service teachers who believe
that Web 2.0 services are easy to use and useful are
Attitudinal Beliefs
Perceived Ease of Use
Perceived Usefulness
Normative Beliefs
Perceived Pressure
Attitude
Subjective
Norm Intention
Perceived Behavioral
Control
Control Beliefs
Perceived General
Self-efficacy
Perceived Web 2.0 Specific
Self-efficacy
0.61(7.89) **
0.25(2.97) **
0.56(7.08) **
0.62(8.10) **
0.19(2.41) *
0.73(9.92) **
0.58(8.34) **
0.50(4.52) **
0.03(0.34)
0.37(3.83)**
N.S.
FACTORS PREDICTING PRE-SERVICE TEACHERS’ ADOPTION OF WEB 2.0 TECHNOLOGIES
Fall 2012 25 RESEARCH IN THE SCHOOLS
more likely to adopt Web 2.0 service for teaching.
Because the coefficient value of perceived
usefulness (β = 0.57) was higher than was the value
of perceived ease of use (β = 0.25), it can be
concluded that the perception toward usefulness is
a stronger predictor than is attitude. We assume
that the difference might be caused by the
affordance of Web 2.0 technologies.
Most Web 2.0 technologies focus on a specific
function that requires a low learning curve (Grant
& Mims, 2009). For example, users can be easily
familiar with the core functions via a short
introduction video clip. However, how to apply
these services to teaching and learning is not
simple.
As shown by the higher coefficient value of
perceived usefulness, teacher educators should
provide successful learning experiences using a
Web 2.0 site in a wider variety of teacher
preparation courses in addition to a few courses
about technology application (Ertmer, 2005;
McRobbie, Ginns & Stein, 2000; Pope, Hare &
Howard, 2002). In order for pre-service teachers to
believe that Web 2.0 service provides a valuable
teaching tool, teacher educators should strengthen
their beliefs about its usefulness through learner-
centered activities and rich resources. Doing so
might reduce uncertainty or hesitation in pre-
service teachers’ perceptions about technology
(Cheon et al., 2010).
Efficacy, perceived control and intention to
act. We found that pre-service teachers’ behavioral
control is another important predictor of their
intention to adopt Web 2.0 technologies. Both
general self-efficacy and specific self-efficacy were
positively related to their perceived behavior
control and, in turn, the behavioral control
predicted positive intention. Pre-service teachers
who perceived more control over Web 2.0
technologies indicated higher intention to adopt the
technologies.
Perceptions of control over the technology
were related to both self-efficacy levels.
Consistent with the findings of prior research,
general self-efficacy was positively related to the
specific task-specific (i.e., Web 2.0 technologies)
self-efficacy. Because the results confirmed the
assumption in TPB, this finding implies that
enhancing pre-service teachers’ confidence in using
Web 2.0 technologies would increase the likelihood
of technology adoption. More specifically, teaching
strategies or training mechanism to increase self-
efficacy likely would be beneficial to pre-service
teachers who have lower technology self-efficacy.
Instead of using the same tool or software in a
course, for example, the freedom of selecting a tool
based on individual preference or capacity likely
would enhance learning autonomy and decrease the
level of anxiety in teacher preparation courses.
Subjective norm and intention to act. This
study did not find support for the relation between
subjective norm and behavioral intention, although
perceived pressure served as a significant
antecedent of subjective norm. We conjecture that
the finding is due to the environments of pre-
service teachers. Because they were not in real
practice, they might not feel any pressure to
integrate technology including Web 2.0
technologies into curricula. This finding is
consistent with what Shiue (2007) reported about
subjective environments weakly influencing the
actual use of technology. Subjective norm,
however, might be more important for in-service
teachers. The norm in a real school setting would
be influenced by more determinants such as
administrative support or pressure and peer use of
technology (e.g., Shiue, 2007).
As teacher educators, we can create a positive
social norm within pre-service teachers’ society.
Previous researchers documented that in-service
teachers tend to use the computer as a tool to
engage students in cognitively challenging tasks if
they regularly participate in professional
interactions beyond their classrooms (Becker &
Riel, 2000). Accordingly, if pre-service teachers
have experiences of producing and sharing
meaningful information with others, their sense of
belonging and kinship could create a more positive
social norm. For example, creating an online
community of pre-service teachers in a university
could be the first step.
General Implications for Teacher Preparation
The proposed adoption model describes a set
of factors related to pre-service teachers’ use of
Web 2.0 technologies. Teacher education programs
could manipulate those factors that facilitate
teacher candidates’ adoption of the use of Web 2.0
technologies. The findings about the importance of
attitude and perceived behavioral control suggest
that teacher education curricula should include
opportunities to learn and to utilize various
functions of Web 2.0 technologies. With more Web
2.0 integration experiences, teacher candidates
should have positive attitudes and higher levels of
confidence to use technology tools.
Limitations and Future Research Directions
This study has at least two limitations that
circumscribe our interpretation and implications.
First, from a methodological standpoint, the
findings should be interpreted with caution due to
the fact that our results were obtained from a
single, self-report questionnaire, and the sample
size was relatively small. In terms of the
intervention, the training session might have been
too short to provide pre-service teachers with
adequate hands-on experiences; thus, it might limit
JONGPIL CHEON, FANNI COWARD, JAEKI SONG, AND SUNHO LIM
Fall 2012 26 RESEARCH IN THE SCHOOLS
the generalizability of the results to other
situations.
We propose several exciting directions for
future research. First, although we included several
antecedent variables, there are others in the
research literature that should be included in future
studies. For example, social pressure as a
determinant of subjective norm might have only
negative connotations, and positive variables such
as support from instructors or colleagues could be
added to the research model. In addition, the
general self-efficacy variables focused on only
Web 2.0 technologies. More general computer self-
efficacy also should be considered to develop a
more systematic and robust conceptual model.
Next, we find it very intriguing to consider a
longitudinal perspective and imagine how the TPB
might help in understanding the emergence of
beliefs, attitudes, and intentions among pre-service
teachers. In addition, this perspective could be
applied to in-service teachers in order to determine
significant predictors of their intentions to integrate
technology. Shiue (2007) discovered other
significant factors (e.g., technical support,
computer access, and professional development)
for technology integration using a decomposed
theory of planned behavior. Further researchers
could produce a revised model for technology
integration that extends the original theory.
Conclusion
The teaching profession is in need of new
teachers with positive attitudes about technology
and the skills to apply them in the classroom
(Russell et al., 2003). This study provides an
analysis of what might motivate new teachers’
technology use via the Theory of Planned
Behavior. Our empirical analysis of pre-service
teachers’ various beliefs provide a picture of what
might influence their attitudes and intentions to
adopt Web 2.0 technologies. Further studies are
needed to explore the pedagogical ways to increase
teacher candidates’ levels of self-efficacy and to
create a supportive social norm.
The lead editors for this article were Anthony J.
Onwuegbuzie and John R. Slate.
References
Abbitt, J. T. (2011). An investigation of
relationship between self-eficacy beliefs
about technology integration and
technological pedagogical content
knwoledge (TPACK) among preservice
teachers. Journal of Digital Learning in
Teacher Education, 27(4), 134-143.
Abbitt, J. T., & Klett, M. D. (2007). Identifying
influences on attitudes and self-efficacy
beliefs towards technology integration
among preservice educators. Electronic
Journal for the Integration of Technology
in Education, 6, 28-42.
Adams, D. A., Nelson, R. R., & Todd, P. A. (1992).
Perceived usefulness, ease of use, and
usage of information technology: A
replication. MIS Quarterly, 16, 227-247.
doi:10.2307/249577
Agarwal, R., & Karahanna, E. (2000). Time flies
when you’re having fun: Cognitive
absorption and beliefs about information
technology usage. MIS Quarterly, 24, 665-
694. doi:10.2307/3250951
Ajzen, I. (1985). From intentions to action: A
theory of planned behavior. In J. Huhl &
J. Beckmann (Eds.), Action-control: From
cognition to behavior (pp. 11-39). Berlin,
Germany: Springer-Verlag.
Ajzen, I. (1991). The theory of planned behavior.
Organizational Behavior and Human
Decision Processes, 50, 179-211.
doi:10.1016/0749-5978 (91)90020-T
Ajzen, I. (2002). Perceived behavioral control, self-
Efficacy, locus of control, and the theory
of planned behavior. Journal of Applied
Social Psychology, 32, 665-683.
doi:10.1111/j.1559-1816.2002.tb00236.x
Bandura, A. (1986). Social foundations of thought
and action: A social cognitive theory.
Englewood Cliffs, NJ: Prentice-Hall.
Bandura, A. (1997). Self-efficacy: The exercise of
control. New York, NY: W. H. Freeman.
Barclay, D. W., Thompson, R., & Higgins, C.
(1995). The partial least squares (PLS)
approach to causal modeling: Personal
computer adoption and use an illustration.
Technology Studies, 2, 285-309.
Becker, H. J. (2000). Findings from the teaching,
learning and computing survey: Is Larry
Cuban right? Report from a National
Study. Irvine, CA: Center for Research on
Information Technology and
Organizations. Retrieved from
http://www.crito.uci.edu/tlc/findings/ccsso
Becker, H. J. (2001, April). How are teachers using
computers in instruction? Paper presented
at the annual meeting of the American
Educational Research Association, Seattle,
WA.
Becker, H. J., & Riel, M. M. (2000). Teacher
professional engagement and
constructivist-compatible computer use
[on-line]. Centre for Research on
Information Technology and
Organisations, University of California,
FACTORS PREDICTING PRE-SERVICE TEACHERS’ ADOPTION OF WEB 2.0 TECHNOLOGIES
Fall 2012 27 RESEARCH IN THE SCHOOLS
Irvine. Retrieved from
http://www.crito.uci.edu/tic/findings.html
Benbasat, I., & Barki, H. (2007). Quo vadis, TAM?
Journal of the Association for Information
Systems, 8, 211-218.
Bontis, N., Crossan, M., & Hulland, J. (2002),
Managing an organizational learning
system by aligning stocks and flows.
Journal of Management Studies, 39, 437-
469. doi:10.1111/1467-6486.t01-1-00299
Cheon, J., Lee, S., Crooks, S. M., & Song, J.
(2012). An investigation of mobile
learning readiness in higher education
based on theory of planned behavior.
Computers & Education, 59, 1054-1064.
doi:10.1016/j.compedu.2012.04.015
Cheon, J., Song, J., Jones, D. R., & Nam, K.
(2010). Influencing pre-service teachers’
intention to adopt Web 2.0 services.
Journal of Digital Learning in Teacher
Education, 27, 53-64.
Chin, W. W. (1998). Issues and opinion on
structural equation modeling. MIS
Quarterly, 22, 7-16.
Chin, W. W., Marcolin, B. L., & Newsted, P. R.
(1996, December). A partial least squares
latent variable modeling approach for
measuring interaction effects: Results
from a Monte Carlo simulation study and
voice mail emotion/adoption study. Paper
presented at the 17th International
Conference on Information Systems,
Cleveland, OH.
Compeau, D., & Higgins, C. A. (1995). Computer
self-efficacy: Development of a measure
and initial test, MIS Quarterly, 19, 189-
211. doi:10.2307/249688
Conner, M., & Armitage, C. J. (1998). Extending
the theory of planned behavior: A review
and avenues for further research. Journal
of Applied Social Psychology, 28, 1429-
1464. doi:10.1111/j.1559-
1816.1998.tb01685.x
Davis, F. D. (1989). Perceived usefulness,
perceived ease of use, and user acceptance
of information technology. MIS Quarterly,
13, 319-340. doi:10.2307/249008
Ertmer, P. A. (2005). Teacher pedagogical beliefs:
The final frontier in our quest for
technology integration? Educational
Technology Research and Development,
53, 25-39.
Evans-Andris, M. (1995) Barrier to computer
integration: micro-interaction among
computer co-ordinators and classroom
teachers in elementary schools, Journal of
Research on Computing in Education, 28,
29-45.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude,
intention, and behavior: An introduction
to theory and research. Reading, MA:
Addison-Wesley
Fornell, C., & Larcker, D. (1981). Evaluating
structural equation models with
unobservable variables and measurement
error. Journal of Marketing Research, 18,
39-50. doi:10.237/3151312
Franklin, T., & van Harmelen, M. (2007). Web 2.0
for content for learning and teaching in
higher Education. Retrieved from
http://www.jisc.ac.uk/media/documents/pr
ogrammes/digitalrepositories/web2-
content-learning-and-teaching.pdf
Gefen, D., & Straub, D.W. (2005). A practical
guide to factorial validity using PLS-
graph: Tutorial and annotated example.
Communications of AIS, 16(1), 91-109.
Gist, M. E., Schwoerer, C. E., & Rosen, B. (1989).
Effects of alternative training methods on
self-efficacy and performance in computer
software training. Journal of Applied
Psychology, 74, 884-891.
doi:10.1037/0021-9010.74.6.884
Grant, M. M., & Mims, C. (2009). Web 2.0 in
teacher education: Characteristics,
implications, and limitations. In T. Kidd &
I. Chen (Eds.), Wired for learning: An
educators guide to Web 2.0 (pp. 343-360).
Charlotte, NC: Information Age
Publishing.
doi:10.4018/9781605667829.ch007
Hartwick, J., & Barki, H. (1994). Explaining the
role of use participation in information
system use. Management Science, 40,
440-465. doi:10.1287/mnsc.40.4.440
Hulland, J. (1999). Use of partial least squares
(PLS) in strategic management research:
A review of four recent studies. Strategic
Management Journal, 20, 195-204.
doi:10.1002/(SICI)1097-
0266(199902)20:2<195::AID-
SMJ13>3.0.CO;2-7
Iacovou, C. L., Benbasat, I., & Dexter, A. S.
(1995). Electronic data interchange and
small organizations: Adoption and impact
of technology, MIS Quarterly, 19, 465-
485. doi:10.2307/249629
International Society for Technology in Education.
(2008). National Educational Technology
Standards and Performance Indicators for
Teachers. Eugene, OR. Retrieved from
http://www.iste.org/standards/nets-for-
teachers.aspx
Jarvenpaa, S. L., Shaw, T. R., & Staples. D. S.
(2004). Toward contextualized theories of
trust: The role of trust in global virtual
teams. Inform. Systems Research, 15, 250-
267. doi:10.1287/isre.1040.0028
Karahanna, E., Straub, D. W., & Chervany, N. L.
(1999). Information technology adoption
JONGPIL CHEON, FANNI COWARD, JAEKI SONG, AND SUNHO LIM
Fall 2012 28 RESEARCH IN THE SCHOOLS
across time: A cross-sectional comparison
of pre-adoption and post-adoption beliefs.
MIS Quarterly, 23, 183-213.
doi:10.2307/249751
Lambert, J., & Cuper, P. (2009). Multimedia
technologies and familiar space: 21st
century teaching for 21st century learners.
Contemporary Issues in Technology and
Teacher Education, 8, 264-276.
Li, G., Liu, M., Li, G., Wang, Z., & Chen, W.
(2011). Approach of Web 2.0 application
pattern applied to the information
teaching. Communications in Computer
and Information Science, 175, 123-127.
doi:10.1007/978-3-642-21783-8_20
Littrell, A. B., Zagumny, M. J., & Zagumny, L. L.
(2005). Contextual and psychological
predictors of instructional technology use
in rural classroom. Educational Research
Quarterly, 29(2), 37-47.
Marakas, G. M., Mun, Y. Y., & Johnson, R. D.
(1998) The multilevel and multifaceted
character of computer self-efficacy:
Toward clarification of the construct and
an integrative framework for research.
Journal Information Systems Research, 9,
126-163. doi:10.1287/isre.9.2.126
McLoughlin, C., & Lee, M. (2008, December).
Mapping the digital terrain: New media
and social software as catalysts for
pedagogical change. Paper presented at
the ASCILITE Melbourne 2008, Deakin
University, Melbourne, Australia.
McRobbie, C., Ginns, I., & Stein, S. (2000). Pre-
service primary teachers’ thinking about
technology and technology education.
International Journal of Technology and
Design Education, 10(1), 81-101.
doi:10.1023/A:1008941520152
Mills, L. B. (2007). The next wave now: Web 2.0.
Education Digest: Essential Readings
Condensed for Quick Review, 73(4), 4-5.
Mishra, P., & Koehler, M. J. (2006). Technological
pedagogical content knowledge: A
framework for integrating technology in
teacher knowledge. Teachers College
Record, 108, 1017-1054.
doi:10.1111/j.1467-9620.2006.00684.x
Moseley, D., & Higgins, S. (1999). Ways forward
with ICT: effective pedagogy using
information and communications
technology for literacy and numeracy in
primary schools. London, England:
Teacher Training Agency.
Mumtaz, S. (2000). Factors affecting teachers’ use
of information and communications
technology: a review of the literature.
Journal of Information Technology for
Teacher Education, 9, 319-341.
doi:10.1080/14759390000200096
Murugesan, S. (2007). Understanding Web 2.0. IT
Professional, 9(4), 34-41.
doi:10.1109/MITP.2007.78
Niederhauser, D. S., & Stoddart, T. (2001).
Teachers’ instructional perspectives and
use of educational software. Teaching and
Teacher Education, 17(1), 15-31.
doi:10.1016/SO742-051X(00)00036-6
Niess, M. L. (2005). Preparing teachers to teach
science and mathematics with technology:
Developing a technology pedagogical
content knowledge. Teaching and Teacher
Education, 21, 509-523.
doi:10.1016/j.tate.2005.03.006
Niess, M. L., Zee, E. H., & Gilow-Wiles, H.
(2010). Knowledge growth in teaching
mathematics/science with spreadsheets:
moving PCK to TPAC through online
professional development. Journal of
Digital Learning in Teacher Education,
27(2), 42-52.
Notani, A. S. (1998). Moderators of perceived
behavioral control’s predictiveness in the
theory of planned behavior: A meta-
analysis. Journal of Consumer
Psychology, 7, 247-271.
Nunnally, J. C., & Bernstein, I. H. (1994).
Psychometric theory (3rd ed.). New York,
NY: McGraw-Hill.
O’Reilley, T. (2005). What is Web 2.0: Design
patterns and business models for the next
generation of software. Retrieved from
http://www.oreillynet.com/pub/a/oreilly/ti
m/news/2005/09/30/what-is-web-20.html
Pedretti, E., Smith-Mayer, J., & Woodrow, J.
(1999). Teaming technology enhanced
instruction in the science classroom and
teacher professional development, Journal
of Technology and Teacher Education, 7,
131-143.
Pope, M., Hare, D., & Howard, E. (2002).
Technology integration: Closing the gap
between what pre-service teachers are
taught to do and what they can do. Journal
of Technology and Teacher Education, 10,
191-203.
Premkumar, G., & Roberts, M. (1999). Adoption of
new information technologies in rural
small business. Omega International
Journal of Management Science, 27, 467-
484. doi:10.1016/S0305-0483(98)00071-1
Prensky, M. (2001). Digital natives, digital
immigrants. On The Horizon, 9(5), 1-6.
Ravitz, J., Wong, Y., & Becker, H. (1999). Report
to participants. Irvine, CA: Center for
Research on Information Technology and
Organizations. Retrieved from
http://www.crito.uci.edu/tlc/findings/speci
al_report/index.htm
FACTORS PREDICTING PRE-SERVICE TEACHERS’ ADOPTION OF WEB 2.0 TECHNOLOGIES
Fall 2012 29 RESEARCH IN THE SCHOOLS
Russell, M., Bebell, D., O'Dwyer, L., & O'Connor,
K. (2003). Examining teacher technology
use: Implications for pre-service and in-
service teacher preparation. Journal of
Teacher Education, 54, 297-310.
doi:10.1177/0022487103255985
Russell, S. (2009). Taking care of business:
Authentic use of Web 2.0 in schools. In
M. Orey, V.J. McClendon & R.M. Branch
(Eds.), Educational media and technology
year book (Vol. 34, pp. 199-209). Springer
Science & Business Media, LLC.
doi:10.1007/978-0-387-09675-9_14.
Segars, A. H., & Grover, V. (1993). Re-examining
perceived ease of use and usefulness: A
confirmatory factor analysis. MIS
Quarterly, 17, 517-525.
doi:10.2307/249590
Shah, P. P. (1998). Who are employees' social
referents? Using a network perspective to
determine referent others. The Academy of
Management Journal, 41, 249-268.
doi:10.2307/256906
Sheingold, K., & Hadley, M. (1990). Accomplished
teachers: Integrating computers into
classroom practice. New York, NY: Bank
Street College of Education, Center for
Technology in Education.
Sheppard, B. H., Hartwick, J., & Warshaw, P. R.
(1988). The Theory of reasoned action: A
meta-analysis of past research with
recommendations for modifications and
future research. Journal of Consumer
Research, 15, 325-343.
doi:10.1086/209170
Shiue, Y. (2007). Investigating the sources of
teachers’ instructional technology use
through the decomposed theory of planned
behavior. Journal of Educational
Computing Research, 36, 425-453.
doi:10.2190/A407-22RR-50X6-2830
Shoffner, M. (2009). Personal attitudes and
technology: Implications for preservice
teacher reflective practice. Teacher
Education Quarterly, 36, 143-161.
doi:10.1016/j.tate.2008.11.012
Smerdon, B., Cronen, S., Lanahan, L., Anderson,
J., Iannotti, N., & Angeles, J. (2000).
Teachers’ tools for the 21st century: A
report on teachers’ use of technology
(NCES 2000-102). Retrieved from
National Center for Education Statistics
website:
http://nces.ed.gov/pubs2000/2000102.pdf
Song, J., & Zahedi, F. M. (2005). A theoretical
approach to Web design in e-commerce: A
belief reinforcement model, Management
Science, 51, 1219-1235.
doi:10.1287/mnsc.1050.0427
Taylor S., & Todd, P. (1995). Decomposition and
crossover effects in the theory of planned
behavior: A study of consumer adoption
intentions. International Journal of
Research in Marketing, 12, 137-156.
doi:10.1016/0167-8116(94)00019-K
van Braak, J., Tondeur, J., & Valcke, M. (2004).
Explaining different types of computer use
among primary school teachers. European
Journal of Psychology of Education, 19,
407-422. doi:10.1007/BF03173218
Vaughan, N., Nickle, T., Silovs, J., & Zimmer, J.
(2010). Moving to their own beat:
Exploring how students use Web 2.0
technologies to support group work
outside of class time. Journal of
Interactive Online Learning, 10, 113-127.
Venkatesh, V., & Davis, F. D. (2000). A theoretical
extension of the technology acceptance
model: Four longitudinal field studies.
Management Science, 46(2), 186-204.
Venkatesh, V., Brown, S. A., Maruping, L. M., &
Bala, H. (2008). Predicting different
conceptualizations of system use: The
competing roles of behavioral intention,
facilitating conditions, and behavioral
expectation. MIS Quarterly, 32, 483-502.
Vockley, M. (2008). Maximizing the impact: Te
pivotal role of technology in 21st century
education system. Retrieved from
http://www.setda.org/web/guest/maximizi
ngimpactreport
Williams, J., & Chinn, S. J. (2009). Using Web
2.0 to support the active learning
experience. Journal of Information System
Education, 20, 165-174.
Wozney, L., Venkatesh, V., & Abrami, P. C.
(2006). Implementing computer
technologies: Teachers’ perceptions and
practices. Journal of Technology and
Teacher Education, 14(1), 173-207.
Yildirim, S. (2000). Effects of an educational
computing course on pre-service and in
service teachers: A discussion and
analysis of attitudes and use. Journal of
Research on Computing in Education, 32,
479-495.
Copyright 2012 by the RESEARCH IN THE SCHOOLS
Mid-South Educational Research Association 2012, Vol.19, No. 2, 30-44
Fall 2012 30 RESEARCH IN THE SCHOOLS
The Revised School Culture Elements Questionnaire: Gender and Grade Level Invariant?
Thomas A. DeVaney, Nan B. Adams, and Flo Hill-Winstead
Southeastern Louisiana University
Mitzi P. Trahan
University of Louisiana at Lafayette
School culture represents the assumptions,
beliefs, values, and habits that constitute the norm
for a school and shapes how the staff thinks, feels,
and acts. Culture also can be defined as “how we
do things around here” and the stories we tell
ourselves (DuFour, 1998). Deal and Peterson
(1990) referred to culture as deep patterns of
values, beliefs, and traditions that have been
formed over the course of the school's history,
consisting of “stable, underlying social meanings
that shape beliefs and behavior over time” (p. 3).
School cultures are complex webs of
traditions and rituals that have evolved over time
(Deal & Peterson, 1990; Schein, 1985). Senge et
al. (2000) stressed that a school's culture is not
static but a continual process in which attitudes,
values, and skills continually reinforce each other.
The concept of school culture provides school
leaders with a holistic way to look at the school and
school reform. The holistic view of school culture
was also espoused by Schoen (2005) who defined
school culture as “the holistic activities and ‘ways
of being and doing’ of those who work in or
participate on a regular basis within a school” (p.
13). Stolp and Smith (1995) believed that a
deepened understanding of school culture would
allow leaders to influence the values, beliefs, and
underlying assumptions held by school community
members.
The importance of school culture was
illustrated in the 1996 Interstate School Leaders
Licensure Consortium (ISLLC) standards and the
revised standards that were adopted in 2008 by the
National Policy Board for Educational
Administration (Council of Chief State School
Officers, 2008). In Standard 2 of the 1996 ISLLC
Standards, an educational leader is described as an
individual “who promotes the success of all
students by advocating, nurturing, and sustaining a
school culture and instructional program conducive
to student learning and staff professional growth”
(p. 18). The 2008 Standards continue to
recognize the importance of school culture by
stating, “An education leader promotes the success
of every student by advocating, nurturing, and
sustaining a school culture and instructional
program conducive to student learning and staff
professional growth” (p. 14).
The role of culture in the preparation of
educational leaders also was illustrated through
standards developed by the National Policy Board
for Educational Administration and adopted by the
National Council for Accreditation of Teacher
Education for review of Educational Leadership
preparation programs. Stated in the standards that
were published in 2002 was that candidates who
completed Educational Leadership preparation
Correspondence concerning this article should be
addressed to Thomas A. DeVaney, SLU 10549
Hammond, LA 70402. Email: [email protected]
The purpose of this research was to examine the psychometric properties of the RSCEQ with respect to
invariance across gender and grade level, using a sample of 901 teachers from 44 schools in southeast
Louisiana. Reliability estimates were consistent with previous research and ranged from .81to .90 on the
actual and .83 to .92 on the preferred scales, respectively. Confirmatory factor analyses supported the 3-
factor structure of both scales and the theoretical assumption that the factors are correlated. Analyses
also provided evidence that the RSCEQ is invariant across gender and grade level. Additional evidence
that supports the use of the RSCEQ as a tool that educational leaders can use to assess and improve
school culture is provided by the results of this investigation.
THOMAS DeVANEY, NAN ADAMS, FLO HILL-WINSTEAD, AND MITZI TRAHAN
Fall 2012 31 RESEARCH IN THE SCHOOLS
programs should have the knowledge and ability to
promote a positive school culture. Additionally,
completers are expected to have the ability to
assess school culture in order to improve school
programs and culture (National Policy Board for
Educational Administration, 2002). These
standards were revised in 2009 and continue to
illustrate the importance of school culture by
stating that an educational leader should “nurture
and sustain a culture of collaboration, trust,
learning, and high expectations” (National Policy
Board for Educational Administration, 2009, p.
12). The standards continue by stating that an
educational leader must have knowledge of “the
elements of school culture and ways it can be
influenced to ensure student success” (p. 12).
Although the importance of school culture has
been illustrated through professional standards
since 1996, its importance also has been recognized
through research activity since the release of the
1996 ISLLC Standards. A search of the
Education Resources Information Center (ERIC)
database for entries related to school culture in
elementary and secondary education between 1997
and 2012 revealed 525 journal articles and 711
ERIC documents, an average of 77.3 entries per
year.
The importance of school culture, as illustrated
through professional standards and publications,
raises an important question for practitioners and
researchers, “How can school culture be
measured?” Maslowski (2006), in a review of six
quantitative, multi-dimensional inventories
designed to measure school culture, concluded that
the instruments contained a common core of scales
but also contained notable differences. One of the
reviewed instruments was the School Culture
Elements Questionnaire, which contained the
unique approach of assessing perceptions of actual
and preferred school culture. The actual scale
requires individuals to respond with respect to how
things actually occur at their school; whereas on
the preferred scale, individuals respond with
respect to how they would like things to be at their
school.
The original 64-item School Culture Elements
Questionnaire (SCEQ) developed by Cavanagh
(1997) was based on the Cultural Elements Model
of School Culture and designed to assess the
following eight domains of school culture: (a)
professional values, (b) teachers as learners, (c)
collegiality, (d) mutual empowerment, (e)
collaboration, (f) shared vision, (g) school wide
planning, and (h) transformational leadership.
Cavanagh (1997) used a sample of 422 high school
and primary school teachers in Western Australia
to examine the psychometric properties of the
actual scale of the SCEQ. Reliability estimates
for the proposed scales ranged from .62 to .70.
Although the development of the SCEQ was based
on eight domains of culture, Cavanagh’s factor
analysis results supported a 6-factor structure
instead of the proposed 8-factor structure.
Other researchers examined the properties of
the SCEQ with samples of teachers in the
southeastern United States. Dellar, Cavanagh,
and Ellett (1998) adapted the original SCEQ by
adding 14 statements and editing language for
appropriateness. Dellar et al.(1988) used samples
of 3,279 and 1,389 teachers to examine the actual
scale of the Revised SCEQ (RSCEQ). They
showed the RSCEQ to be a useful measure of
school culture and identified three predominant
factors: (a) Shared Leadership and Vision, (b)
Collegial Teaching and Learning, and (c)
Professional Commitment. In addition to analyses
of the total sample, Dellar et al. (1998) examined
the factor structure of the RSCEQ’s actual scale
based on grade level (i.e., elementary, middle, and
high school). Although they identified a common
core of factors, differences among grade levels
were identified with respect to the proportions of
variance explained.
In 2001, Olivier continued the refinement of
the original SCEQ, leading to the creation of a
20-item RSCEQ used to measure professional
school culture. Olivier examined the structure of
the RSCEQ actual scale using a sample of 1,437
elementary level teachers from 12 school districts
in Louisiana. Factor analysis results supported
the 3-factor structure identified by Dellar et al.
(1998) and Ellet, Rugutt, and Cavanagh (2000).
Lorraine (2012) also used factor analysis based on
a sample of 356 teachers to support the 3-factor
structure of the actual scale.
Overall, the validation studies for the SCEQ
and RSCEQ indicate that these instruments
measure constructs representative of general
professionalism and perceived elements of school
culture. However, several limitations have been
identified in previous examinations of the
psychometric properties of these instruments.
First, neither Cavanagh (1997), Olivier (2001), nor
Lorraine (2012) examined the factor structure for
subgroups. It appears that the only researchers to
examine the factor structure for subgroups were
Dellar et al. (1998). Although their results
supported a common factor structure for the
subgroups based on grade level, their research also
contained the same limitation as did that of
Cavanagh, Olivier, and Lorraine. Specifically, the
researchers did not examine the structure of the
preferred scale. Therefore, the purpose of the
current research was to continue the analysis of the
20-item RSCEQ by examining the psychometric
properties of both scales, actual and preferred, with
respect the 3-factor structure and invariance across
gender and grade level.
THE REVISED SCHOOL CULTURE ELEMENTS QUESTIONNAIRE
Fall 2012 32 RESEARCH IN THE SCHOOLS
Method
Procedure
The Revised School Culture Elements
Questionnaire (RSCEQ) was distributed to faculty
in PreK-12 schools in southeast Louisiana as part
of a class project for students enrolled in graduate
courses in Educational Leadership. Because the
distribution of the surveys was a required activity
for the students and different professors were
involved, the distribution involved traditional
paper-and-pencil delivery and electronic delivery.
The method of delivery was determined by the
professor in order to meet the needs of the class.
Regardless of delivery, students were required to
obtain consent from their school to administer the
surveys. For the paper-and-pencil delivery method,
a copy of the survey and cover letter was sent to the
students. They were then required to make the
necessary copies and arrange for the distribution
and collection of the surveys. For students who
used the electronic distribution, an email that
contained a link to the cover letter was sent to the
students. They were then required to distribute the
link to their faculty using an appropriate delivery
method.
After the paper-and-pencil surveys were
returned, they were entered into an Excel
spreadsheet, and random checks were conducted to
determine the accuracy of the data processing.
For the electronic submissions, the data were
exported from www. surveymonkey.com in an
Excel format. Prior to processing the paper-and-
pencil surveys, a sample survey was submitted
electronically, and the results were downloaded in
an Excel format. The downloaded file was used as
the template into which the responses from paper-
and-pencil surveys were entered. The use of the
template ensured an exact match with respect to the
number and order of the columns contained in the
downloaded response files and hand processed
response files. Because the formats of the two types
of files were identical, the files were easily merged
into one file for analysis. The final Excel files
were imported into SPSS 17.0 for the final
analyses. All computed scores were obtained using
the guidelines provided by the RSCEQ developers.
Sample Surveys were completed by 901 teachers in 44
schools throughout parishes in southeast Louisiana.
The number of respondents per school ranged from
a minimum of 1 to a maximum of 46, and 33
schools (75.0%) had at least 10 faculty responses.
In Louisiana, the school performance labels for
schools ranged from Academically Unacceptable to
Five Stars. The schools that are included in the
current analyses ranged from One Star to Four
Stars with 75% of the schools identified as Two or
Three Star schools based on the School Report
Cards available from the Louisiana Department of
Education. The most frequent gender was female
(81.1%), and the most frequently reported highest
degree earned was a Bachelor’s degree (59.3%).
Respondents also reported being certified in the
field in which they taught (89.3%). Table results
also illustrate uniform distributions with respect to
age and number of years at the current school.
Finally, demographics show that the sample
included teachers at the elementary, middle, and
high school levels.
An examination of the data indicated that 98
individuals had one or more missing responses on
the actual scale and 77 individuals had one or more
missing responses on the preferred scale. To
minimize the impact of missing values, individuals
with missing values were omitted from the
analysis. Therefore, the sample sizes were 803 and
824 for the actual and preferred scale analyses,
respectively. An examination of the respondent
characteristics for these samples showed that the
distributions were similar to the overall sample,
which suggests that a systematic pattern to explain
the missing values was not present.
Instrument
The RSCEQ measures perceptions related to
three dimensions: (a) shared leadership and vision,
(b) collegial teaching and learning, and (c)
professional commitment (see Table 1 for
conceptual definitions of the dimensions). For each
dimension, the RSCEQ requires respondents to rate
their actual school culture and how they would
prefer the culture to be at their school. The actual
scale addresses "how I and my school actually are"
and requires participants to respond to the
statements according to their perceptions of how
things actually occur in their schools. Conversely,
the preferred scale indicates participants' opinions
about the school in which they wish to work,
thereby detailing their preferences for
characteristics of an ideal school. Actual and
preferred perceptions are assessed by responding to
20 statements using a 4-point Likert-type scale
ranging from 1 (Strongly Disagree) to 4 (Strongly
Agree).
The questionnaire was formatted into three
columns. The center column contained the 20
statements while the left and right columns
contained the response options for the actual and
preferred scales, respectively (see Appendix A).
THOMAS DeVANEY, NAN ADAMS, FLO HILL-WINSTEAD, AND MITZI TRAHAN
Fall 2012 33 RESEARCH IN THE SCHOOLS
Table 1
Conceptual Definitions of RSCEQ Dimensions
Dimension Definition
Shared Leadership/Vision An ongoing process to accomplish school goals that reflect interpersonal
roles and relationships among organizational members grounded in the
norms, values, and beliefs of cooperation, sharing, support, and
encouragement in work tasks and sensitivity to the problems and
difficulties expressed by colleagues.
Collegial Teaching and Learning A continual growth process in which teachers prioritize the need to
continue to learn as an organizational member for the purpose of
enhancing learning through collaborative efforts in order to personally
and collectively benefit all students and staff; characterized by
collaborative work, shared planning, personal and group reflection,
dialogue among teachers, and incorporation of educational research.
Professional Commitment A continuous process that provides opportunities to enhance the
professional effectiveness of teachers through a commitment to the
continuous improvement of the learning process and a commitment
among teachers to serve as sources of help and support for colleagues
within the organization.
(Olivier, 2001).
Statistical Analyses
To examine the properties of the actual and
preferred RSCEQ scales, reliability estimates were
calculated for the total sample and subgroups using
SPSS 17.0, and confirmatory factor analyses
(CFAs) were conducted using AMOS 17.0. The
model used for the factor analyses was based on the
results of previous studies and specified a 3-factor
structure with independent factors (see Figure 1).
Based on reporting guidelines recommended by
Meyers, Gamst, and Guarino (2006), the following
fit measures were reported for each total group
CFA: (a) chi square, (b) normed fit index (NFI), (c)
comparative fit index (CFI), and (d) root mean
square error or approximation (RMSEA). The
following criteria were used to evaluate the
goodness of fit for the specified model: (a) chi
square, p ≥ .05, (b) NFI > .90, (c) CFI > .95, and
(d) RMSEA < .10 (Meyers et al., 2006).After the
scales were examined using the total sample,
subgroup analyses were conducted to test for three
types of invariance: (a) metric, (b) factor
covariance, and (c) error variance and covariance.
Metric invariance implies that the unstandardized
factor loadings for each indicator are equal across
groups. Factor covariance invariance focuses on
the inter-relationships between the factors across
groups. Error variance and covariance invariance
examines the equality of error variances across
groups (Blunch, 2008).
Results
Reliability Analyses
Actual scale. The results of the reliability
analyses for the actual scale are contained in Table
2. The reliability estimates for the total sample
ranged from .83 for Professional Commitment to
.90 for Shared Leadership and Vision and were
consistent with values reported by Olivier (2001)
and Lorraine (2012) for the 20-item RSCEQ. For
the subgroup analyses, the results showed that the
reliability estimates for subgroups based on gender
and grade level remained consistent with the values
for the total sample. The estimates ranged from
.83 to .89 for females and .83 to .90 for males.
Finally, the reliability estimates for elementary
respondents ranged from .82 to .90, and the
estimates for high school respondents ranged from
.83 to .88.
THE REVISED SCHOOL CULTURE ELEMENTS QUESTIONNAIRE
Fall 2012 34 RESEARCH IN THE SCHOOLS
Table 2
Reliability Estimates for Total Sample and Subgroups on RSCEQ Actual Scale
RSCEQ Dimension
Sample
n
Shared
Leadership/Vision
Collegial Teaching
and Learning
Professional
Commitment
Total 803 .90 .85 .83
Gender
Female 655 .89 .86 .83
Male 115 .90 .81 .83
School Level
Elementary 474 .90 .85 .82
Secondary 227 .88 .84 .83
Note. Subgroup sample sizes may not equal total sample size due to missing values on the subgroup indicator.
Figure 1. Confirmatory factor analysis model for the RSCEQ assuming uncorrelated factors. Leadership =
Shared leadership and vision. Professionalism = Professional commitment. Collegiality = Collegial teaching and
learning.
THOMAS DeVANEY, NAN ADAMS, FLO HILL-WINSTEAD, AND MITZI TRAHAN
Fall 2012 35 RESEARCH IN THE SCHOOLS
Preferred scale. The results of the reliability
analyses for the preferred scale are presented in
Table 3. For the total sample, the reliability
estimates ranged from a low of .87 for Professional
Commitment to a high of .91 for Collegial
Teaching and Learning. For the subgroup analyses,
the results showed that the reliability estimates for
subgroups based on gender and grade level
remained consistent with the values for the total
sample. Reliability estimates ranged from .88 to .91
for females and .83 to .89 for males. Finally, for
the reliability estimates based on grade level, the
values ranged from .87 to .90 for elementary
respondents and .88 to .92 for high school
respondents.
Table 3
Reliability Estimates for Total Sample and Subgroups on RSCEQ Preferred Scale
RSCEQ Dimension
Sample
n
Shared
Leadership/Vision
Collegial Teaching
and Learning
Professional
Commitment
Total 824 .90 .91 .87
Gender
Female 672 .91 .91 .88
Male 121 .86 .89 .83
School Level
Elementary 493 .90 .90 .87
Secondary 226 .91 .92 .88
Note. Subgroup sample sizes may not equal total sample size due to missing values on the subgroup indicator.
Confirmatory Factor Analyses
Actual scale. In order to examine the
underlying structure of the actual scale, a
confirmatory factor analysis was conducted using a
3-factor model with uncorrelated factors. The
results of the analysis are presented in Table 4.
An examination of the standardized regression
weights showed that all weights were greater than
.60 and supported the proposed item to factor
pattern for the 3-factor model. The chi-square
value of 1,586.8 (df = 170, p < .001) indicated that
the specified model was not an acceptable fit to the
data. The analysis also resulted in values of .795
for NFI, .812 for CFI, and .102 for RMSEA.
Similar to the chi-square result, these values
suggested that the specified model did not provide
a good fit for the data. These results suggested
that an alternative model may provide a better fit
for the data.
Although previous researchers treated the
factors as independent or uncorrelated, Cavanagh
(1997) acknowledged that, theoretically, the factors
should be correlated. Therefore, in order to
improve the original model, it was modified so that
the factors were correlated, and the analysis was
repeated (See Figure 2). The correlations among
the factors ranged from .60 to .86 and supported
Cavanagh’s belief that the factors are related. As
shown in Table 5, the analysis produced a chi-
square value of 722.58 (df = 167, p < .001).
Although the chi-square value was statistically
significant and did not support the model as a good
fit, by allowing the factors to be correlated, the chi-
square value was reduced by 54.5%.
Additionally, the values for two of the fit indices
suggested that the model provided a good fit. The
NFI (.914) exceeded the criterion of .90, and the
RMSEA (.06) was below the criterion of .10.
These results indicated that allowing the factors to
be correlated provided a better fitting model.
Consequently, this model was used for the
remaining CFAs.
THE REVISED SCHOOL CULTURE ELEMENTS QUESTIONNAIRE
Fall 2012 36 RESEARCH IN THE SCHOOLS
Table 4
Confirmatory Factor Analysis Results for Model with Uncorrelated Factors: Actual Scale
Item Leadership Collegial
Teaching
Professional
Commitment
Administrators provide visible, ongoing support for new
school programs and ideas. .681
Administrators are sympathetic with problems and
difficulties encountered by teachers in their work. .771
Administrators work to ensure the cooperation of
teachers. .812
Administrators visibly encourage teachers to be the best
that they can be in the classroom. .725
Teachers and administrators work cooperatively in
developing new school programs and policies. .728
Teachers receive the assistance they need from
administrators and colleagues to enhance the quality of
teaching and learning in their classroom. .795
Leadership roles are equally shared by teachers and
administrators. .694
Teachers are willing to help each other when problems
arise. .606
Teachers share classroom experiences with each other to
improve their understanding of students learning. .709
Teachers openly share problems with each other. .649
Teachers professionally share and learn from one
another. .786
Teachers encourage each other to use professional
judgment when making decisions. .658
Teachers feel comfortable in providing suggestions to
colleagues about ways in which to improve teaching and
learning in their classrooms. .680
Teachers spend time together to informally discuss ways
to improve the school. .620
Teachers give priority to helping their students develop
higher order thinking skills.
.677
Teachers incorporate the findings of educational research
into their own teaching and learning practices.
.622
Teachers believe that all students can learn. .645
Teachers are committed to professional growth to
improve teaching and learning.
.731
Teachers adequately plan teaching and learning activities
to accommodate individual differences among students.
.714
Teachers spend time in professional reflection about
their work.
.651
Model Fit Indices
χ2 NFI CFI RMSEA
1586.80* .795 .812 .102
*p < .001
THOMAS DeVANEY, NAN ADAMS, FLO HILL-WINSTEAD, AND MITZI TRAHAN
Fall 2012 37 RESEARCH IN THE SCHOOLS
Table 5
Confirmatory Factor Analysis Results for Model with Correlated Factors: Actual Scale
Item Leadership Collegial
Teaching
Professional
Commitment
Administrators provide visible, ongoing support for new
school programs and ideas. .683
Administrators are sympathetic with problems and
difficulties encountered by teachers in their work. .751
Administrators work to ensure the cooperation of
teachers. .802
Administrators visibly encourage teachers to be the best
that they can be in the classroom. .720
Teachers and administrators work cooperatively in
developing new school programs and policies. .726
Teachers receive the assistance they need from
administrators and colleagues to enhance the quality of
teaching and learning in their classroom. .793
Leadership roles are equally shared by teachers and
administrators. .695
Teachers are willing to help each other when problems
arise. .614
Teachers share classroom experiences with each other to
improve their understanding of students learning. .694
Teachers openly share problems with each other. .603
Teachers professionally share and learn from one another. .783
Teachers encourage each other to use professional
judgment when making decisions. .687
Teachers feel comfortable in providing suggestions to
colleagues about ways in which to improve teaching and
learning in their classrooms. .692
Teachers spend time together to informally discuss ways
to improve the school. .610
Teachers give priority to helping their students develop
higher order thinking skills.
.656
Teachers incorporate the findings of educational research
into their own teaching and learning practices.
.623
Teachers believe that all students can learn. .635
Teachers are committed to professional growth to
improve teaching and learning.
.758
Teachers adequately plan teaching and learning activities
to accommodate individual differences among students.
.702
Teachers spend time in professional reflection about their
work.
.655
Model Fit Indices
χ2 NFI CFI RMSEA
722.58* .914 .932 .061
*p < .001
THE REVISED SCHOOL CULTURE ELEMENTS QUESTIONNAIRE
Fall 2012 38 RESEARCH IN THE SCHOOLS
Subgroup analyses by gender. To test for
model invariance across gender, a CFA was
conducted with gender as the grouping variable.
The results of the analysis are presented in Table 6
and suggest that the model was factor invariant
with respect to the measurement weights (metric
invariant), structural covariances (factor covariance
invariant), and measurement residuals (error
variance invariant). These results indicate that not
only did the model fit both genders but also the
measurement weights were equal for males and
females. Further, the factor covariance and error
variance invariances suggest that the factors were
inter-related in the same way for male and female
respondents and the error variances were equal
across gender.
Subgroup analysis by grade level. To test for
model invariance across grade level, a CFA was
conducted with grade level (elementary or high
school) as the grouping variable. The results of
the analysis are presented in Table 7 and suggest
that the model was factor invariant with respect to
the measurement weights (metric invariant),
structural covariances (factor covariance invariant),
and measurement residuals (error variance
invariant). These results indicate that not only did
the model fit both grade levels but also the
measurement weights were equal across groups.
Further, the factor covariance and error variance
invariances suggest that the factors were inter-
related in the same way for both grade levels and
the error variances were equal across grade levels.
Table 6
Chi Square Tests for Invariance of Actual Scale Across Gender: Actual Scale
Model df χ2 χ
2 change df change p
Unconstrained 334 887.74
Equal measurement weightsa 351 901.33 13.59 17 .696
Equal structural covariancesb 357 906.79 5.46 6 .486
Equal error variancesc 377 937.98 31.19 20 .053
a results based on the assumption that the unconstrained model was correct.
b results based on the assumption that the model with equal measurement weights was correct.
c results based on the assumption that the model with equal measurement weights and structural covariances was
correct.
Table 7
Chi Square Tests for Invariance of Actual Scale Across Grade Level: Actual Scale
Model df χ2 χ
2 change df change p
Unconstrained 334 846.49
Equal measurement weightsa 351 858.25 11.76 17 .815
Equal structural covariancesb 357 863.64 5.39 6 .495
Equal error variancesc 377 873.02 9.37 20 .978
a results based on the assumption that the unconstrained model was correct.
b results based on the assumption that the model with equal measurement weights was correct.
c results based on the assumption that the model with equal measurement weights and structural covariances was
correct.
Preferred scale. Because prior researchers
did not examine the properties of the preferred
scale, the best fitting model (see Figure 2) from the
actual scale analysis was used. Results of the
analysis provided support for the identified model.
The values of the correlation coefficients among
the factors ranged from .894 to .969, and
standardized regression weights were greater than
.60 and supported the 3-factor model (see Table 8).
Although the chi-square value (859.54, df = 167)
indicated a statistically significant difference was
present between the model and observed data, the
fit measures suggested that the model provided a
good fit for the data. The NFI (.925) and RMSEA
(.071) exceeded the respective criterion for
considering the model a good fit for the data.
Additionally, the CFI value (.939) approached the
criterion (.95) for classifying the model as a good
fit.
THOMAS DeVANEY, NAN ADAMS, FLO HILL-WINSTEAD, AND MITZI TRAHAN
Fall 2012 39 RESEARCH IN THE SCHOOLS
Table 8
Confirmatory Factor Analysis Results for Model with Correlated Factors: Preferred Scale
Item Leadership Collegial
Teaching
Professional
Commitment
Administrators provide visible, ongoing support for new
school programs and ideas. .644
Administrators are sympathetic with problems and
difficulties encountered by teachers in their work. .742
Administrators work to ensure the cooperation of
teachers. .786
Administrators visibly encourage teachers to be the best
that they can be in the classroom. .811
Teachers and administrators work cooperatively in
developing new school programs and policies. .772
Teachers receive the assistance they need from
administrators and colleagues to enhance the quality of
teaching and learning in their classroom. .843
Leadership roles are equally shared by teachers and
administrators. .692
Teachers are willing to help each other when problems
arise. .706
Teachers share classroom experiences with each other to
improve their understanding of students learning. .794
Teachers openly share problems with each other. .720
Teachers professionally share and learn from one another. .834
Teachers encourage each other to use professional
judgment when making decisions. .804
Teachers feel comfortable in providing suggestions to
colleagues about ways in which to improve teaching and
learning in their classrooms. .773
Teachers spend time together to informally discuss ways
to improve the school. .754
Teachers give priority to helping their students develop
higher order thinking skills.
.688
Teachers incorporate the findings of educational research
into their own teaching and learning practices.
.707
Teachers believe that all students can learn. .687
Teachers are committed to professional growth to
improve teaching and learning.
.769
Teachers adequately plan teaching and learning activities
to accommodate individual differences among students.
.797
Teachers spend time in professional reflection about their
work.
.754
Model Fit Indices
χ2 NFI CFI RMSEA
859.54* .925 .939 .071
*p < .001
THE REVISED SCHOOL CULTURE ELEMENTS QUESTIONNAIRE
Fall 2012 40 RESEARCH IN THE SCHOOLS
Figure 2. Revised confirmatory factor analysis model for the RSCEQ assuming correlated factors. Leadership =
Shared leadership and vision. Professionalism = Professional commitment. Collegiality = Collegial teaching and
learning.
Subgroup analysis by gender. To test for
model invariance across gender, a confirmatory
factor analysis was conducted with gender as the
grouping variable. The results of the analysis are
presented in Table 9 and suggested that the model
was invariant with respect to the measurement
weights (metric invariant) and structural
covariances (factor covariance invariant). These
results indicated that not only did the model fit both
subgroups but also the measurement weights were
equal across gender and the factors were inter-
related in the same way for male and female
respondents. However, the results also indicated
that statistically significant differences were
present between the error variances associated with
the scale items for male and female respondents.
Subgroup analysis by grade level. To test for
model invariance across grade level, a confirmatory
factor analysis was conducted with grade level
(elementary or high school) as the grouping
variable. The results of the analysis are presented in
Table 10 and suggested that the model was
invariant with respect to the measurement weights
(metric invariant). However, the results also
indicated that the model was not invariant with
respect to the structural covariances or error
variances. These results suggested that not only did
the model provide a good fit at both grade levels
but also the measurement weights were equal
across grade level. However, these results also
suggested that a statistically significant difference
was present in the way the factors were related and
the error variances associated with the scale items
for the two grade levels.
THOMAS DeVANEY, NAN ADAMS, FLO HILL-WINSTEAD, AND MITZI TRAHAN
Fall 2012 41 RESEARCH IN THE SCHOOLS
Table 9
Chi Square Tests for Invariance of Preferred Scale Across Gender : Preferred Scale
Model Df χ2 χ
2 change df change p
Unconstrained 334 1118.10
Equal measurement weightsa 351 1139.27 21.16 17 .219
Equal structural covariancesb 357 1148.24 8.97 6 .175
Equal error variancesc 377 1454.50 306.26 20 <.001
a results based on the assumption that the unconstrained model was correct.
b results based on the assumption that the model with equal measurement weights was correct.
c results based on the assumption that the model with equal measurement weights and structural covariances was
correct.
Table 10
Chi Square Tests for Invariance of Preferred Scale Across Grade Level: Preferred Scale
Model df χ2 χ
2 change df change p
Unconstrained 334 1012.55
Equal measurement weightsa 351 1034.72 22.17 17 .178
Equal structural covariancesb 357 1053.17 18.45 6 .005
Equal error variancesc 377 1281.14 227.96 20 <.001
a results based on the assumption that the unconstrained model was correct.
b results based on the assumption that the model with equal measurement weights was correct.
c results based on the assumption that the model with equal measurement weights and structural covariances was
correct.
Discussion
In 2006, Maslowski noted an increase in the
prevalence of research related to the topic of school
culture. One possible stimulus for this increase
was the development and release of the 1996
Interstate School Leaders Licensure Consortium
(ISLLC) standards in which the need for
educational leaders to understand the role of school
culture was explicitly acknowledged. Of
particular interest for the current study was the
expectation that school culture and climate be
assessed on a regular basis. The increased
attention that has been given to school culture, as
evidenced by the increase in research activity and
its continued inclusion in professional standards for
educational leaders (e.g., revised ISLLC standards
and original and revised ELCC standards), has
resulted in the need for instruments to assess school
culture. One such instrument is the RSCEQ
(Olivier, 2001).
The RSCEQ was derived from the original
SCEQ developed by Cavanagh (1997) and provides
a unique feature among school culture instruments
by collecting data that reflect the perceptions of
faculty concerning not only the current school
culture but also the preferred culture. Researchers
who have reported the results of developmental
work leading to the current RSCEQ (Cavanagh,
1997; Dellar et al., 1998; Olivier, 2001) have only
focused on the factor structure and properties of the
actual scale. Further, the original examination by
Cavanagh (1997) and later examination of the
current 20-item version by Olivier (2001)
addressed the properties only for the total sample.
Therefore, the purpose of the current study was to
examine the psychometric properties of the actual
and preferred scales of the RSCEQ for the total
THE REVISED SCHOOL CULTURE ELEMENTS QUESTIONNAIRE
Fall 2012 42 RESEARCH IN THE SCHOOLS
sample and subgroups based on gender and grade
level (i.e., elementary or high school).
The results of this study are based on a sample
of 901 elementary through high school teachers
across 44 schools in southeast Louisiana. The
psychometric properties of the RSCEQ were
examined through the calculation of score
reliabilities and use of confirmatory factor
analyses. The reliability analyses produced total
sample estimates that ranged from .83 to .90 on the
actual scale and .87 to .91 for the preferred scale.
Estimates for subgroups based on gender and grade
level (i.e., elementary or high school) were
consistent with the total sample estimates.
The original model for the confirmatory factor
analyses was based on findings by previous
researchers who examined the actual scale and
assumed independent factors. However, the
measures of fit suggested that the model shown in
Figure 1 did not provide a good fit for the data.
Consequently, the model was modified to align
with the theoretical assumption made by Cavanagh
(1997) that the factors were related. The results
for this model showed a substantial decrease in the
chi square value, and multiple measures of fit
satisfied the respective criterion for identifying a
good fitting model. These results provide
evidence to support not only the three-factor
structure of the data but also the theoretical
assumption that the factors are correlated.
The multi-group analyses provided evidence
that the RSCEQ scales were invariant across
gender and grade level. For the actual scale, the
scale was invariant with respect to measurement
weights, factor covariances, and error variances
across gender and grade level. Additionally, the
results of the preferred scale analyses indicated that
the scale was invariant with respect to the
measurement weights and factor covariance across
gender. However, across grade level, the
preferred scale was only invariant with respect to
the measurement weights. Therefore, although
the results indicate that statistically significant
differences were not present between
corresponding measurement weights, statistically
significant differences were related to the
relationships among the three factors at the
elementary and high school levels.
Because the current findings support the three-
factor structure of the RSCEQ for males and
females and elementary and high school teachers,
this instrument provides a tool that educational
leaders can use to assess and describe the climate
and culture among these subgroups. Further,
because the actual scale was invariant across
gender and grade level, the RSCEQ can be used by
educational leaders and researchers to compare
responses across these subgroups. However, the
preferred scale was not fully invariant across
genders or grade level. Although Byrne (2004)
noted that testing for full invariance (e.g., testing
for invariance of error parameters) is an overly
restrictive analysis, comparing results obtained
from the preferred scale across gender and grade
level should be made with caution.
Recommendations for Further Research
The results of this study provide additional
evidence that supports the theoretically based
structure of the RSCEQ that includes three
correlated factors. However, continued research
that examines its properties should be conducted.
Because the samples for the current study and
previous validation study of the 20-item RSCEQ
(Olivier, 2001) were based on teachers in
Louisiana, future research should utilize teachers
from other geographic locations in order to
strengthen the population validity of the
instrument. Also, as noted in the description of
the sample, variation was present in the number of
responses per school. Because of differences
among schools regarding the size of faculties and
the lack of available data regarding total faculty
size, response rates per school could not be
determined. However, it can be assumed that the
voluntary nature of participation resulted in
variations in per school response rates. Using the
teacher as the unit of analysis, instead of
aggregating data to the school level, should have
minimized the impact of per school response rate;
however, future research should attempt to utilize
techniques that will maximize the response rates
within schools.
The lead editors for this article were John R. Slate
and Anthony J. Onwuegbuzie.
References
Blunch, N. J. (2008). Introduction to structural
equation modeling using SPSS and
AMOS. Thousand Oaks, CA: Sage.
Byrne, B. M. (2004). Testing for multigroup
invariance using AMOS Graphics: A road
less traveled. Structural Equation
Modeling, 11, 272-300.
doi:10.1207/s15328007sem1102_8
Cavanagh, R. F. (1997). The culture and
improvement of Western Australia’s
senior secondary schools (Unpublished
doctoral dissertation). Curtain University,
Western Australia.
Council of Chief State School Officers. (2008).
Educational leadership policy standards:
ISLLC 2008. Washington, DC: Author.
Retrieved from http://www.npbea.org
THOMAS DeVANEY, NAN ADAMS, FLO HILL-WINSTEAD, AND MITZI TRAHAN
Fall 2012 43 RESEARCH IN THE SCHOOLS
Deal, T. E., & Peterson, K. D. (1990). The
principal's role in shaping school culture.
Retrieved from ERIC database.
(ED325914)
Dellar, G., Cavanagh, R., & Ellett, C. D. (1998,
April). Changing the context of learning
environment research. Paper presented at
the annual meeting of the American
Educational Research Association, San
Diego, CA.
DuFour, R. (1998). Schools and learning
communities: Learning-centered schools
grow from strong cultures. Journal of Staff
Development, 19(1), 58-59.
Ellett, C. C., Rugutt, J. K., & Cavanagh, R. (2000,
April). Conceptualizing and measuring
school environments from cultural
perspective: Phase three of the
development of a teacher perceptions
measure. Paper presented at the annual
meeting of the American Educational
Research Association, New Orleans, LA.
Lorraine, R. R. D. O. (2012). A study of the
relationship between teachers’ perception
of principal’s leadership practices and
school culture to student achievement
(Doctoral dissertation). Available from
ProQuest Dissertations and Theses
database. (UMI No. 3482494)
Maslowski, R. (2006). A review of inventories for
diagnosing school culture. Journal of
Educational Administration, 44(1), 6-35.
doi:10.1108/09578230610642638
Meyers, L. S., Gamst, G., & Guarino, A. J. (2006).
Applied multivariate research design and
interpretation. Thousand Oaks, CA: Sage.
National Policy Board for Educational
Administration. (2002). Standards for
advanced programs in educational
leadership for principals, superintendents,
curriculum directors, and supervisors.
Austin, TX: Author. Retrieved from
http://www.npbea.org
National Policy Board for Educational
Administration. (2009). Educational
leadership program standards: ELCC
revised standards – First draft for
advanced programs that prepare
principals, superintendents, curriculum
directors, and supervisors. Austin, TX:
Author. Retrieved from
http://www.npbea.org
Olivier, D. F. (2001). Teacher personal and school
culture characteristics in effective
schools: Toward a model of a professional
learning community. (Doctoral
dissertation). Available from ProQuest
Dissertations and Theses database. (UMI
No. 3016566)
Schein, E. H. (1985). Organizational culture and
leadership. San Francisco, CA: Jossey-
Bass.
Schoen, L. (2005). Conceptualizing, describing,
and comparing school cultures: A
comparative case study of school
improvement processes. (Doctoral
dissertation). Available from ProQuest
Dissertations and Theses database. (UMI
No. 3167154)
Senge, P., Cambron-McCabe, N., Lucas, T., Smith,
B., Dutton, J., & Kleiner, A. (2000).
Schools that learn: A fifth discipline
fieldbook for educators, parents, and
everyone who cares about education. New
York, NY: Doubleday.
Stolp, S., & Smith S. C. (1995). Transforming
school culture: Stories, symbols, values, &
the leaders role. University of Oregon:
ERIC Clearinghouse on Educational
Management.
THE REVISED SCHOOL CULTURE ELEMENTS QUESTIONNAIRE
Fall 2012 44 RESEARCH IN THE SCHOOLS
Appendix A
Revised School Culture Elements Questionnaire
This questionnaire contains a number of statements about things which occur in some schools. After reading
each of the statements carefully, you are asked to judge each response according to two criteria: (1) “how you
and your school actually are…” and (2) “you would prefer that you or your school would be…” You are to
indicate the extent to which you agree or disagree with each of the statements. The “actual scale” applies to
how “YOU AND/OR YOUR SCHOOL ACTUALLY ARE” and the “prefer scale” describes what you
would “PREFER TO BE OR WOULD PREFER YOUR SCHOOL TO BE LIKE.”
SCALE: 1=Strongly Disagree (SD) 2=Disagree (D) 3=Agree (A) 4=Strongly Agree (SA)
“ACTUAL” SCALE
SD D A SA
STATEMENTS
“PREFER” SCALE
SD D A SA
Administrators provide visible, ongoing support for new
school programs and ideas.
Teachers are willing to help each other when problems
arise.
Teachers give priority to helping their students develop
higher order thinking skills.
Administrators are sympathetic with problems and
difficulties encountered by teachers in their work.
Teachers share classroom experiences with each other to
improve their understanding of students learning.
Teachers incorporate the findings of educational research
into their own teaching and learning practices.
Administrators work to ensure the cooperation of teachers.
Teachers openly share problems with each other.
Teachers believe that all students can learn.
Administrators visibly encourage teachers to be the best
that they can be in the classroom.
Teachers professionally share and learn from one another.
Teachers are committed to professional growth to improve
teaching and learning.
Teachers and administrators work cooperatively in
developing new school programs and policies.
Teachers encourage each other to use professional
judgment when making decisions.
Teachers adequately plan teaching and learning activities
to accommodate individual differences among students.
Teachers receive the assistance they need from
administrators and colleagues to enhance the quality of
teaching and learning in their classroom.
Teachers feel comfortable in providing suggestions to
colleagues about ways in which to improve teaching and
learning in their classrooms.
Teachers spend time in professional reflection about their
work.
Leadership roles are equally shared by teachers and
administrators.
Teachers spend time together to informally discuss ways
to improve the school.
Copyright 2012 by the RESEARCH IN THE SCHOOLS Mid-South Educational Research Association 2012, Vol.19, No. 2, 45-61
Fall 2012 45 RESEARCH IN THE SCHOOLS
Examining Variations in Programmatic Delivery on Teaching Candidates’ Sense of Efficacy
S. Michael Putman The University of North Carolina at Charlotte
The current climate in the United States,
markedly fixed on teacher effectiveness and student learning, has resulted in considerable pressure at multiple levels in the educational hierarchy, from elementary schools through institutions of higher education (Ball & Forzani, 2009; Darling-Hammond, 2010). The demands have been especially impactful on colleges of education as reformers request evidence of programmatic rigor while simultaneously calling for the preparation of teachers who are content-area experts able to work with diverse populations (Darling-Hammond, 2010). Numerous measures have been explored by coordinators of teacher education programs to address these challenges (see Darling-Hammond, 2010); yet, a common assertion is that attention must be directed towards designing authentic teaching-learning environments that will empower preservice teaching candidates to translate theory presented within university courses into their practices more effectively
(Moore, 2003). Field experiences have been cited as representing a mechanism to facilitate this process because they allow candidates with limited training to apply pedagogical skills and strategies in controlled settings under the guidance of a mentor (Berliner, 1985; Zeichner, 2010). However, recent research examining the impact of differing field-based experiences has been characterized as sparse at best (Shanahan, 2008). Although generally recognizing the paradigm that more is better, there are multiple considerations that must be examined to maximize the benefits associated with field experiences, including duration, context, opportunities for practice, and guidance (Gurvitch & Metzler, 2009; Hammerness, Darling-Hammond, & Bransford, 2005; Putman & Borko, 2000).
Acknowledging that field experiences represent an important vehicle for developing practical knowledge and impacting teaching behaviors, another very influential variable on practices is teaching candidates’ beliefs about their abilities (Woolfolk Hoy & Spero, 2005). Research has implicated teacher preparation as an important mechanism for understanding the impact of beliefs on practices and behaviors (Hammerness et al., 2005; Pajares, 1992). However, as a result of the complex, multifaceted nature of the construct and
Correspondence concerning this article should be addressed to S. Michael Putman, Charlotte, 9201 University City Boulevard, Charlotte, NC 28223 E-Mail: [email protected]
Researchers have recommended that teacher education candidates receive multiple opportunities for systematic implementation of practices within authentic contexts. These experiences form an important foundation for candidates’ beliefs about their abilities in classrooms. Yet, few researchers have examined the relationship between the context for delivery of the experiences and the development of teaching efficacy. This study involved investigation of the impact of variations in 2 teacher education courses along several dimensions of teaching efficacy, consisting of 43 teaching candidates enrolled in an undergraduate elementary education program, who were selected via convenience sampling. These candidates comprised two groups: (a) those enrolled in the blocked semester (n = 18) that involved both a pedagogical foundations course (i.e., included objectives focused on developing and extending knowledge of instructional planning and investigating theories and principles of effective classroom management) and a course that extended the content of the pedagogical foundations course within a focused practicum experience; and (b) those enrolled in a traditional course delivery format (n = 25), wherein teaching candidates enrolled in each course separately and completed the requirements in separate semesters. Participants in both groups were administered an instrument that assessed efficacy towards the beginning of the course and during the last class of the practicum course. Findings indicated that participation in an intensive, field-based placement (i.e., blocked group) yielded growth in several related areas of teaching efficacy. Implications are discussed as related considerations for teacher education programs as they seek to structure field experiences to maximize growth in candidates’
S. MICHAEL PUTMAN
Fall 2012 46 RESEARCH IN THE SCHOOLS
variations in contexts where teaching experiences occur, previous investigations examining the relationship between efficacy and experiences have not resulted in definitive conclusions (Knoblauch & Woolfolk Hoy, 2008; Woolfolk Hoy & Spero, 2005). Researchers have called for additional studies that investigate features of pathways to teaching that are powerful for preparing teachers and demonstrate a related effect on efficacy for teaching (e.g. Grossman & McDonald, 2008; Zeichner & Conklin, 2005). This research was undertaken to answer such calls. The following research question was used to frame the investigation: What is the impact of variations in programmatic delivery on teaching candidates’ general and domain-specific sense of teaching efficacy?
Review of Literature
Teacher Preparation Preparing teaching candidates to enter the complex and challenging environment of today’s schools is a daunting task. Programs, whether traditional or alternative, must attend to multiple elements, including where, when, and how teacher preparation is structured, to ensure candidates are prepared for the rigors of the classroom (Zeichner, 2010). Regardless of the challenges inherent with the preparation of future teachers, however, the current accountability movement in the United States has resulted in a focus on the influence and impact of teacher quality on student achievement and educational outcomes (Hanushek & Rivkin, 2012; Henry, Kershaw, Zulli, & Smith, 2012). As a result, university-based programs are under increased scrutiny and criticism has mounted regarding an apparent lack of evidence of their effectiveness (Hanushek & Rivkin, 2012).
Although such debates regarding teacher preparation are highly politicized and charged, teacher education programs themselves might have contributed to the current state of affairs. Traditionally, teacher preparation programs have been characterized as focusing on preparation of candidates within a theoretical paradigm as opposed to facilitating development of skills through opportunities to practice them with authentic classroom contexts (Ball, Sleep, Boerst, & Bass, 2009). The emphasis behind this approach is to develop a sound understanding of pedagogy in combination with content knowledge prior to the application of these elements within a school setting (Korthagen & Kessels, 1999).
Programs operating within this paradigm have provided teaching candidates multiple opportunities to engage in “simulated planning” (Grossman & McDonald, 2008, p. 190) and reflection, but fewer opportunities to practice teaching in authentic settings under normal
teaching conditions (Grossman & McDonald, 2008). This likely contributes to a lack of connection between information presented in university coursework and practices observed within field experiences. Numerous researchers have cited the theory-practice gap discussed by candidates in teacher education programs (Allsopp, DeMarie, Alvarez-McHatton, & Doone, 2006; Korthagen, Loughran, & Russell, 2006). As Grossman, Hammerness, and McDonald (2009) point out,
Principles developed in the absence of assisted practice lack the depth required … to enact such principles in practice. At the same time, learning to enact instructional routines in the absence of a developing sense of the principles underlying such routines reinforces a view of teaching as a set of techniques. (p. 278)
When theory-practice connections are lacking or when there is dissonance between theory presented in coursework and actual practice observed in the classroom, teaching candidates are more likely to adopt the practices they observe in the classroom, regardless of whether it conflicts with information presented in teacher education courses (see Moore, 2003; Putman, 2009).
Perhaps exacerbating the problems associated with the lack of coherence between teacher education and the experiences of teaching candidates is the fact that there is often too little communication between classroom supervisors and university faculty. It is not uncommon for cooperating teachers to have limited information regarding the content presented within the campus-based courses and university instructors often know very little about the practices taking place in the P-12 classroom where candidates are placed (Darling-Hammond, 2010). As a result, opportunities to implement practices described within coursework vary greatly, even within the context of partnerships established with professional development schools (Moore, 2003; Zeichner, 2007).
To create effective teachers, research suggested that the learning process of teaching candidates should include multiple opportunities that engage teaching candidates as learners and build from their prior knowledge and understandings within the environments that they will teach (Borko & Putman, 1996; Hammerness et al., 2005). As described by Schön (1983), the emphasis should be placed on “theory-in-action” (p. 50). Candidates should engage in "work that is done as part of action or practice, like that done in the actual riding of a bicycle or the actual making of a medical diagnosis” (Cook & Brown, 1999, p. 387). This can be accomplished through carefully structuring and scaffolding candidate learning in sustained and intense field experiences that occur within programs.
EXAMINING VARIATIONS IN PROGRAMMATIC DELIVERY ON TEACHING CANDIDATES’ SENSE OF EFFICACY
Fall 2012 47 RESEARCH IN THE SCHOOLS
Field Experiences Field experiences are “school-based
opportunities in which candidates may observe, assist, tutor, instruct, participate in service learning projects, or conduct applied research” (National Council for Accreditation of Teacher Education, 2008, p. 32). In short, they represent contexts where teaching candidates have the opportunity to explore the relationship between theory and action. The importance of the field experience within the development of teacher proficiency has been emphasized for more than a century. John Dewey (1904/1964) advanced this notion as he advocated for the use of learning to teach through extensive practice in real classrooms. Dewey believed that this facilitated opportunities for experimentation with methods and practices, similar to the apprenticeship models that proliferated in many industries during the same time period.
Contemporary educational theorists still support the importance of practical experiences in the classroom to advance teaching candidates’ knowledge, skills, and dispositions as they offer what Grossman and McDonald (2008) referred to as “approximations of practice” (p. 190). These approximations are the result of being able to develop teaching strategies in settings of reduced complexity where candidates do not have to attend to all elements within a teaching situation. Field experiences have the greatest impact on the development of the skills and practices of candidates when they are explicitly linked to content presented in methods courses (Darling-Hammond, 2006). This has led some universities to combine coursework with extended internships or residencies that occur in structured school-based settings. Such activities allow teaching candidates to benefit from the guidance of university faculty and staff as well as school personnel (see Castle, Fox, & O’Hanlan Souder, 2006; Fraser, 2007). It also allows teacher educators to coordinate information provided within their methods courses to the specific practices and expertise of teachers in schools (Zeichner, 2010). The coordination of practices as well as the joint support and guidance for teaching candidates is crucial for an understanding of their development and impact.
However, teacher education has not reached a juncture where there is consistent synchronization between university coursework and practices demonstrated in the field. Reinforcing the traditional paradigm, theory may be introduced in coursework, but field experiences do not provide specific opportunities to enact the principles advanced within the theory presented as part of the coursework (Korthagen & Kessels, 1999). Instead, Montecinos et al. (2011) found that during the early phases of field experiences, candidates’ primary tasks were focused on “observing, gathering, and analyzing information” (p. 285) as opposed to
engaging in teaching practices that would advance their pedagogical skills. The current challenge is that research on methods, structures, and conditions for field experiences has not produced definitive conclusions for maximizing the links between theory and practice (see Cochran-Smith & Zeichner, 2005; Wilson, Floden, & Ferrini-Mundy, 2002). Allsopp et al. (2006) provided several suggestions to foster these connections, including increasing the involvement of the supervising teacher, creating course assignments that demonstrate greater applicability to the field experience, and modifying the way courses are scheduled. Additional approaches for connecting the theory taught within teacher education programs and field-based experiences include:
• Developmentally sequenced field experiences aligned with theory-based and practical teaching assignments (Capraro, Capraro, & Helfeldt, 2010);
• Clearly articulated purposes that are shared amongst all stakeholders, including classroom supervisors (Dodds, 1989); and
• Altering the quantity and duration of the field experiences to match the needs of teaching candidates (Burant & Kirby, 2002).
When teacher preparation programs focus on these critical variables, they can link them with experiences focused upon reflection to strengthen connections between personal experiences and classroom pedagogy. The result of such activities will be teaching candidates who are better prepared to engage in the complex behaviors necessary of effective teachers (see Darling-Hammond, Hammerness, Grossman, Rust, & Shulman, 2005; Zeichner & Conklin, 2005).
Teachers’ Sense of Efficacy
Although field experiences represent a significant factor in developing skills and pedagogy, teachers’ beliefs about their abilities represent another important influence on their classroom behaviors as they mediate the relationship between knowledge and action. Teaching efficacy—to which this construct often is referred—was derived from Bandura’s (1986) social cognitive theory (SCT). Definitions of teachers’ sense of efficacy vary; yet, most share the common characteristic of being focused upon a teacher’s belief in her or his ability to affect student learning and behavior. In their seminal work, Tschannen-Moran and Woolfolk Hoy (2001) stated that the construct represented a teacher’s “judgment of his or her capabilities to bring about desired outcomes of student engagement and learning” (p. 783).
Teachers who demonstrate high levels of efficacy for teaching are more likely to implement effective methods of instruction (Ashton & Webb,
S. MICHAEL PUTMAN
Fall 2012 48 RESEARCH IN THE SCHOOLS
1986; Bandura, 1997), to persist during difficult teaching situations (Dembo & Gibson, 1985; Woolfolk, Rosoff, & Hoy, 1990), and to be successful at maintaining student engagement (Ashton & Webb, 1986). Efficacious teachers also are more likely to seek ways to improve their teaching methods through alternative methods of instruction and experiment with innovative instructional materials (Allinder, 1994; Ashton & Webb, 1986). All of these practices are associated with improvements in student achievement. On the other hand, a low sense of efficacy has been associated with non-differentiation of instruction, lack of interest in collaboration with other teachers, and negative views toward inclusion (Soodak, Podell, & Lehman, 1998).
Teaching efficacy is complex and multifaceted, especially in regard to preservice teaching candidates. Posnanski (2007) noted that teaching candidates’ efficacy is enhanced through participation in university courses that employed a constructivist paradigm and included both modeling and opportunities to apply knowledge in contextually relevant settings. Similarly, Clift and Brady (2005) observed a positive impact on efficacy when teaching candidates were provided a combination of observation, simulation, and opportunities to work with small groups of students. However, questions remain regarding the best method to combine these elements within a teacher education program to maximize the impact on teaching efficacy (Erawan, 2011). The lack of certainty in how effectively to merge the three elements likely is exacerbated by the fact that teaching candidates generally perceive themselves as being effective at implementing a variety of instructional practices and management strategies prior to field experiences (Rimm-Kaufman & Sawyer, 2004). In fact, given the malleability of the beliefs of preservice teachers, precise measurement of teaching efficacy might be difficult due to the subtle changes that occur as candidates complete other programmatic requirements, including field experiences and coursework prior to student teaching (see Chambers, 2003; Oh, Ankers, Llamas, & Tomjoy, 2005; Wagler & Moseley, 2005). Field Experiences and Teaching Efficacy
There is little doubt that teaching candidates need field experiences that offer safe opportunities to teach and to receive feedback while simultaneously being able to observe and to implement effective teaching strategies (Morrell & Carroll, 2003). From a social cognitive perspective (Bandura, 1997), mastery experiences, such as those that occur within practica and student teaching, provide the greatest impact on efficacy. Research has established that when teaching candidates faced and succeeded in challenging
situations within field experiences, stronger efficacy beliefs were developed (Gurvitch & Metzler, 2009). Accordingly, the earlier such experiences can be introduced in the teacher education program, the more durable teaching candidates’ sense of efficacy becomes as the beliefs are based on actual experiences (Woolfolk Hoy & Spero, 2005). When teaching candidates are not provided sufficient experiences linked to practices present in a real classroom, they are more likely to experience the previously described positive, yet unrealistic, sense of efficacy (Rimm-Kaufman & Sawyer, 2004). For example, receiving high scores on classroom management plans that are not implemented within a classroom might inflate views regarding successful implementation of strategies and teaching candidates might be easily dismayed in the context of a true teaching experience (see Putman, 2009).
The majority of research on the impact of field experiences on efficacy has occurred within student teaching experiences. Results have not been found to be definitive in regard to the specific impact of field experiences and might be dependent on the form of efficacy that is examined. For example, Hoy and Woolfolk (1990) indicated that general teaching efficacy improved as a result of participation in field experiences that occurred in conjunction with coursework, but declined during student teaching. However, in studies focused on personal teaching efficacy, which is more domain specific, researchers have noted improvements in efficacy within student teaching (Knoblauch & Woolfolk Hoy, 2008; Woolfolk Hoy & Spero, 2005). In fact, Knoblauch and Woolfolk Hoy (2008) found that teaching candidates’ efficacy increased regardless of the setting (e.g., rural, suburban, and urban) in which student teaching occurred. Still other researchers have found no differences or declines in efficacy in preservice candidates with regard to the completion of student teaching (Oh et al., 2005; Wagler & Moseley, 2005). Variations might be attributable to differences within preparation as candidates engage in a range of field experiences prior to student teaching (Hammerness et al., 2005). Some early field experiences offer a variety of support and feedback mechanisms, whereas student teaching immerses candidates in a complex atmosphere, which might not always match their prior expectations of the day-to-day realities of the classroom (Charalambous, Philippou, & Kyriakides, 2008; Zeichner & Conklin, 2005).
The aforementioned support might come in the form of assistance and guidance within structured feedback from classroom supervisors and university personnel. Tschannen-Moran and Woolfolk Hoy (2007) noted that when feedback and guidance were present, they contributed significantly to efficacy development. Zeichner
EXAMINING VARIATIONS IN PROGRAMMATIC DELIVERY ON TEACHING CANDIDATES’ SENSE OF EFFICACY
Fall 2012 49 RESEARCH IN THE SCHOOLS
(1992) found that undergraduate teaching candidates became more confident and successful when provided with adequate feedback within field experiences. Improvements in candidates’ sense of efficacy also can occur within the context of self-reflective activities. Graham (2006) determined that when candidates were given multiple opportunities to reflect upon and to engage in dialogue about their practices and the overall experiences of teaching and observing within field experiences, it had a significant influence on efficacy. In short, candidates need to believe that they are guided and supported by mentors, such as the classroom supervisor and university faculty member. On the other hand, when this support is withdrawn, efficacy levels are likely to drop (Tschannen-Moran, Woolfolk Hoy, & Hoy, 1998).
Recent research findings have expanded the knowledge-base regarding field experiences as well as teaching efficacy; yet, examinations of preservice candidates’ efficacy for teaching have yielded inconsistencies and raised multiple questions about the development of beliefs as candidates participate in various field experiences (see Chambers & Hardy, 2005; Knoblauch & Woolfolk Hoy, 2008; Oh et al. 2005; Woolfolk Hoy & Spero, 2005). As a result, it has been proposed that additional investigations of methods to strengthen and to develop efficacy in teaching candidates are necessary as programs seek to support candidates in their development (Henson, 2003). This research was undertaken to answer such calls; yet, unlike many of the other investigations wherein researchers have attempted to develop conclusions relative to field experiences, in this investigation, it was sought to examine the impact of variations in programmatic delivery, including coursework and field experiences, that occurred prior to student teaching on teaching efficacy. It was hoped that such an examination would provide researchers and developers of teacher education programs insights on how efficacy changes over time within a supportive environment that maximized connections between theory and practice, as well as on potential methods to alleviate the malleability of efficacy beliefs shown within previous investigations (see Pajares, 1992; Woolfolk Hoy & Spero, 2005).
Method
Context
The context for this research was a 4-year, undergraduate elementary education program in a Midwestern public university with a total enrollment of approximately 20,000 students. The research was conducted within a required two-course sequence typically completed in Year 3 of the program, prior to eligibility for student teaching. The first class in this sequence, which
will be referred to as the pedagogical foundations course, includes objectives focused on developing and extending knowledge of instructional planning and investigating theories and principles of effective classroom management, as applied to instructional contexts. As part of the course requirements, candidates develop two lesson plans central to a topic or theme that require the inclusion of effective instructional strategies identified by Marzano, Pickering, and Pollock (2004). Candidates also create a classroom management plan that introduces their ideas regarding the development of rules, procedures, and classroom organization, citing relevant theories of experts within the field. The pedagogical foundations course does not include a field experience; thus, preservice candidates are not provided with the opportunity to teach their lessons, nor do they implement any aspects of the classroom management plan as part of typical delivery format for the course. Instead, candidates are required to complete 10 hours of observation within an elementary classroom of their choosing. Upon completion of the observations, candidates are required to write a reflective paper that summarizes the experiences, citing relevant course content and documenting their growth as future educators.
The second course in the sequence extends the content of the pedagogical foundations course within a focused practicum experience. The objectives for candidates enrolled in the practicum experience include: demonstrating proficiency in planning and instructional practice, investigating the various perspectives on student achievement and the factors that affect it, and applying theories and strategies associated with classroom management strategies to teaching situations. Candidates experience 4 to 6 hours of observation, conduct 1 to 6 hours of whole class instruction, and conduct 24 to 28 hours of small group instruction within the practicum under the guidance of the university instructor and cooperating teacher at a field site. Lessons for the classroom instruction component are drawn from a 10-lesson unit plan developed by the candidates that focuses upon a specific topic as dictated by the curricular topics of the cooperating teaching. Feedback is provided to the candidate on lesson plans as well as teaching experiences by the university instructor in collaboration with the cooperating teacher.
This investigation focused on two variations of the delivery of the aforementioned classes. The first variation represented an intense blocked semester wherein candidates enrolled in both of the previously described courses simultaneously. All content and programmatic requirements for the pedagogical foundations course as well as the focused practicum experience were included and assigned within what will henceforth be referred to as the blocked semester or blocked experience. All
S. MICHAEL PUTMAN
Fall 2012 50 RESEARCH IN THE SCHOOLS
activities for the blocked semester, including seminars, occurred in a local elementary school. In the second variation, which will be referred to as the traditional format or traditional experience, candidates enrolled in each course separately and completed the requirements described previously in separate semesters.
Participants
Non-probability, convenience sampling (Cohen, Manion, & Morrison, 2000; Onwuegbuzie & Collins, 2007) was used to select 43 participants enrolled in an undergraduate elementary education program. All participants had been accepted into the teaching curriculum associated with the major and had achieved junior-level status and, thus, were enrolled in the pedagogical foundations course, which was previously described. The participants comprised two groups: those enrolled in the blocked semester and those participating in the traditional course delivery format. Self-selection sampling was used to identify participants within the blocked semester because each teaching candidate had chosen to enroll in this method of delivery for the two courses. Eighteen candidates were enrolled in the blocked semester, 16 females and two males. All participants were White and the sample had a mean age of 20.17 years (SD = 0.38). Random purposeful sampling (see Onwuegbuzie & Collins, 2007) was used to select the participants from the traditional course delivery format. A list of all candidates (n = 81) enrolled in the traditional sections of the pedagogical foundations course who granted consent to participate in the study was compiled. Every third name from the list was selected until 25 names were generated. The selection process resulted in 22 females and three males. The participants’ self-identified race-ethnicity was White (92%) and African-American (8%). The mean age of the participants in the traditional format was 21.28 years (SD = 3.47).
Instrument
Tschannen-Moran and Woolfolk Hoy’s (2001) Teachers’ Sense of Efficacy Scale (TSES) was the instrument utilized for the assessment of efficacy of all participants. The TSES was selected due to its recognized acceptance within the field and its demonstrated use with preservice teachers (see O’Neill & Stephenson, 2012; Putman, 2012; Tschannen-Moran & Woolfolk Hoy, 2001). There are two versions of the TSES—a long form, which consists of 24 items, and a short form, which comprises 12 items from the long form. For this research, the short form was selected because the author and instructors collaboratively determined that the items in this form were more aligned than were the items in the long form, with participants’ previous experiences in the program as well as with the content of the courses included within the
research. The TSES also was designed to include domain-specific subscales that measure three related constructs of efficacy. Each subscale consisted of four items relating to one of the following three constructs: (a) student engagement (e.g., How much can you do to motivate students who show low interest in school work?), (b) instructional strategies (e.g., To what extent can you craft good questions for your students?), and (c) classroom management (e.g., How much can you do to control disruptive behavior in the classroom?). Calculation of the teachers’ sense of efficacy score as well as subscale scores is based upon the sum of most positive responses of the items, which are written along a 9-point continuum that ranges from 1 (nothing) to 9 (a great deal), with additional anchors at 3 (very little), 5 (some influence), and 7 (quite a bit).
The short form of the instrument has been shown to yield a high overall score reliability (α = .90), as has each sub-scale: α = .86 for student engagement, α = .81 for instructional strategies, and α = .86 for management (see Tschannen-Moran & Woolfolk Hoy, 2001). Reliability measurements for the current data were .91, .80, .77, and .88 for the overall score and each subscale, respectively. The construct-related validity of TSES was measured through its correlation with other existing scales of teaching efficacy and suggested that the TSES effectively assesses the construct of efficacy (see Tschannen-Moran & Woolfolk Hoy, 2001).
Procedure
The research design represented a quasi-experimental design, namely, a nonequivalent control group design. Prior to beginning the activities associated with this research, the author met with the two instructors, referred to as Instructor #1 and Instructor #2, who were scheduled to teach the sections of the pedagogical foundations and focused practicum courses offered within the traditional format for the 2009-2010 academic year. Instructor #1, a tenure-track assistant professor and former elementary school teacher, had previous experience teaching each of the classes and was scheduled to teach three sections of the foundations course during the fall semester when the research was scheduled to begin. Instructor #2, an adjunct faculty member and former elementary school teacher and principal, also had taught both classes multiple times. She was scheduled to teach the courses associated with the blocked experience as well as one section of the foundations course and one section of the focused practicum experience during the fall semester. Instructor #2 had previously agreed to participate in the research based on her teaching assignment within the blocked experience and both instructors agreed to allow the author to recruit students from
EXAMINING VARIATIONS IN PROGRAMMATIC DELIVERY ON TEACHING CANDIDATES’ SENSE OF EFFICACY
Fall 2012 51 RESEARCH IN THE SCHOOLS
their respective classes within the traditional program.
Recruitment activities consisted of the author visiting all sections of the pedagogical foundations course and presenting information about the research during the first class of the semester. Students were provided with consent forms to complete immediately or to return by the second class, during which the initial administration of the survey was to be completed. All students enrolled in the blocked experience granted consent to participate. Within the foundations courses associated with the traditional program, 84% (n = 81) granted consent to participate.
All participants who granted consent, regardless of participation in the blocked or traditional experience, completed the TSES by the beginning of the third class of the fall semester in which they were enrolled in the pedagogical foundations course. The instrument was administered electronically, with the instructors of the respective courses reading the directions for administration provided within the instrument. To ensure the alignment of course content and delivery for the two experiences, the two instructors engaged in collaborative planning, using the department’s master syllabi for each of the two courses to guide their discussions. The focused practicum courses for both instructors occurred in the same elementary school, which is a professional development school for the university, and followed established, departmental guidelines for teaching and observing. The cooperating teachers had multiple years of experience supervising and working with preservice teachers within the context of both courses. Using the same process for delivery, all participants in the traditional experience completed the TSES a second time, referred to as the middle administration, at the beginning of the spring semester when they enrolled in the focused practicum experience. Instructor #1 and Instructor #2 taught all sections of the focused practicum experience and, thus, administered the instrument using the provided directions.
Within the blocked experience, Instructor #1 was the instructor of record for both the pedagogical foundations and practicum courses. To ensure continuity and coherence between coursework and the field experience, the instructor met multiple times with the cooperating teachers, all of whom had extensive experience working with preservice teachers within the context of a professional development school associated with the university, prior to the start of the course. The instructor and teachers jointly established a weekly observation and instruction schedule that facilitated the development of topics for seminars on various aspects of classroom management and instructional theory. Also, they developed a schedule to provide
candidates with opportunities to meet the departmental teaching and observation requirements for the practicum experience. There was a deliberate effort made by the instructor and cooperating teachers to connect academic and practitioner knowledge in support of student teachers’ learning how to enact specific teaching practices. To ensure that multiple opportunities for teaching and observing at different grade levels were present, teaching candidates equally split the semester into one primary and one upper elementary classroom. Of note, participants in the blocked experience did not complete the middle administration due to the shortened time frame to note changes in scores.
The final administration of the TSES was completed by all participants again during the last class of the practicum course. As with previous administrations, the instrument was administered electronically with directions provided by Instructor #1 or Instructor #2. For those in the blocked group, this represented the same semester as the first administration. It is important to note that for the traditional group, the final administration occurred the next semester after the first administration due to the extended time frame for the completion of both courses. To ensure parsimony and coherence within the analysis, only the administrations of the TSES that occurred at the beginning of the foundations course and end of the focused practicum course were used for the statistical analyses associated with this research.
Data Analysis
Scores for total efficacy and for each of domain-specific areas of efficacy (student engagement, instructional strategies, and classroom management) were computed for each participant and for each test administration. Descriptive statistics, including means and standard deviations, were calculated for the total efficacy scores as well as for each of the subscale scores for each administration of the surveys in each contextual variation for delivery (blocked, traditional). To ensure equality between the groups, an independent samples t test was utilized to compare scores on the first administration of the TSES. Accordingly, a Bonferroni adjustment was utilized to diminish the risk of a Type I error, yielding a statistical significance level of .0125. The same procedure was used within the statistical analysis to measure differences in the means for the final administration of the TSES. A paired samples t test with a Bonferroni adjustment then was performed to examine changes from initial administration within the pedagogical foundations course to the final administration at the end of the practicum course for each of the experiences. Finally, effect sizes were calculated using Cohen’s d to note the magnitude of the effects on efficacy scores.
S. MICHAEL PUTMAN
Fall 2012 52 RESEARCH IN THE SCHOOLS
Results Descriptive statistics for all variables for the
first and final administrations of the TSES are shown in Table 1. Figures 1-4 graphically depict changes from the first administration of the TSES to the final administration Within the figures, the middle administration of the TSES completed by participants in the traditional experience also is reflected to provide additional insights into the changes in efficacy that were experienced by this group. Figure 1 illustrates that both the blocked and traditional groups experienced growth on the overall scores of the TSES between the first and final administration of the TSES. However, although the rate of growth appears initially to be very similar for the two groups, this was not maintained while the traditional group was enrolled in the practicum course, as indicated by the increasing gap between the two groups’ scores. A similar pattern was shown in Figure 2 with respect to overall change between the first and final administration of the instrument; but, in this case, the sense of efficacy for student engagement for the traditional group demonstrated a downward trend during their practica, thereby increasing the differences noted between the two groups’ scores.
As shown in Figure 3, the traditional group began the two-course sequence with a more efficacious view of their capabilities to implement instructional strategies. During the two-course sequence, the group’s scores increased slightly between each administration of the TSES, but the overall growth in the scores remained relatively flat. On the other hand, the blocked group demonstrated an increased sense of efficacy on this subscale, surpassing the traditional group and demonstrating what turned out to be a statistically significant change from the first to final administration of the TSES (see Table 1). Figure 4 depicts scores on the efficacy for classroom management subscale. In this case, each group displayed the largest change in scores from first administration to final administration among the three subscales contained in the instrument. Similar to what was observed in Figures 1 and 2, the slope of the growth was nearly parallel initially and the blocked group started and finished with a higher sense of efficacy for the construct. However, unlike the outcomes depicted in those figures, the scores of the traditional group demonstrated only a slight deterioration during the practicum.
To examine the potential effects of self-selection sampling bias associated with participation in the blocked experience, an independent samples t test was computed comparing the scores of each group on the first administration of the TSES on the primary scale as well as each of the subscales. To minimize the likelihood of a Type I error associated with
multiple tests, a Bonferroni adjustment was completed and a statistical significance level of .0125 (.05/4) was used in evaluating the results of each test. Although the participants in the blocked semester showed slightly higher efficacy at the time of the first administration of TSES in three of the four areas measured, no statistically significant differences were noted between the groups for the overall score on the TSES (t = 0.37, p > .05), student engagement (t = 1.43, p > .05), instructional strategies (t = -1.501, p > .05), and classroom management (t = 0.60, p > .05). Thus, additional comparisons were deemed acceptable to compare the effects of participation within the two experiences and the risk of self-selection sampling bias was deemed to be minimal.
Results of the t tests with the Bonferroni adjustments to compare scores on the first and final administration of the TSES and each subscale also are shown in Table 1. The participants in the blocked experience demonstrated statistically significant growth for the overall score as well as for each subscale, respectively, with the greatest difference in classroom management. Effect sizes also were large, using Cohen’s (1988) criteria. The largest effect size was noted for efficacy for instructional strategies. Examining scores for the participants in the traditional format revealed the largest gain within the subscale for classroom management, which was statistically significant. The participant’s total TSES score increased as did the scores pertaining to the remaining subscales; however, growth did not reach statistical significance. Employing Cohen’s (1988) criteria, the effect sizes were noted to be small for instructional strategies, moderate for the total TSES score and student engagement, and high for classroom management.
The independent samples t test with a Bonferroni adjustment to compare the means of the blocked group to those of the traditional group on the final administration of the TSES indicated statistically significant differences favoring the blocked group for general teaching efficacy, effective strategies, instructional strategies, and classroom management (see Table 2). To facilitate further examinations of these differences, the effect sizes were calculated for each variable using Cohen’s (1988) d criteria. Using these criteria, effects were noted to be large for the general scale as well as each for subscale (see Table 2).
EXAMINING VARIATIONS IN PROGRAMMATIC DELIVERY ON TEACHING CANDIDATES’ SENSE OF EFFICACY
Fall 2012 53 RESEARCH IN THE SCHOOLS
Figure 1. Graphic representation of the mean TSES scores of the blocked and traditional groups for each administration of the instrument.
Figure 2. Graphic representation of the mean scores on the TSES student engagement subscale for the blocked and traditional groups for each administration of the instrument.
Figure 3. Graphic representation of the mean scores on the TSES instructional strategy subscale for the blocked and traditional groups for each administration of the instrument.
Figure 4. Graphic representation of the mean scores on the TSES classroom management subscale for the blocked and traditional groups for each administration of the instrument.
S. MICHAEL PUTMAN
Fall 2012 54 RESEARCH IN THE SCHOOLS
Table 1 Descriptive Statistics and Paired Samples t Tests
1st administration Final administration
TSES M SD M SD Cohen’s
d ta
Blocked (n = 18)
Total 67.06 8.32 83.63 8.83 1.93 5.46**
SE 25.56 4.08 29.81 3.76 1.08 3.06** IS 17.50 2.73 23.25 2.35 2.26 6.38** CM 24.00 3.83 30.56 3.75 1.73 4.89**
Traditional (n = 25)
Total 65.92 10.48 72.44 8.13 0.68 2.46
SE 23.80 3.72 25.64 4.05 0.46 1.67 IS 19.00 3.34 19.68 2.34 0.23 0.83 CM 23.12 4.94 27.12 3.10 0.95 3.43**
Note. SE = Student Engagement; IS = Instructional Strategies; CM = Classroom Management aBlocked, df = 35; Traditional, df = 49 **p < .01 Table 2 Comparison Final Administration Scores of Traditional and Blocked Groups
Traditional (n = 25)
Blocked (n = 18)
TSES M SD
M SD Mean Score Differences
t (df = 41)
Cohen’s d
Total Score 72.44 8.13 83.63 8.83 11.19 4.15** 1.12 SE 25.64 4.05 29.81 3.76 4.17 3.31** 0.95 IS 19.68 2.34 23.25 2.35 3.57 4.76** 1.23 CM 27.12 3.10 30.56 3.75 3.44 3.19** 0.92 Note. SE = Student Engagement; IS = Instructional Strategies; CM = Classroom Management **p < .01
Discussion
Acknowledging efficacy for teaching as a potentially important factor in helping prospective teachers improve their teaching and impact student learning, researchers in teacher education have recommended that candidates receive multiple opportunities for systematic experimentation with teaching practices in authentic contexts (Grossman & McDonald, 2008). Early teaching experiences within a preparation program have the potential for long-term implications. It is, therefore, important that we examine the impact of programs that include field experiences on teachers’ beliefs and knowledge, as
well as examine structures within field experiences that maximize their influence (Sleeter, 2005).
This investigation demonstrated that efficacy was positively impacted through participation in both the traditional and blocked experiences as demonstrated by the overall increases in the means of each group for the total score on the TSES as well as those for each domain-specific area. These findings were consistent with those of Posnanski (2007) and Redmon (2007) and reinforced several tenets of Bandura’s (1997) theories inherent within the development of general efficacy. In essence, all candidates were provided with opportunities to plan lessons as well as to observe and to teach in
EXAMINING VARIATIONS IN PROGRAMMATIC DELIVERY ON TEACHING CANDIDATES’ SENSE OF EFFICACY
Fall 2012 55 RESEARCH IN THE SCHOOLS
classrooms; thus, mastery and vicarious experiences likely contributed to increases in efficacy scores for candidates, regardless of the specific experience. However, it is notable that differential rates of growth were observed between candidates that showed no statistically significant differences at the outset of the research. This will be explored further with probable explanations linked to specific characteristics attributable to the design and implementation of blocked experience.
Although the middle administration of the TSES was not utilized in statistical analyses, it is important to engage in a visual examination of Figures 1-4, which include this administration. With the exception of the efficacy relative to the implementation on instructional strategies (i.e., Figure 3), the participants in the traditional format initially experienced a rate of growth in efficacy similar to those in the blocked group while completing the foundations course. This increase in efficacy occurred within a semester that was theory driven and lacked any field experience, although the candidates did engage in 10 hours of observation. Clift and Brady (2005) found that combinations of observations and simulations, which also were utilized to develop strategies for classroom management in the course, had a positive impact on efficacy. Thus, this might account for some of the initial growth comparable to the participants in the blocked semester. It could be surmised that the growth also might represent an instance of inflated views of personal capabilities, which has been associated with preservice teaching candidates (see Rimm-Kaufman & Sawyer, 2004). The teaching candidates might have identified with the teacher or university instructor during observations of instruction, yet lacked an adequate measure of their own capabilities because there was no actual teaching experience. As a result, when these candidates entered classrooms during the field experience course, the rate of growth of their efficacy levels became flatter or, in the case of student engagement, decreased, thereby confirming that candidates need a combination of opportunities to see and to work with children when engaged in completing coursework within their teacher education programs (Clift & Brady, 2005).
On the other hand, the blocked experience incorporated several principles that have been shown to have a positive impact on teaching efficacy. The most significant difference between this experience and the traditional one is represented by the incorporation of the pedagogical foundations course within a field experience and the use of the latter as the foundation from which content could emanate. Structuring the experience within this paradigm facilitated the planning of the format and schedule for the blocked semester as a collaborative effort between the university faculty member and classroom teachers who would supervise the teaching
candidates. There was a deliberate effort to connect academic and practitioner knowledge in support of candidates learning how to enact specific teaching practices advocated within coursework. The latter facet is important because the faculty member could trust the competent and skillful teachers to demonstrate and to model positive attitudes and effective instructional behaviors in classrooms. It ensured also candidates were able to benefit from mutual guidance from the university faculty member and cooperating teachers (see Castle et al., 2006).
Within the blocked experience, candidates received multiple opportunities to observe and to implement instructional and management-related strategies described within coursework immediately in an authentic context because of the coherence in planning the blocked experience. The capacity to make connections between theory and practice was immediate, specific, and tied to events that occurred within the context of the classroom (Allsopp et al., 2006; Darling-Hammond, 2006). Essential content discussed one afternoon in class could be observed or implemented to meet the needs of specific students the next morning. These experiences then could be deconstructed within the context of the seminar in the blocked experience. The demonstration of a skill or practice in conjunction with discussions as part of university coursework has been shown to be important in such situations because the modeling likely helped the teaching candidate not only understand the pedagogical technique, but also understand her/his own practices in relation to the demonstration (Bandura, 1997; Carlson & Gooden, 1999; Hodson & Hodson, 1998). In essence, candidates in the blocked experience were able to experience a constant state of connection and reference, both externally focused as they watched teachers and individually as they engaged in teaching practices, maximizing their growth in each area of efficacy. The traditional experience, on the other hand, was extended over two or more semesters. As a result, information introduced as part of the pedagogical foundations course might not have been observed or utilized until a period of time had elapsed, limiting candidates’ ability to connect practice and theory. One participant from the traditional experience who had only completed the foundations course noted, “While I am starting to understand some techniques for classroom management, I will need a lot more practice utilizing those techniques for me to become confident in my ability to effectively use them” (personal communication, January 23, 2010).
As students in the blocked experience engaged in lesson planning, course assignments were directly applicable to teaching practices within the context (see Allsopp et al., 2006). Students wrote plans that incorporated specific instructional strategies that they knew would be taught to various groups of students with whom they were familiar and, in many cases,
S. MICHAEL PUTMAN
Fall 2012 56 RESEARCH IN THE SCHOOLS
had developed personal knowledge through day-to-day interactions. These formally planned teaching tasks were integrated into the learning experiences of the candidates and feedback on performance and remediation was immediately provided by the cooperating teacher and/or university instructor. Examining the overall impact of these practices with regard to instructional strategies (see Figure 3), the slope of the growth for this area of efficacy is much steeper for the participants in the blocked group than for those in the traditional experience. Immersion in the blocked experience facilitated continuous growth because students could make connections between the alignment of their instruction completed in the classroom and skills taught as part of coursework (see Allsopp et al., 2006; Capraro et al., 2010).
The feedback provided by the instructor or cooperating teacher also represented an important component in field experiences as Bandura (1997) suggested that it can modify individuals’ beliefs about their abilities (i.e., self-efficacy). Within the blocked experiences, the immediacy and specificity of feedback was a likely cause for the significant improvements demonstrated by the candidates. Verbal persuasion could occur directly in the context of the teaching experience or shortly following in conjunction with daily seminars. Noting the impact of assistance and structured feedback from supervisors and faculty in general (see Tschannen-Moran & Woolfolk Hoy, 2007), the ability to relate the feedback directly to experiences and course content while fresh in the candidates’ memory added direct relevance due to the distinct frame of reference. The daily seminars also offered the opportunity for candidates to debrief in a setting that included other teacher candidates. Teaching candidates could connect common problems or issues related to implementation of course principles amongst the group, facilitating the realization that any problems might not be individual specific, but instead were shared. In essence, the candidates gained greater proficiency in recognizing their own instructional capabilities and teaching competence due to the daily shared experiences and feedback of the group.
The latter notion of group experiences is significant because Tschannen-Moran and Woolfolk Hoy (2007) found verbal persuasion in the form of collegial and community support to be impactful on self-efficacy beliefs. Although collective efficacy was not measured as part of this research, Instructor #2 (personal communication, February 2, 2010) described a sense of shared purpose that developed amongst participants. These candidate-to-candidate relationships, coupled with sustained instructor-to-candidate and cooperating teacher-to-candidate relationships, supported the development of a strong professional community. Researchers have noted that such communities enhance efficacy through the rapport that develops amongst members, including
the candidates, classroom supervisors, and faculty members (Bandura, 1997; Clifford & Green, 2004; Tschannen-Moran, Woolfolk Hoy, & Hoy, 1998). Perhaps inhibiting this relationship for the traditional group was the extended time period between experiences as well as the traditional 3-day meeting schedule for coursework. In essence, students were unable to establish these relationships because class composition might have consisted of other students with whom they had little familiarity.
Limitations
This research study has several limitations that must be acknowledged. A small sample size and purposive sampling might have influenced the findings. In particular, participants from the blocked experience might not be representative of the general population of teaching candidates enrolled in the program due to their decision to participate in the program. Further, no methods, beyond comparisons of initial TSES scores, were employed to verify or to ensure through data collection that participants were similar. Additionally, Tschannen-Moran and Woolfolk Hoy (2001) noted that efficacy might vary across contexts, disciplines, and student populations. This would indicate an additional limitation for this study because no specific measures were in place to control the content areas in which the teaching candidates planned and taught lessons. For example, an individual who believed they were efficacious in mathematics might have planned lessons in this content area, which might have led to improvements in efficacy that were greater than if the student had been required to plan a lesson for social studies. Similar conclusions might be noted in relation to the grade levels in which the experiences occurred. Finally, although the author served as a course advisor for both Instructor #1 and Instructor #2, no external fidelity checks of instructional practices were conducted; thus, this must be recognized as a limitation.
Implications for Research
The findings of the current study can be used to inform examinations of teacher education programs as they seek to introduce prospective teachers to effective strategies and techniques and to reinforce these practices in such a way that they will be maintained during field and novice teaching experiences. Additional insights are presented that will address potential research activities for colleges of education to extend their capability to facilitate the development of a positive sense of efficacy within their candidates.
This research study did not include specific observations of practice, but teacher educators and researchers likely would gain valuable information by correlating changes in efficacy with actual
EXAMINING VARIATIONS IN PROGRAMMATIC DELIVERY ON TEACHING CANDIDATES’ SENSE OF EFFICACY
Fall 2012 57 RESEARCH IN THE SCHOOLS
observations of practice. Increases in efficacy scores demonstrated by the candidates in this study showed perceptions of improvement, but observational data would have provided validity of the effectiveness of the blocked approach and further established the relationship between efficacy and practice. To facilitate this process, teacher preparation programs should consider correlating information gained through observation protocols among multiple measures to triangulate the impact of efficacy and to verify the relationship between efficacy and behaviors. Ideally, the same measures would be used throughout the program to ensure the ability to examine changes in both scores and practices longitudinally. The long-term impact of this is represented by improvements in candidates’ abilities more accurately to differentiate among different aspects of practices (i.e., implementing effective strategies or classroom management techniques), which Edwards, Higley, Zeruth, and Murphy (2007) noted was a skill that many candidates lacked.
Self-reflection on practices remains an important component within the development of personal efficacy when engaged in field experiences. One challenge associated with the intensity of the blocked semester described by Instructor #1 (personal communication, February 2, 2010) was that taking a methods course concurrently with a field experience limited candidates’ ability to engage in critical reflection outside of the seminar. Although connections could be made and discussed in the context, because the experience was so intensive due to the requirements of being in the classroom daily as well as engaging in lesson planning and maintaining a rigorous academic schedule of other university classes, candidates had difficulty reflecting on the big picture. To alleviate this, videos might serve as an important self-reflective tool that would help candidates feel less overwhelmed and create more time for reflection. Van Es and Sherin (2002) found that using video within instruction slowed down what is often perceived as a fast pace within the classroom and allowed preservice teachers to notice and to recall particular, specific aspects of the classroom that were not evident to them during teaching. Thus, the use of videos within field experiences might facilitate candidates’ ability to examine practices and to reflect upon their actions, creating more accurate beliefs about or perceptions of ability. It is recommended that teacher educators explore this form of technology for its potential to help preservice teachers study and learn from their experiences to improve their practices and beliefs (Rosean, Lundenberg, Cooper, Fritzen, & Terpstra, 2008; van Es & Sherin, 2002).
Finally, the daily seminars conducted in the blocked format offered the opportunity for candidates to debrief in a setting that included other teacher candidates. Teaching candidates could connect common problems or issues related to
implementation of course principles amongst the group, facilitating the realization that any problems might not be individual specific, but instead were shared. In essence, the candidates gained greater proficiency in recognizing their own instructional capabilities and teaching competence due to the daily shared experiences and feedback of the group. The latter notion of group experiences is significant because Tschannen-Moran and Woolfolk Hoy (2007) found collegial and community support to be impactful on efficacy. This investigation indicated a sense of shared purpose among participants in the blocked experience. Thus, additional research on candidate-to-candidate relationships, coupled with sustained instructor-to-candidate and cooperating teacher-to-candidate relationships, could help educators further understand the conditions that support the development of a strong professional community among preservice teaching candidates and its related impact on efficacy.
Experts continue to agree that the beliefs of the teacher impact her/his success in the classroom and that these beliefs have a long-term impact on effective teaching, classroom management and, subsequently, student learning (Woolfolk Hoy & Spero, 2005). As a result, educational stakeholders at all levels, but especially those in teacher education, must direct specific attention towards those variables that are likely to demonstrate the most significant impact on the formation of positive beliefs. This includes carefully structuring and scaffolding learning experiences that will enable teaching candidates to connect the theoretical knowledge gained with university classrooms to the practical knowledge enacted in the authentic classroom contexts. Doing so should create the effective teaching force that is sought as well as ensure that this teaching force remains in the profession and impacts the children they serve.
The lead editors for this article were Anthony J. Onwuegbuzie and John R. Slate.
References Allinder, R. M. (1994). The relationship between
efficacy and the instructional practices of special education teachers and consultants. Teacher Education and Special Education, 17, 86-95. doi:10.1177/088840649401700203
Allsopp, D. H., DeMarie, D., Alvarez-McHatton, P., & Doone, E. (2006). Bridging the gap between theory and practice: Connecting courses with field experiences. Teacher Education Quarterly, 33(1), 19-35. Retrieved from http://www.teqjournal.org/Back%20Issues/
S. MICHAEL PUTMAN
Fall 2012 58 RESEARCH IN THE SCHOOLS
Volume%2033/VOL33%20PDFS/33_1/07allsoppetal-33_1.pdf
Ashton, P. T., & Webb, R. B. (1986). Making a difference: Teachers’ sense of efficacy and student achievement. New York, NY: Longman.
Ball, D. L., & Forzani, F. M. (2009). The work of teaching and the challenge for teacher education. Journal of Teacher Education, 60, 497- 511. doi:10.1177/0022487109348479
Ball, D. L., Sleep, L., Boerst, T., & Bass, H. (2009). Combining the development of practice and the practice of development in teacher education. Elementary School Journal, 109, 458-474. doi:10.1086/596996
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, CA: Prentice-Hall.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York, NY: W. H. Freeman and Company.
Berliner, D. (1985). Laboratory settings and the study of teacher education. Journal of Teacher Education, 36, 2-8. doi:10.1177/002248718503600601
Borko, H., & Putman, R. (1996). Learning to teach. In D. Berliner & R. Calfee (Eds.), Handbook of educational psychology (pp. 673-708). New York, NY: MacMillan.
Burant, T. J., & Kirby, D. (2002). Beyond classroom-based early field experiences: Understanding an “educative practicum” in an urban school and community. Teaching and Teacher Education 18, 561-575. doi:10.1016/S0742-051X(02)00016-1
Capraro, M. M., Capraro, R. M., & Helfeldt, J. (2010). Do differing types of field experiences make a difference in teacher candidates’ perceived level of confidence? Teacher Education Quarterly, 37, 131-154. Retrieved from http://aggiestem.tamu.edu/papers/articles/DrRCapraro/Do%20DifferingTypes.pdf
Carlson, R. D., & Gooden, J. S. (1999). Mentoring pre-service teachers for technology skills acquisition. In J. Price, J. Willis, D. A. Willis, M. Jost, & S. Boger-Mehall (Eds.), Proceedings of Society for Information Technology & Teacher Education International Conference 1999 (pp. 1313-1318). Chesapeake, VA: AACE.
Castle, S., Fox, R. K., & O’Hanlan Souder, K. (2006). Do professional development schools (PDSs) make a difference? A comparative study of PDS and non-PDS teacher candidates. Journal of Teacher Education, 57, 65-80. doi:10.1177/0022487105284211
Chambers, S. M. (2003, February). The impact of length of student teaching on the self-efficacy and classroom orientation of preservice teachers. Paper presented at the Annual Meeting of the Southwest Educational Research Association, San Antonio, TX.
Chambers, S., & Hardy, J. (2005). Length of time in student teaching: Effects on classroom control orientation and self-efficacy beliefs. Educational Research Quarterly, 28(3), 3-9. Retrieved from http://www.eric.ed.gov/PDFS/EJ718127.pdf
Charalambous, C. Y., Philippou, G. N., & Kyriakides, L. (2008). Tracing the development of preservice teachers’ efficacy beliefs in teaching mathematics during fieldwork. Educational Studies in Mathematics, 67, 125-142.
Clifford, G., & Green, H. (2004). Teacher efficacy and beliefs. The Journal of Educational Research, 22(1), 30-41.
Clift, R. T., & Brady, P. (2005). Research on methods courses and field experiences. In M. Cochran-Smith & K. M. Zeichner (Eds.), Studying teacher education: The report of the AERA panel on research and teacher education (pp. 309–424). Mahwah, NJ: Lawrence Erlbaum Associates Publishers.
Cochran-Smith, M., & Zeichner, K. (2005). Executive summary. In M. Cochran-Smith & K. Zeichner (Eds.), Studying teacher education: The report of the AERA panel on research and teacher education (pp. 1-37). New York, NY: Routledge.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.
Cohen, L., Manion, L., & Morrison, K. (2000). Research methods in education (5th ed.). New York, NY: Routledge.
Cook, J., & Brown, J. S. (1999). Bridging epistemologies: The generative dance between organizational knowledge and organizational knowing. Organization Science, 10, 381-400. doi:10.1287/orsc.10.4.381
Darling-Hammond, L. (2006). Powerful teacher education. San Francisco, CA: Jossey-Bass.
Darling-Hammond, L. (2010). Teacher education and the American future. Journal of Teacher Education, 61(1-2), 35-47. doi:10.1177/0022487109348024
Darling-Hammond, L., Hammerness, K., Grossman, P., Rust, F., & Shulman, L. (2005). The design of teacher education programs. In L. Darling-Hammond & J. Bransford (Eds.), Preparing teachers for a changing world (pp. 390-441). San Francisco, CA: Jossey Bass.
EXAMINING VARIATIONS IN PROGRAMMATIC DELIVERY ON TEACHING CANDIDATES’ SENSE OF EFFICACY
Fall 2012 59 RESEARCH IN THE SCHOOLS
Dembo, M. H., & Gibson, S. (1985). Teachers’ sense of efficacy: An important factor in school improvement. The Elementary School Journal, 86, 173–184. doi:10.1086/461441
Dewey, J. (1904/1964). The relation of theory to practice in education. In R. D. Archambault (Ed.), John Dewey on education: Selected writings (pp. 314-388). Chicago, IL: University of Chicago Press.
Dodds, P. (1989). Trainees, field experiences, and socialization into teaching. In T. J. Templin & P. G. Schempp (Eds.), Socialization into physical education: Learning to teach (pp. 81-105). Indianapolis, IN: Benchmark.
Edwards, M. N., Higley, K., Zeruth, J. A., & Murphy, P. K. (2007). Pedagogical practices: Examining preservice teachers’ perceptions of their abilities. Instructional Science, 35, 443-465. doi:10.1007/s11251-006-9014-1
Erawan, P. (2011). A path analysis for factors affecting pre-service teachers’ teaching efficacy. American Journal of Scientific Research, 13, 47-58. Retrieved from http://www.eurojournals.com/AJSR_13_04.pdf
Fraser, J. W. (2007). Preparing America’s teachers: A history. New York, NY: Teachers College Press.
Graham, B. (2006). Conditions for successful field experiences: Perceptions of cooperating teachers. Teaching and Teacher Education, 22, 1118-1129. doi:10.1016/j.tate.2006.07.007
Grossman, P., Hammerness, K., & McDonald, M. (2009). Redefining teaching, re-imagining teacher education. Teachers and Teaching: Theory and Practice, 15, 273-289. doi:10.1080/13540600902875340
Grossman, P., & McDonald, M. (2008). Back to the future: Directions for research in teaching and teacher education. American Educational Research Journal, 45, 184-205. doi:10.3102/0002831207312906
Gurvitch, R., & Metzler, M. W. (2009). The effects of laboratory-based and field-based practicum experience on pre-service teachers’ self-efficacy. Teaching and Teacher Education, 25, 437–443. doi:10.1016/j.tate.2008.08.006
Hammerness, K., Darling-Hammond, L., & Bransford, J. (2005). How teachers learn and develop. In L. Darling-Hammond & J. Bransford (Eds.), Preparing teachers for a changing world (pp. 358-389). San Francisco, CA: Jossey-Bass.
Hanushek, E. A., & Rivkin, S. G. (2012). The distribution of teaching quality and implications for policy. Annual Review of Economics, 4, 131-157.
doi:10.11.46/annurev-economics-080511-111001
Henry, G. T., Kershaw, D. C., Zulli, R. A., & Smith, A. A. (2012). Incorporating teacher effectiveness into teacher preparation program evaluation. Journal of Teacher Education, 63, 335-355. doi:10.1177/0022487112454437
Henson, R. K. (2003). Relationships among preservice teachers’ self-efficacy, task analysis, and classroom management beliefs. Research in the Schools, 10(1), 53-62.
Hodson, D., & Hodson, J. (1998). Science education as enculturation: Some implications for practice. School Science Review, 80(290), 17-24.
Hoy, W. K., & Woolfolk, A. E. (1990). Socialization of student teachers. American Educational Research Journal, 27, 279–300. doi:10.3102/00028312027002279
Knoblauch, D., & Woolfolk Hoy, A. (2008). “Maybe I can teach those kids.” The influence of contextual factors on student teachers’ efficacy beliefs. Teaching and Teacher Education, 24, 166-179. doi:10.1016/j.tate.2007.05.005
Korthagen, F., & Kessels, J. (1999). Linking theory and practice: Changing the pedagogy of teacher education. Educational Researcher, 28, 4-17. doi:10.2307/1176444
Korthagen, F. A., Loughran, J. J., & Russell, T. (2006). Developing fundamental principles for teacher education programs and practices. Teaching and Teacher Education, 22, 1020-1041. doi:10.1016/j.tate.2006.04.022
Marzano, R. J., Pickering, D. J., & Pollock, J. E. (2004). Classroom instruction that works: Research-based strategies for increasing student achievement. Alexandria, VA: Association for Supervision and Curriculum Development.
Montecinos, C., Walker, H., Rittershaussen, S., Nunez, C., Contreras, I., & Solis, M. C. (2011). Defining content for field-based coursework: Contrasting perspectives of secondary preservice teachers and their teacher preparation curricula. Teaching and Teacher Education, 27, 278-288. doi:10.1016/j.tate.2010.09.001
Moore, R. (2003). Reexamining the field experiences of preservice teachers. Journal of Teacher Education, 54, 31-42. doi:10.1177/0022487102238656
Morrell, P., & Carroll, J. (2003). An extended examination of preservice elementary teachers’ science teaching self-efficacy. School Science and Mathematics, 103, 246-
S. MICHAEL PUTMAN
Fall 2012 60 RESEARCH IN THE SCHOOLS
252. doi:10.1111/j.1949-8594.2003.tb18205.x
National Council for Accreditation of Teacher Education. (2008). Professional standards for the accreditation of teacher preparation institutions. Washington, DC: Author.
Oh, D. M., Ankers, A. M., Llamas, J. M., & Tomjoy, C. (2005). Impact of pre-service student teaching experience on urban school teachers. Journal of Instructional Psychology, 32(1), 82-98. Retrieved from http://www.freepatentsonline.com/article/Journal-Instructional-Psychology/132048075.html
O’Neill, S., & Stephenson, J. (2012). Exploring Australian pre-service teachers sense of efficacy, its sources, and some possible influences. Teaching and Teacher Education, 28, 535-545. doi:10.1016/j.tate.2012.01.008
Onwuegbuzie, A. J., & Collins, K. M. T. (2007). A typology of mixed methods sampling designs in social science research. The Qualitative Report, 12, 281-316. Retrieved from http://www.eric.ed.gov/PDFS/EJ800183.pdf
Pajares, F. (1992). Teachers' beliefs and educational research: Cleaning up a messy construct. Review of Educational Research, 62, 307-332. doi:10.2307/1170741
Posnanski, T. J. (2007). A redesigned Geoscience content course’s impact on science teaching self-efficacy beliefs. Journal of Geoscience Education, 55, 152-157. Retrieved from http://nagt.org/files/nagt/jge/abstracts/posnanski-v55p152.pdf
Putman, R., & Borko, H. (2000). What do new views of knowledge and thinking have to say about research on teachers’ learning? Educational Researcher, 29, 4-15. doi:10.3102/0013189X029001004
Putman, S. M. (2009). Grappling with classroom management: The orientations of preservice teachers and impact of student teaching. The Teacher Educator, 44, 232-247. doi:10.1080/08878730903180226
Putman, S. M. (2012). Investigating teacher efficacy: Comparing preservice and inservice teachers with different levels of experience. Action in Teacher Education, 34, 26-40.
doi:10.1080/01626620.2012.642285 Redmon, J. R. (2007, October). Impact of teacher
preparation upon teacher self-efficacy. Paper presented at the Annual Meeting of the American Association for Teaching and Curriculum, Cleveland, Ohio. Retrieved from http://eric.ed.gov/PDFS/ED500013.pdf
Rimm-Kaufman, S. E., & Sawyer, B. E. (2004). Primary-grade teachers’ self-efficacy beliefs, attitudes toward teaching, and discipline and teaching practice priorities in relation to the responsive classroom approach. Elementary School Journal, 104, 321–341. doi:10.1086/499756
Rosean, C. L., Lundenberg, M., Cooper, M., Fritzen, A., & Terpstra, M. (2008). Noticing noticing: How does investigation of video records change how teachers reflect on their experiences? Journal of Teacher Education, 59, 347-360. doi:10.1177/0022487108322128
Schön, D. A. (1983). The reflective practitioner: How professionals think in action. New York, NY: Basic Books.
Shanahan, C. H. (2008). Essential fieldwork for the preparation of teachers of reading for urban settings. In L. C. Wilkinson, L. M. Morrow, & V. Chou (Eds.), Improving literacy achievement in urban schools: Critical elements in teacher preparation (pp. 105-120). Newark, DE: International Reading Association.
Sleeter, C. E. (2005). Un-standardizing curriculum: Multicultural teaching in the standards-based classroom. New York, NY: Teachers College Press.
Soodak, L. C., Podell, D. M., & Lehman, L. R. (1998). Teacher, student and school attributes as predictors of teachers’ responses to inclusion. Journal of Special Education, 31, 480-497. doi:10.1177/002246699803100405
Tschannen-Moran, M., & Woolfolk Hoy, A. (2001). Teacher efficacy: Capturing and elusive construct. Teaching and Teacher Education, 17, 783-805. doi:10.1016/S0742-051X(01)00036-1
Tschannen-Moran, M., & Woolfolk Hoy, A. (2007). The differential antecedents of self-efficacy beliefs of novice and experience teachers. Teaching and Teacher Education, 23, 944-956. doi:10.1016/j.tate.2006.05.003
Tschannen-Moran, M., Woolfolk Hoy, A., & Hoy, W. K. (1998). Teacher efficacy: Its meaning and measure. Review of Educational Research, 68, 202-248. doi:10.2307/1170754
van Es, E. A., & Sherin, M. G. (2002). Learning to notice: Scaffolding new teachers’ interpretations of classroom interactions. Journal of Technology and Teacher Education, 10, 571-596. Retrieved from http://thorndike.tc.columbia.edu/~david/MTSU4083 /Readings/vanes.pdf
Wagler, R., & Moseley, C. (2005). Preservice teacher efficacy: Effects of a secondary education
EXAMINING VARIATIONS IN PROGRAMMATIC DELIVERY ON TEACHING CANDIDATES’ SENSE OF EFFICACY
Fall 2012 61 RESEARCH IN THE SCHOOLS
methods course and student teaching. Teacher Education & Practice, 18, 442-457.
Wilson, A., Floden, R., & Ferrini-Mundy, J. (2002). Teacher preparation research: An insider's view from the outside. Journal of Teacher Education, 53, 190-204. doi:10.1177/0022487102053003002
Woolfolk, A. E., Rosoff, B., & Hoy, W. K. (1990). Teachers’ sense of efficacy and their beliefs about managing students. Teaching and Teacher Education, 6, 137-148 doi:10.1016/j.tate.2005.01.007
Woolfolk Hoy, A., & Spero, R. B. (2005). Changes in teacher efficacy during the early years of teaching: A comparison of four measures. Teaching and Teacher Education, 21, 343-356. doi:10.1016/j.tate.2005.01.007
Zeichner, K. (1992). Rethinking the practicum in the professional development school partnership. Journal of Teacher Education, 43, 296-307. doi:10.1177/0022487192043004009
Zeichner, K. (2007). Professional development schools in a culture of evidence and accountability. School-University Partnerships: The Journal of the National Association of Professional Development Schools, 1(1), 9-17.
Zeichner, K. (2010). Rethinking the connections between campus courses and field experiences in college- and university-based teacher education. Journal of Teacher Education, 61, 89-99. doi:10.1177/0022487109347671
Zeichner, K., & Conklin, H. (2005). Teacher education programs. In M. Cochran-Smith & K. Zeichner (Eds.), Studying teacher education: The report of the AERA panel on research and teacher education (pp. 645-735). New York, NY: Routledge.
Copyright 2012 by the RESEARCH IN THE SCHOOLS Mid-South Educational Research Association 2012, Vol.19, No. 2, 62-74
Fall 2012 62 RESEARCH IN THE SCHOOLS
Research Teaching Pedagogy: Lessons Learned From an Arabic Language Intervention Pilot Study
Thomas W. Christ University of Bridgeport
Hosam Elmetaher
Ubon Ratchathani University, Thailand
Teaching mixed methods and action research
to graduate students has great advantages but also poses interesting challenges. Many students who take research methodology courses at institutions such as the University of Hawaii are English Second Language learners with scholarships from a variety of sources including the Ford Foundation and the East-West research center. These students are under stringent timelines: 2 years for a masters or 4 years for a doctoral degree. Time becomes a significant constraint when conducting their research, and often they take their first research methodology course in the second year of the program after they have already collected data for their projects. Many students do not have a clear understanding of research processes or the importance of clearly articulating the purpose, research questions, methodology, and procedures (Christ, 2007, 2010; Creswell, 2010; Creswell & Plano Clark, 2010; Greene, 2007, 2008; Mertens,
2009; Teddlie & Tashakkori, 2009). Scholarship recipients at the master’s level must collect data during the first summer of their programs regardless of knowledge, or lack thereof, about research methodology. This often results in backward research planning (i.e., the process of collecting data before fully developing the methodology and procedures, including redesigning research questions, changing typology classification, modifying the methodology and methods; cf. Christ, 2009) out of necessity due to restrictions in available data and time constraints. Furthermore, many students indicate they are conducting mixed methods research without knowledge of this methodology or the training necessary to understand applicable procedures.
Students who take an introductory mixed methods research course in their second year after data are collected have fewer research design options from which to choose. Supporting these students becomes more complex because the methodology and data analysis procedures are restricted by the available data set. Backward research planning a thesis or pilot project clearly is not ideal but is commonly the challenge faced by
Correspondence concerning this article should be addressed to Thomas W. Christ, University of Bridgeport Carlson Hall, Room 116, 126 Park Ave Bridgeport, CT 06604. Email: [email protected]
Teaching mixed methods and action research to graduate students has great advantages but also poses interesting challenges. The purpose of this article is to make transparent the process by which a mixed methods research instructor helped inform a student-researcher of the limitations in a pilot study that was compromised by methodology, including sampling conditions and analysis procedures. The student-researchers’ Communicative Language Teaching (CLT) intervention was proposed initially to be examined via a pretest-posttest design intended to determine whether video clips were an effective intervention to teach Arabic language and Egyptian customs to non-native speakers. Although data included videotaped teaching sessions, interviews, observations, expert testimonials, and pre- and post-surveys, the research design, sample size, and analysis procedures were insufficient to assess causality. At best, the data collected before the student-researcher had taken an introductory mixed methods research course were useful to understand the participants’ perceptions about the effectiveness of CLT techniques for learning. This article provides specific suggestions with applicable references about how the CLT pilot study could be improved and presents a rationale for classifying the study as an exploratory multimethod study, or an arrested action research study, rather than as a mixed methods effectiveness study as originally proposed. Advice for novice researchers concerning paradigms, stages within the research process, interactive research elements, typologies, and issues of credibility are provided along with a 7-step planning process designed to improve novice researchers’ projects.
THOMAS W. CHRIST AND HOSAM ELMETAHER
Fall 2012 63 RESEARCH IN THE SCHOOLS
those teaching research and mentoring students (Christ, 2009). This was the case in the Communicative Language Teaching (CLT) Arabic language intervention study. By the time the student-researcher entered the introductory mixed methods research class, the intervention designed by the student and an interim advisor had been administered and data including video, surveys, interviews, and expert testimonial had been collected. Opportunities to redesign and to improve the project were severely constricted as a result of time limitations.
When working with students who have begun their projects before having a clear understanding of methodological options, the research-teacher often becomes a mentor who advises students how to refine the research design and procedures. In the case of the CLT pilot study, rich data were collected that logically connected to the topic under investigation, but the research design precluded the opportunity to demonstrate causality and the limited sample size adversely affected both statistical power and the generalizability of the findings. Thus, it became necessary to provide the student-researcher curricular advice and written feedback as a way to support choosing the most applicable methodology and procedures. Although the student first defined the methodology as representing mixed methods research, other approaches including multimethod research (Brewer & Hunter, 1989), action research (Mertler, 2008), and case study (Yin, 2003) needed to be considered.
Christ (2010, 2013), Frels, Onwuegbuzie, Leech, and Collins (2012), Ivankova (2010), and Mertens (2010) all have documented pedagogical challenges and strategies to teaching mixed methods research, but no published works were found that specify what can be undertaken when the focus of the research project changes as a result of data limitations. Thus, the purpose of this article is to present a case study that illustrates a common mistake made by beginning researchers: collecting data that do not adequately reflect what the research questions/hypotheses intended or that could not possibly be analyzed in the ways in which the researcher assumed. Often, researchers end up having to reconsider the analytic approach and/or research questions/hypotheses; however, this should be a rare event and certainly discouraged. This article presents the challenge faced by the research-teacher in the CLT pilot study, to provide information to the student-researcher about why demonstrating a causal relationship between the intervention and speaking and listening skills as proposed by the student-researcher under advisement from an interim advisor was impossible due to improper sampling procedures and limited measures and available data. Determining the strength of the relationship
as originally conceived via the limited data set would have resulted in meaningless conclusions. However, other methodological options were available for analyzing descriptively the numerical data, interview data, expert testimonials, and student-researchers’ self-reflections. Methodological Options
According to Brewer and Hunter (2006), multimethod research entails different styles of research combined in the same research project and is not restricted to the use of both quantitative and qualitative data. For example, qualitative participant observations could be combined with qualitative in-depth interviewing, or quantitative surveys and norm-referenced measures could be included in a quasi-experimental research design. In the case of the CLT study, the student-researcher hoped to demonstrate causal inferences using a quasi-experimental design with survey and observation data presented as descriptive statistics to examine a possible post-intervention difference. Limitations of the research design would have made causal inferences invalid. As an alternative, the student-researcher could have conducted qualitative research techniques to assess causality (cf. Maxwell, 2004) but unfortunately did not have knowledge of this approach when the data were collected. Once the constraints of the available data sets and various methodological options were considered, the causal line of reasoning was abandoned.
During the introductory mixed methods course, the student-researcher was provided with specific readings about various research typologies (cf. Christ, 2009). Emphasis was placed on determining how to analyze and to merge the quantitative and qualitative data strands (Christ, 2008a, 2008b, 2010) when considering methodological options. After reading the provided curricula and creating methodological maps, the student-researcher focused on four methodological approaches: mixed, multi, case study, and arrested action research (i.e., a single research cycle; cf. Mertler, 2008). The most difficult distinction for the student-researcher was the difference between multi- and mixed methodologies. After reviewing Pat Bazeley’s definition as quoted in Johnson, Onwuegbuzie, and Turner’s (2007) article, the distinction between multimethod and mixed methods research was further clarified:
Multimethods research is when different approaches or methods are used in parallel or sequence but are not integrated until inferences are made. Mixed methods research involves the use of more than one approach to or method of design, data collection, data analysis within a single program of study, with integration of the different approaches or methods occurring during the program of
RESEARCH TEACHING PEDAGOGY: LESSONS LEARNED FROM AN ARABIC LANGAUGE INTERVENTION PILOT STUDY
Fall 2012 64 RESEARCH IN THE SCHOOLS
study, and not just at its concluding point. (p. 119)
This definition helped the student-researcher to make the distinction between multi- and mixed methods. For the CLT intervention, considering multi- versus mixed research promoted a healthy class discussion about the value of different forms of data when merged to give meaning and what can be concluded when quantitative data are used for descriptive purposes in an intervention study. The CLT Pilot Study
The second language learning literature reveals that Grammar Translation Methods and rote memorization are the hallmark of current language instruction techniques. For learning to occur, foreign language learners need receptively to hear, to process, to comprehend, and to express in conversation words, phrases, and sentences for language skills to maintain (Gordon, 2001; Littlewood, 1981; Medd & Whitmore, 2001). Second language acquisition literature indicates that the Communicative Language Teaching approach is a viable alternative to Audiolingualism and Grammar Translation Methods. Because no literature was found specific to CLT techniques used when teaching the Arabic language, the strength of the pilot study is in exploring participant perceptions about the intervention. Although the original intent of the study was to examine whether CLT was an effective intervention for increasing Arabic communication skills with non native speakers, constraints in data and time limitations precluded a causal research design, although exploring the phenomenon and describing the results were possible.
Video recordings were used in three unique ways in this study: first as a source of information about how the Arabic language and culture was introduced as part of the intervention; second, to establish whether CLT techniques were used when teaching students; and third, if the CLT techniques were used, to determine whether they showed signs of helping, engaging, and improving the participants’ Arabic language proficiency. Two research questions were initially proposed to the student-researcher’s interim advisor before data were collected.
(a) To what extent does the CLT curriculum improve and develop the foreign language learners’ listening and speaking skills?
(b) What are the perceptions of the participants concerning their experiences with CLT strategies as a method of increasing Arabic speaking and listening skills?
After collecting the data and reviewing curricula assigned in the introductory mixed methods research course, the student-researcher realized that the first question, posed to the interim
advisor about generating a causal research hypothesis, could not be answered reliably given the limited data, time remaining in program, and restricted sample size. Demonstrating a causal relationship with any degree of credibility was less than ideal because the student-researcher also was the Arabic language teacher in this CLT project. After extensive discussions, review of the literature, and written feedback, the causal hypothesis was abandoned and the participants’ experience question remained, a somewhat weaker guiding question, but the best question that could be addressed given the available data. The student-researcher determined that multimethod exploratory case study was more applicable than was a quasi-experimental or a mixed methods research design (Creswell, 2008).
To clarify why this decision was made, it is important to present to the readers how the existing data restricted methodological options. In the CLT study, pre- and post-speaking and listening tasks were conducted to examine whether the participants’ skills improved as a result of the intervention. The independent variable was defined by the student-researcher as the “CLT curriculum and strategies,” whereas the dependent variable was considered to represent “ability in Arabic language, listening, and speaking skills.” The student-researcher’s first indication that a quasi-experimental research design was less than ideal occurred when attempting to define both the independent and dependent variables. In this case, after reviewing the literature, the student-researcher recognized that CLT is a construct with numerous definitions and teaching strategies that vary considerably. Although the intervention was operationally defined for the study, determining the dependent variables used to measure increases in performance turned out to be more challenging. As originally designed, the only quantitative measures were scores from a 6-item learning task (used as pre- and post-measures) and a 6-item rating scale provided to the study participants. Coded interviews and testimonies from experts who observed the videotaped teaching sessions were indicated by the student-researcher to be the qualitative markers of change. The CLT Intervention
Communicative Language Teaching is "an approach to foreign or second language teaching which emphasizes that the goal of language learning is communicative competence which seeks to make meaningful communication and language use a focus of all classroom activities" (Richards & Schmidt, 2002, p. 90). The CLT curriculum objectives designed for Arabic foreign language learners were taught over six 1-hour classes (see Figure 1). The pre-interview provided information about why they were interested in learning Arabic
THOMAS W. CHRIST AND HOSAM ELMETAHER
Fall 2012 65 RESEARCH IN THE SCHOOLS
and if they knew any Arabic writing and reading skills. Pre-intervention speaking and listening tasks were administered and video-recorded to establish a baseline. Post-intervention speaking and listening tasks then were administrated and video-recorded as a way to identify change in the participants’ levels of language proficiency. Data collection procedures are visually represented for clarity in Figure 2.
The task involved eliciting the following information via six items that were administered pre- and post-intervention: the study participant’s name, country, and university, followed by questions about food. Three professors and three language instructors then observed videos of CLT being taught followed by listening and speaking measures being re-administrated. This procedure provided the opportunity for determining whether CLT strategies were being used and whether changes in speaking and listening skills could be gauged by the panel of experts as well as by the Arabic language student-researcher-teacher. Post-surveys and interviews allowed for the Arabic language students to offer their opinions about the intervention and their perceptions of the class in general. Specifically, the participants were asked to
comment on whether they believed the intervention changed their expressive and receptive Arabic communication skills. According to the student-researcher, the main goal of the intervention was not to assess the degree to which the participants mastered the course content, but to measure the degree to which the target language skills (speaking and listening) had been improved and developed.
The insurmountable flaw in the original research design that necessitated the causal research hypothesis to be abandoned and a different research design to be chosen was the notion that a causal relationship could be measured between the CLT intervention and levels of Arabic language proficiency. Although results in terms of descriptive statistics were presented as a way to demonstrate that the intervention improved the Arabic language skills of the students, inferences regarding the causal nature of the relationship were not made by the student-researcher due to the limitations stemming from use of a quasi-experimental research design, small sample size, non-random sampling scheme, short duration of the intervention, and restricted type and amount of data collected.
Main goal: Improving and developing speaking and listening skills for the foreign language learners
(Through)
Culture
Content objectives (CBI) Language Objectives Expressions/idioms
About Customs and traditions
Grammar Values Gestures The restaurant
Words, structures, and sentences Intonations Oral language skills
Figure 1. Course objectives.
Six CLT class sessions
Teaching with focus on speaking and listening skills
CLT Curriculum Pre- and post-tasks (Confirmatory Quantitative) Participants (Intervention) Bounded by Pre- and post-interview and survey (Exploratory Qualitative) Bounded by
Language experts’ observations Numbers (Professors and language instructors) Arabic as foreign language
Literature Review Familiarity with Arabic reading and writing Difficulties in Arabic oral communication
Experienced the old teaching methods
Figure 2. Data collection procedures.
RESEARCH TEACHING PEDAGOGY: LESSONS LEARNED FROM AN ARABIC LANGAUGE INTERVENTION PILOT STUDY
Fall 2012 66 RESEARCH IN THE SCHOOLS
Research Methodology
The student-researcher first proposed a
concurrent mixed methods design (Creswell, 2008) for the CLT study based on the assumption that quantitative pre- and post-task data, interviews, observations, and open-ended survey items could demonstrate whether participants’ listening and speaking skills improved as a result of the intervention. After reviewing the existing data set and project timeline, it was clear to both the student-researcher and mixed methods research-teacher that answering the first research question with confidence due to the small number of participants and limited data was not possible. Because research projects are culminating activities designed for learning, it was more important for the student-researchers in the introductory class logically to deduce applicable methodologies and procedures for their projects. Assigned literature in the course (cf. Christ, 2009 for a detailed description of curricula and activities) helped the student-researcher to consider several methodological alternatives for the CLT pilot study. Helpful tools for determining methodological typologies included a simple decision-making matrix (see Table 1) and methodological maps. The methodological typology for the CLT study was determined by bounding the case considering: (a) the topic: CLT strategies; (b) time: six classes; (c) participants: college students, teacher, expert observers; (d) location: university setting; and (e) data: surveys, interviews, videotaped teaching sessions, and pre-post measures (Christ, 2010). The student-researcher created several methodological maps as a way to help define potential research designs (see Figure 3).
The student-researcher who had a very basic understanding of paradigms indicated that pragmatism (of the middle philosophy; cf. Johnson & Onwuegbuzie, 2004) was applicable for this study because the quantitative (pre- and post-tasks) and qualitative (interview, observation, survey) data were collected, analyzed, and merged to understand better the research problem. Although pragmatism was specified by the student-researcher, other philosophical assumptions that are emerging in the mixed methods literature such as critical realism (Christ, 2010; Maxwell & Mitappalli, 2010) and the transformative (Greene, 2007) stance were not considered because they were optional readings in the introductory research course. Although a philosophical worldview (ontology, epistemology, axiology) was requested
to be included in the CLT study, it was not articulated. Christ (2013), Creswell and Plano Clark (2010), and Teddlie and Tashakkori (2009) all specify that when designing research, it is important to specify one’s own worldview and the purpose of the study to help clarify that the research questions, methodology, and procedures are all consistent. Most helpful was the opportunity for the student-researcher to delineate the purpose of the study and to determine how the case was bound as a way to determine an appropriate methodology. If the purpose and parameters of the study were defined from the onset, the predisposition towards seeking to examine a causal relationship could have been avoided.
Data included in the study were self-reflections, pre- and post-speaking tasks, pre-intervention interview responses, and post-intervention responses. The pre- and post-speaking task raw scores were compared by the student-researcher with the intent of assessing changes in language skills. If the student-researcher had wanted to test for differences in the population by assessing whether differences found in the sample are statistically significant, an expected effect size would have had to be determined. With such a small convenience sample, even if applicable procedures had been followed, the results could not demonstrate with validity changes in language skills due to inadequate statistical power. The responses from the pre- and post-participant interviews and surveys, and the feedback from experts who reviewed the videotaped teaching sessions to determine how CLT teaching strategies were used were analyzed for themes using a simplified form of hand analysis (color marking themes in transcripts for quotable passages transferred to tables for review). In essence, the analysis of the qualitative data lacked sufficient rigor. Given the time constraints of 16 weeks in the introductory mixed methods research course, students with prior qualitative research training (e.g., method of constant comparison, discourse analysis, componential analysis, keywords-in-context, analytic induction, word count, domain analysis, taxonomic analysis; cf. Leech & Onwuegbuzie, 2007, 2008, 2011) were at an advantage to their peers because these techniques were only assigned as optional readings in the course. The student-researcher indicated that the three data strands (i.e., speaking task responses, interview responses, video responses) allowed for data triangulation as a way to increase credibility of the findings (Maxwell, 2005; Patton, 1999).
THOMAS W. CHRIST AND HOSAM ELMETAHER
Fall 2012 67 RESEARCH IN THE SCHOOLS
Table 1 Mixed Method, Multimethod, and Action Research Criteria
Figure 3. Methodological map.
Methodology Options Mixed Method Multimethod Action Research Purpose
Confirmatory Explanatory Exploratory
Exploratory Explanatory
Confirmatory
Self Reflective
Sequencing Concurrent Sequential
Concurrent Sequential
Sampling Purposeful Convenience Convenience
Generalization Possible Possible Not Possible
RESEARCH TEACHING PEDAGOGY: LESSONS LEARNED FROM AN ARABIC LANGAUGE INTERVENTION PILOT STUDY
Fall 2012 68 RESEARCH IN THE SCHOOLS
Pilot Study Results
According to the student-researcher’s analysis,
the CLT strategies offered participants the chance to talk, to interact, and to negotiate meaning using activities designed to enhance practicing communication. Strategies, including web-links to videos, were introduced to the participants as a way of enhancing their interest in the language through exposure to a different culture, traditions, and values. Students learned by studying, speaking, and searching for video materials, which is a hands-on approach promoted in the CLT literature (Lee & VanPatten, 2003; Littlewood, 1981). Searching links, role playing, and storytelling, according to the participants, allowed them to interact, to communicate, and to learn from each other.
Excerpts of the six sessions were observed by three professors and three English language instructors familiar with CLT teaching techniques. According to these six language experts, video as a source of curricula emerged as a particularly effective teaching strategy. One English language instructor approved of using “media and sounds in the class.” Another indicated the power of “video to gain attention of students.” Another comment by a professor included “Beginning the class with a short video on Egypt was a great idea.” The expert observers stated that many of the prominent CLT teaching strategies as indicated in the literature were observed (Liao, 2000a, 2000b; Littlewood, 1981).
Clearly, the data collected and analyzed by the CLT student-researcher were descriptive in nature and useful for understanding how the strategies were viewed by participants, the CLT teacher, and language experts, but could not be used to assess statistical significance (i.e., p values) or practical significance (i.e., effect size) (cf. Onwuegbuzie & Daniel, 2005) due to the convenience sample and small number of participants in the study. The six pre- and post-speaking and listening tasks conducted by the 10 participants in the study were presented simply in tables as a visual display demonstrating increased language performance due to the intervention.
As an alternative to presenting raw data in tables, the student-researcher could have used the test of multiple group differences devised by Onwuegbuzie and Levin (2005), who conceptualized three methods for testing the trend of multiple group differences. Onwuegbuzie and Levin indicated that the Binomial Test of Trend is the most notable, in which the binomial distribution is used to determine whether the number of group differences in the same direction (positive/negative) should be regarded as either a statistically real or a statistically chance finding. Specifically, as conceptualized by Onwuegbuzie,
Levin, and Ferron (2011), for a 5% level of statistical significance,
when a study includes at least five outcome measures (for directional alternatives) or at least six outcome measures (for nondirectional alternatives), the binomial formula can be used to determine the probability of obtaining the observed proportion of findings in the same direction, under the null hypothesis specification that p (the success probability) and q (the failure probability) are each equal to .5 (i.e., assuming that a difference in one direction is as likely as obtaining a difference in the opposite direction). If the observed proportion differs statistically from .5, an effect-size index (typically the observed proportion itself, which serves as an unbiased estimator of p), along with a corresponding confidence interval, could be reported and interpreted. (p. 130) Thus, using this procedure in the case of the
CLT study, the probability that all six post-test scores were higher than was the pretest scores is less than .05, which indicates a statistically significant finding could have been used to address the first research question. Unfortunately, the Binomial Test of Trend was not used for statistical analysis in the case of the CLT pilot study because several of the post-test items were modified, precluding a meaningful comparison of the results.
Participants asked to comment on their experience with CLT strategies and their views about Arabic communication skills reported self-perceived improvements as a result of learning CLT strategies. One participant indicated “During the class I felt my listening and speaking were improved. I can distinguish some words in Arabic conversation now. I learned some necessary words and sentences in Arabic conversation.” Another participant stated the following:
I knew almost nothing when I began this course. Now I know some useful phrases and the numbers. Although I made many mistakes and still have not mastered everything presented in the class, I came to understand a great deal of what the teacher spoke, I was able to produce language independently, and my answers were understandable to both the teacher and fellow students. Analysis of the post-surveys, according to the
student-researcher, also revealed that the participants found communicative activities, opportunities to practice, and visual aids helpful to improve their Arabic language skills. A series of quotations from the 10 participants was presented as evidence to support these conclusions, such as the following: “language has been presented very vividly”; “I think I will remember and also be more capable of productive use”; “great interactions
THOMAS W. CHRIST AND HOSAM ELMETAHER
Fall 2012 69 RESEARCH IN THE SCHOOLS
between cultures, between individuals…Yes, definitely, there were lots of activities which enabled everyone to participate”; and “I most enjoyed the group activities where I spoke with other students. Knowing that other students were also experiencing difficulty with pronunciation I felt less pressure.”
Although the selected quotations present some evidence that support categories that the student-researcher presented in the conclusions section of the study, it was not clear how these categories were created. Credibility of the categories and themes could have been much improved if four steps suggested by Constas (1992) had been made explicit: (a) where the responsibility or authority for the creation of categories resided (i.e., participants, programs, investigative, literature, or interpretive); (b) what the grounds were on which one could justify the existence of a given set of categories (i.e., external, rational, referential, empirical, technical, or participative); (c) what was the source of the name used to identify a given category (i.e., participants, programs, investigative, literature, or interpretive); and (d) at what point during the research process the categories were specified (i.e., a priori, a posteriori, or iterative).
Pre- and post-interviews used in the CLT study according to the student-researcher “revealed that motivation to learn Arabic differed for all the participants.” Although the student-researcher had been instructed by his interim advisor to use a pretest-posttest (pre-experimental) design, it was not until after data were collected for the study that readings assigned in the mixed methods class revealed that pre-experimental designs are considered to be extremely weak because it is impossible to make trustworthy causal statements. Gay and Airasian (2003) indicated that pre-experimental designs are so weak that they simply should be avoided because findings would be so questionable that they could not be useful for most purposes. Clearly, the CLT study, which was a one-group pretest-posttest design, could not control for the countless threats to internal validity (e.g., history, maturation, testing, instrumentation, regression) and external validity (e.g., pretest-treatment interaction).
Post-videotaped interview results, according to the student-researcher, support the use of communication activities and technology aids as an alternative to memorization. Quotations provided to support this conclusion included the following: “I think the activities were very important for us to learn…because during those activities we had to communicate with each other in Arabic what we learned in class”; “It is very comfortable engaging in communication and interaction among people”; and “we learned to communicate and to interact by the class activities.”
The student-researcher concluded that CLT strategies appeared to improve students’ abilities to communicate as seen in the higher post-speaking and listening scores. With inferential tests being invalid due to the small sample size (cf. Onwuegbuzie & Levin, 2005), this conclusion was unwarranted; yet, the qualitative data including student reports of their language performance, and observations of use, suggested improved Arabic speaking skills. Experts who viewed the videotapes also supported this conclusion. The student-researcher’s qualitative findings did align with prior research about CLT (Galloway, 1993; Gatbonton & Segalowitz, 2005; Jones & Wang, 2001; Lee & VanPatten, 1995, 2003; Liao, 2000a, 2000b) adding a level of credibility to the student-centered theme that emerged.
The most important process of conducting a pilot study and taking an introductory mixed methods research course was not the knowledge gained from the pilot study itself, but the ability to understand a variety of research methodologies, articulate numerous limitations inherent in the initial proposal (i.e., pre-experimental design, convenience sample, small sample size, no control group, threats to verification and trustworthiness of the qualitative interview data analyses techniques), and what could be undertaken to improve the study in the future.
Conclusions
Backward research planning, a term used to
describe how the student-researcher collected data before a clearly articulated research design was concluded, is common. The CLT example demonstrates the limitations faced by the student-researcher and why researchers should consider available time and specify the purpose of the study when designing the research questions and accompanying methodology. Because mixed and action research have increased in popularity over the past 10 years, many students indicate that their studies are mixed without understanding the qualities of these research designs (Christ, 2010; Teddlie & Tashakkori, 2009). Fortunately, the number of experts with knowledge about multi-, mixed-, and action research also has increased, offering opportunities for appropriate guidance. As Tashakkori and Creswell (2008) indicated, it is the responsibility of “stewards of the social-behavioral research enterprise” (p. 291) to help those new to the arena who are struggling with terminology, methods, and methodology. The CLT pilot project highlights several points that stewards of mixed and action research should consider, including a reasonable level of rigor, criteria for establishing legitimacy (Collins, Johnson, & Onwuegbuzie, 2012) and authenticity (Guba & Lincoln, 1989), in terms of stated methodology, paradigmatic
RESEARCH TEACHING PEDAGOGY: LESSONS LEARNED FROM AN ARABIC LANGAUGE INTERVENTION PILOT STUDY
Fall 2012 70 RESEARCH IN THE SCHOOLS
consistency, and credible evidence. For example, the CLT study demonstrates how numerical data increased the knowledge gained about the phenomenon, but the data that were collected could not be used to demonstrate a causal relationship. The CLT pilot study highlights several shortcomings that can inform those new to the field of research: the importance of discussing threats to verification, trustworthiness, authenticity, credibility, transferability, and dependability of qualitative data (e.g., Creswell, 2007; Guba & Lincoln, 1989; Lather, 1993; Lincoln, 1995; Lincoln & Guba, 1985; Maxwell, 1997, 2004, 2005; Miles & Huberman, 1994). Many of these threats could have been countered via one or more of the following 18 qualitative procedures: (a) prolonged engagement; (b) persistent observations; (c) triangulation of data sources or participants; (d) contextualization of observations; (e) method of constant comparison; (f) checking for representativeness of sources of data; (g) checking for researcher effects; (h) weighing the evidence; (i) examining extreme cases; (j) checking for spurious relations; (k) examining rival explanations; (l) looking for negative evidence; (m) obtaining feedback from informants; (n) leaving an audit trail; (o) thick description; (p) assessing structural relationships; (q) use of referential material; and (r) theoretical sampling (for a complete explanation, see Onwuegbuzie & Leech, 2007a, 2007b).
This article is not meant to limit new and freely developed typological combinations or procedures that are emerging in the mixed and action research literature; rather, it is an appeal to those who teach methodology to maintain a high standard for logical and sound empirically based research. The challenge for student-researchers is to look holistically at study conditions and the limitations inherent in their research skills, available data, and time when determining the most appropriate methodological design (e.g., case, mixed, action research design) that they plan to use.
Four steps can help those considering the most appropriate methodological classification. First, researchers should become familiar with the underpinnings of a minimum of five paradigmatic views: constructivism, critical, transformative, pragmatism, and postpositivism (Christ, 2010). These five paradigmatic views have multiple definitions and it is critical for student-researchers to understand at least the basics of each. For example, there are at least two prominent types of constructivism that should be reviewed (a) radical constructivism (cf. Glasersfeld, 1996), which focuses more on the individual knower and acts of cognition; and (b), social constructionism (cf. Berger & Luckmann [1966] or Schwandt [1994, 2001]), which focuses more on social processes
and interaction (e.g., symbolic interactionism) and ethnomethodology that emphasize the participant’s interpretation of the situation; how social people identify, produce, and reproduce social actions; and how they come to share an intersubjective understanding of specific life circumstances. Key to both forms of constructivism is the rejection of scientific realism (i.e., the view that our theories map out real features of the world) and scientific objectivity (i.e., the view that it is possible to obtain accurate and reliable interpretations of the way the world is). Similarly, there are several forms of critical realism, with distinctions between U.S. and British versions (cf. Danermark, Ekström, Jakobsen, & Karlsson, 2002; Maxwell & Mittappali, 2010; Sayer, 1992); yet, most forms of critical realism are closely associated with Bhaskar’s view who denied there is any objective or certain knowledge of the world and that there is the possibility of alternative valid accounts of any phenomenon. Maxwell and Mittappali (2010) summarized an important tenet in the definition of critical realism: “All theories about the world are grounded in a particular perspective and worldview, and all knowledge is partial, incomplete, and fallible” (p.150).
Second, four levels of the research process should be considered by researchers as they construct their projects: (a) the epistemological level, (b) theoretical perspectives, (c) methodology, and (d) methods. Crotty (1998) summarized these four levels in his excellent text, which should be strongly considered as mandatory reading for students in introductory mixed methods research courses because it is easily understood and clarifies these complex concepts.
Third, the problem, purpose, and research questions should be understood as representing interactive elements in a research study (Christ, 2009; Maxwell, 2005). Maxwell (1997) originally published his interactive qualitative research model, which included five components: goals, conceptual framework, validity, methods, and the research question. Christ (2010) expanded Maxwell’s methodological graphic display more readily to fit mixed and action research studies: (a) Goals: personal, practitioner, policy, and intellectual; (b) Methodology: research design, typologies; (c) Methods/procedures: sampling, data collection, analysis, merging, bounding the case; (d) Validity: reliability, triangulation (data sources, paradigms, methods, participants), credibility, member check, extended observations, thick descriptions, expert audit; (e) Conceptual Framework: existing theories, researchers’ worldviews, paradigms; and (f) Research Questions: overarching and subsumed.
Fourth, there are numerous typologies that should be considered. Teddlie and Tashakkori (2009, p. 141) developed a useful tool that lists
THOMAS W. CHRIST AND HOSAM ELMETAHER
Fall 2012 71 RESEARCH IN THE SCHOOLS
seven criteria for determining potential typologies, number of approaches, strands, implementation process, integration stages, priority, purpose, and theoretical/ideological perspectives. In terms of potential methodological typologies, the CLT study could have been classified as a mixed method, multimethod, arrested action research, or a case study design. Defining these four typologies can help the reader to determine which methodology was most appropriate and why. For example, mixed methods research, according to Johnson et al. (2007), is research in which
a researcher or team of researchers combines elements of qualitative and quantitative research approaches (viewpoints, data, inferences) for the broad purposes of breadth and depth of understanding and corroboration. A mixed methods study would involve mixing within a single study; a mixed method program would involve mixing within a program of research, and the mixing might occur across a closely related set of studies. (p. 123) Creswell’s (2006) definition incorporates
paradigms and philosophical assumptions, and theoretical perspectives, as well as research questions and interpretations. In short, mixed methods encompass the totality of all phases of research, not just the methods. Brewer and Hunter’s (1989) definition of multimethod research indicates that both quantitative and qualitative research approaches are used, but they remain relatively independent until the interpretation stage.
Reason and Bradbury (2008) defined action research as a “participatory process concerned with developing practical knowing in the pursuit of worthwhile human purposes. It seeks to bring together action and reflection, theory and practice, in participation with others, in pursuit of practical solutions to issues of pressing concern to people.” (p. 4). Gorard (2010, pp. 241-242) simply defined research as “a spiral which has no clear beginning or end, in which activities (phases) overlap, can take place simultaneously, and iterate.” which is an appropriate description of the central tenet in action research.
In the case of the CLT study, numerical and verbal data briefly were analyzed separately, resulting in exploratory descriptions that were compared. Available data could not confirm if the CLT strategies improved listening and speaking skills, precluding a causal mixed methods classification; but the self-reflective activities were useful for improving student-researcher’s own teaching practices, which is critical in action research. Where the CLT study fell short of being classified as action research was the study did not follow a reiterative process of planning, acting, developing, and reflecting (Bradbury, 2007; Greenwood & Levin, 2001). Without repeated cycles of reflection and intervention modification
(Mertler, 2008), including a plan for improvements to the intervention and expressed modifications to teaching praxis (Christ, 2010), the project only could be classified as arrested action research.
Implications for Research Teachers For many student-researchers interested in
creating and evaluating an intervention, action research is an ideal form of confirmatory mixed methods research in practice (e.g., intervention improves performance, self reflection improves praxis), but it typically requires following several steps: (a) topic and intervention identification, (b) intervention administration and data collection, (c) data analysis and reflection leading to intervention modifications, (d) repeated iterations, and (e) dissemination to stakeholders. Action research is an ideal medium for learning how to conduct research because it is exploratory, confirmatory, and explanatory requiring data collection, analysis, and reflexive action (Reason & Torbert, 2001). The first step of an action research study often represents an exploratory design asking a how question with focus upon improving a situation, then moves through cycles of data collection, analysis, and reflection, becoming an explanatory and/or confirmatory design that determines if the actions have an intended effect (Christ, 2009). The action research cycles are sequential, which is one form of mixed methods design (Creswell & Plano Clark, 2007; Tashakkori &Teddlie, 1998; Teddlie & Tashakkori, 2009).
Scholars who advise students conducting a thesis or dissertation should encourage the pilot testing process to determine if a study is logical, representative of the proposed typology, and demonstrates credible and valid findings. They should provide specific written feedback to highlight any flaws and suggest readings the student-researchers can use to improve their projects in the future. In the case of the CLT study, the descriptive data presented in tables and themes that emerged from interviews and expert opinions demonstrate knowledge. Yet, the findings clearly cannot show causation between the intervention and the acquisition of speaking and listening skills. The numerical results, expert testimonials, and student interviews combined only could provide the student-researcher insight about her or his own teaching habits, the value of CLT strategies, and whether the intervention appeared to improved students’ expressive and receptive language skills. Most meaningful were themes that emerged about the use of video as a medium for increasing student engagement and the acquisition of language skills. Clearly, the use of backward research planning is far from ideal, but it is quite common at the thesis level. Most critical when first
RESEARCH TEACHING PEDAGOGY: LESSONS LEARNED FROM AN ARABIC LANGAUGE INTERVENTION PILOT STUDY
Fall 2012 72 RESEARCH IN THE SCHOOLS
engaging in research is to learn the basics, conduct the research, and learn from the shortcomings.
Strategies including following a project outline with headers for each section and creating methodological diagrams (Christ, 2009) help student-researchers to learn about the strengths and challenges associated with various forms of mixed-, multi-, and action research. By introducing paradigmatic views, terminology, methodological maps, and rubrics for evaluating research at the proposal stage (Christ, 2010), the likelihood is that the final research project will be rigorous, logical, and, perhaps, even publishable.
The lead editors for this article were Anthony J. Onwuegbuzie and John R. Slate.
References Berger, P., & Luckmann, T. (1966). The social
construction of reality: A treatise in the sociology of knowledge. Garden City, NY: Anchor Books.
Bradbury, H. (2007). Quality and actionability: What action researchers offer from the tradition of pragmatism. In A. B. (Rami) Shani, S. A. Mohrman, W. Pasmore, B. Stymne, & N. Adler (Eds.), Handbook of collaborative management research (pp. 583–600). Thousand Oaks, CA: Sage.
Brewer, J., & Hunter, A. (1989). Multimethod research: A synthesis of styles. Newbury Park, CA: Sage.
Brewer, J., & Hunter, A. (2006). Foundations of multimethod research: Synthesizing styles (2nd ed.). Thousand Oaks, CA: Sage.
Christ, T. W. (2007). A recursive approach to mixed methods research in a longitudinal study of postsecondary education disability support services. Journal of Mixed Methods Research, 1, 226–241. doi:10.1177/1558689807301101
Christ, T. W. (2008a). A cross-case analysis of leadership qualities in three postsecondary disability support centers. Journal of Ethnographic & Qualitative Research, 2, 223-230.
Christ, T. W. (2008b). Technology support services in postsecondary education. A mixed methods study. Technology and Disability, 20(1), 25-35.
Christ, T. W. (2009). Designing, teaching, and evaluating two complementary mixed methods research courses. Journal of Mixed Methods Research, 3, 292–325. doi:10.1177/1558689809341796
Christ, T. W. (2010). Teaching mixed methods and action research: Pedagogical, practical, and evaluative considerations. In A.
Tashakkori & E. Teddlie (Eds.), Sage handbook of mixed methods in social & behavioral research (2nd ed., pp. 643-676). Thousand Oaks, CA: Sage.
Christ, T. W. (2013). The worldview matrix as a strategy when designing mixed methods research. International Journal of Multiple Research Approaches, 7, 110-118.
Collins, K. M. T., Onwuegbuzie, A. J., & Johnson, R. B. (2012). Securing a place at the table: A review and extension of legitimation criteria for the conduct of mixed research. American Behavioral Scientist, 56, 849-865.
Constas, M. A. (1992). Qualitative analysis as a public event: The documentation of category development procedures. American Educational Research Journal, 29, 253-266.
Creswell, J. (2006, April). Continuing the discourse: Advocates for and challengers to mixed methods research. Symposium conducted at the American Education Research Association Mixed Methods SIG Business Meeting, San Francisco.
Creswell, J. (2007). Qualitative inquiry and research design: Choosing among five approaches. (2nd ed.).Thousand Oaks, CA: Sage.
Creswell, J. (2008). Educational research: Planning, conducting, and evaluating quantitative and qualitative research. Upper Saddle River, N.J.: Pearson/Merrill Prentice Hall.
Creswell, J. (2010). Mapping the developing landscape of mixed methods research. In A. Tashakkori & E. Teddlie (Eds.), Sage handbook of mixed methods in social and behavioral research (2nd ed., pp.45-68). Thousand Oaks, CA: Sage.
Creswell, J., & Plano Clark, V. (2007). Designing and conducting mixed methods research. Thousand Oaks, CA: Sage.
Creswell, J., & Plano Clark, V. (2010) Designing and conducting mixed methods research (2nd ed.). Thousand Oaks, CA: Sage.
Crotty, M. (1998). Foundations of social research: Meaning and perspective in the research process. Thousand Oaks, CA: Sage.
Danermark, B., Ekström, M., Jakobsen, L., & Karlsson, J. C. (2002). Explaining society: Critical realism in the social sciences. London, England: Routledge.
Frels, R. K., Onwuegbuzie, A. J., Leech, N., & Collins, K. M. T. (2012). Challenges to teaching mixed research courses. The Journal of Effective Teaching, 12, 23-44.
Galloway, A. (1993). Communicative language teaching: An introduction and sample
THOMAS W. CHRIST AND HOSAM ELMETAHER
Fall 2012 73 RESEARCH IN THE SCHOOLS
activities. In, ERIC Digests (pp. 1-8). Washington DC: ERIC Clearinghouse on languages and Linguistics . (Reprinted from www.eric.ed.gov)
Gatbonton, E., & Segalowitz, N. (2005). Rethinking communicative language teaching: A focus on access to fluency. The Canadian Modern Language Review, 61, 325 353. doi:10.1353/cml.2005.0016
Gay, L. R., & Airasian, P. (2003). Educational research: Competencies for analysis and application (7th ed.). Upper Saddle River, NJ: Pearson Education.
Glasersfeld, E. V. (1996). Radical constructivism: A way of knowing and learning. London, England: Falmer Press.
Gordon, D. (2001). Practical suggestions for supporting speaking and listening in classrooms. In P. G. Smith (Ed.), Talking classrooms: Shaping children's learning through oral language instruction (pp. 57-73). Newark, DE: International Reading Association.
Gorard, S. (2010). Research design as independent of methods. In A. Tashakkori & C. Teddlie (Eds.), Sage handbook of mixed methods in social and behavioral research (2nd ed., pp. 237-252). Thousand Oaks, CA: Sage.
Greene, J. C. (2007). Mixed methods in social inquiry. San Francisco, CA: Jossey-Bass.
Greene, J. C. (2008). Is mixed methods social inquiry a distinctive methodology? Journal of Mixed Methods Research, 2, 7-22. doi:10.1177/1558689807309969
Greenwood, D., & Levin, M. (2001). Pragmatic action research and the struggle to transform universities into learning communities. In P. Reason & H. Bradbury (Eds.), Handbook of action research (pp.103-113). Thousand Oaks, CA: Sage.
Guba, E. G., & Lincoln, Y. S. (1989). Fourth generation evaluation. Newbury Park, CA: Sage.
Ivankova, N. (2010). Teaching and learning mixed methods in computer-mediated environment: Educational gains and challenges. International Journal of Multiple Research Approaches, 4, 49-65. doi:10.5172/mra.2010.4.1.049
Johnson, R. B., & Onwuegbuzie, A. J. (2004). Mixed methods research: A research paradigm whose time has come. Educational Researcher, 33(7), 14-26. doi:10.3102/0013189X033007014
Johnson, R., Onwuegbuzie, A., & Turner, L. (2007). Towards a definition of mixed methods research. Journal of Mixed Methods Research, 1, 112–133. doi:10.1177/1558689806298224
Jones, N. B., & Wang, S. H. (2001). Applying communicative language teaching in a Taiwanese elementary school. Retrieved from http://www.eric.ed.gov/ERICDocs/data/ericdocs2sql/content_storage_01/0000019b/80/19/bf/31.pdf
Lather, P. (1993). Fertile obsession: validity after poststructuralism. The Sociological Quarterly, 34, 673–693. doi:10.1111/j.1533-8525.1993.tb00112.x
Lee, J. F., & VanPatten, B. (1995). Making communicative language teaching happen. The McGraw-Hill foreign language professional series, vol 1. New York, NY: McGraw-Hill.
Lee, J. F., & VanPatten, B. (2003). Making communicative language teaching happen. Boston, MA: McGraw-Hill.
Leech, N. L., & Onwuegbuzie, A. J. (2007). An array of qualitative analysis tools: A call for data analysis triangulation. School Psychology Quarterly, 22, 557-584. doi:10.1037/1045-3830/22/4/557
Leech, N. L., & Onwuegbuzie, A. J. (2008). Qualitative data analysis: A compendium of techniques for school psychology research and beyond. School Psychology Quarterly, 23, 587-604. doi:10.1037/1045-3830.23.4.587
Leech, N. L., & Onwuegbuzie, A. J. (2011). Beyond constant comparison qualitative data analysis: Using NVivo. School Psychology Quarterly, 26, 70-84. doi:10.1037/a0022711
Liao, X. Q. (2000a). Communicative language teaching innovation in China: Difficulties and solutions. Retrieved from http://www.eric.ed.gov/ERICDocs/data/ericdocs2sql/content_storage_01/0000019b/80/16/52/0b.pdf
Liao, X. Q. (2000b). Communicative language teaching: Approach, design and procedure. Retrieved from http://www.eric.ed.gov/ERICDocs/data/ericdocs2sql/content_storage_01/0000019b/80/16/67/40.pdf
Lincoln, Y. S. (1995) Emerging criteria for qualitative and interpretive research. Qualitative Inquiry, 3, 275–289. doi:10.1177/107780049500100301
Lincoln, Y. S., & Guba, E. G. (1985) Naturalistic inquiry. Thousand Oaks, CA: Sage.
Littlewood, W. (1981). Communicative language teaching: An introduction. Cambridge, England: Cambridge University Press.
Maxwell, J. A. (1997). Designing a qualitative study. In L. Bickman & D. J. Rog (Eds.), Handbook of applied social research
RESEARCH TEACHING PEDAGOGY: LESSONS LEARNED FROM AN ARABIC LANGAUGE INTERVENTION PILOT STUDY
Fall 2012 74 RESEARCH IN THE SCHOOLS
methods (pp. 69–100). Thousand Oaks, CA: Sage.
Maxwell, J. A. (2004). Causal explanation, qualitative research, and scientific inquiry. Educational Researcher, 33(2), 3-11. doi:10.3102/0013189X033002003
Maxwell, J. A. (2005). Qualitative research design: An interactive approach. Applied Social Research Methods Series, no. 41. Thousand Oaks, CA: Sage.
Maxwell, J. A., & Mitappalli, K. (2010). Realism as a stance for mixed methods research. In A. Tashakkori & C. Teddlie (Eds.), Sage handbook of mixed methods in social and behavioral research (2nd ed., pp. 145-167). Thousand Oaks, CA: Sage.
Medd, S. K., & Whitmore, K. F. (2001). What’s in your backpack? Exchanging funds for language knowledge in an ESL classroom. In P. G. Smith (Ed.), Talking classrooms: Shaping children's learning through oral language instruction (pp. 42-65). Newark, DE: International Reading Association.
Mertens, D. M. (2009). Transformative research and evaluation. New York, NY: Guilford Press.
Mertens, D. M. (2010). Philosophy in mixed methods teaching: The transformative paradigm as illustration. International Journal of Multiple Research Approaches, 4, 9-18. doi:10.5172/mra.2010.4.1.009
Mertler, C. (2008). Action research: Teachers as researchers in the classroom (2nd ed.). Thousand Oaks, CA: Sage.
Miles, M., & Huberman, M. (1994). Qualitative data analysis: An expanded sourcebook. (2nd ed.) Thousand Oaks, CA: Sage.
Onwuegbuzie, A. J., & Daniel, L. G. (2005). Editorial: Evidence-based guidelines for publishing articles in Research in the Schools and beyond. Research in the Schools, 12(2), 1-11. Retrieved from www.msera.org/download/Rits_editorial_12_2.pdf
Onwuegbuzie, A. J., & Leech, N. L. (2007a). A call for qualitative power analyses. Quality & Quantity: International Journal of Methodology, 41, 105-121.
Onwuegbuzie, A. J., & Leech, N. L. (2007b). Validity and qualitative research: An oxymoron? Quality & Quantity: International Journal of Methodology, 41, 233-249. doi:10.1007/s11135-006-9000-3
Onwuegbuzie, A. J., & Levin, J. R. (2005). Strategies for aggregating the statistically nonsignificant outcomes of a single study. Research in the Schools, 12(1), 10-19.
Onwuegbuzie, A. J., Levin, J. R., & Ferron, J. M. (2011). A binomial test of group
differences with correlated outcome measures. The Journal of Experimental Education, 79, 127-142. doi:10.1080/00220970903352761
Patton, M. Q. (1999). Enhancing the quality and creditability of qualitative analysis. HSR: Health Services Research, 34, 1189-1208.
Reason, P., & Bradbury, H. (2008). Introduction. In P. Reason & H. Bradbury (Eds.), Sage handbook of action research (pp. 1-10). Thousand Oaks, CA: Sage.
Reason, P., & Torbert, W.R., (2001). The action turn: Towards a transformative social science. Concepts and Transformation, 6(1), 1-37.
Richards, J. C., & Schmidt, R. W. (2002). Longman dictionary of language teaching and applied linguistics. London, England: Longman.
Sayer, A. (1992). Method in social science: A realist approach, London, England: Routledge.
Schwandt, T. (1994). Constructivist, interpretivist approaches to human inquiry. In N. K. Denzin, & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 118-137). Thousand Oaks: Sage.
Schwandt, T. A. (2001). Dictionary of qualitative inquiry (2nd ed.). Thousand Oaks, CA: Sage.
Tashakkori, A., & Creswell, J. (2008). Envisioning the future stewards of the social-behavioral research enterprise. Journal of Mixed Methods Research, 2, 291-295. doi:10.1177/1558689808322946
Tashakkori, A., & Teddlie, C. (1998). Mixed methodology: Combining the qualitative and quantitative approaches. Thousand Oaks, CA: Sage.
Teddlie, C., & Tashakkori, A. (2009). Foundations of mixed methods research: Integrating quantitative and qualitative approaches in the social and behavioral sciences. Thousand Oaks, CA: Sage.
Yin, R. (2003). Case study research: Design and methods (3rd ed.). Thousand Oaks, CA: Sage.
Copyright 2012 by the RESEARCH IN THE SCHOOLS Mid-South Educational Research Association 2012, Vol.19, No. 2, 75-89
Fall 2012 75 RESEARCH IN THE SCHOOLS
Unpacking the Effects: Identifying School and Teacher Factors and Their Influence on Teachers’ Intentions to Stay or Leave the Profession
Amy L. Sedivy-Benton University of Arkansas at Little Rock
Carrie J. Boden-McGill Texas State University
As we move full speed into the 21st Century, many things are changing, ranging from how we communicate and work to how we learn (Schleicher, 2010). These changes are major factors in how the United States will have to shape and adjust in order to compete in the global marketplace. One reoccurring issue that has not seemed to cease in the last decade is the ever-growing focus on the condition of education in the United States (Aud et al., 2012). As the globalization of jobs, politics, and economics become commonplace in the everyday lives of individuals in the United States, schools, teachers, and students are compared to their counterparts across the world. On a consistent basis, the United States scores at the bottom for both student achievement and teacher status (Organisation for Economic Co-operation and Development, 2012). Implementation of change in the United States school system tends to be incredibly slow, and the results from these actions cannot realistically be seen in the near future, as students need to
matriculate for results to be measured. For example, A Nation at Risk was published in 1983 by National Commission on Excellence in Education, and no substantial changes were noted in education until the implementation of No Child Left Behind in 2001 (U.S. Department of Education, 2005). This lack of change could be because schools needed time to determine what was not working and how to adjust. However, the expectation of the federal government seems to be contrary, and results are expected in a much shorter timeframe as further mandates are being passed in a shorter period. It has taken less than 10 years for the federal government to implement Race to the Top as an attempt to increase the standing and academic achievement of United States schools (U.S. Department of Education, 2010).
Since the passage of the No Child Left Behind Act of 2001 (U.S. Department of Education, 2010), which was also revisited again in 2010, districts, schools, and teachers have keenly focused on the performance of students and the movement and gains that have been made regarding test scores. The largest push in this movement has been to focus on the accountability of individuals involved with education. The U. S. Department of Education has just issued its fourth round of Teacher Incentive Fund grants. A focus of these
Correspondence concerning this article should be addressed to Amy L. Sedivy-Benton, 2801 South University Avenue DKSN 314, Little Rock, AR 72204. Email: [email protected]
Teacher turnover is costly in its financial implications and negative impact on student learning. Documented in many recent studies many teachers choose to transfer to a more preferable work setting or leave the field. For this study, the researchers conducted an analysis of the most recent School and Staffing Survey (SASS) data from the National Center for Education Statistics. Analysis methods entailed implementing a nested design using Hierarchal Linear Modeling and yielded results that indicated school level contextual factors such as teacher influence on school, teacher perception of control, and teacher perceived support played a significant role in teachers’ intentions to leave or remain in the field. An understanding of these factors allows policymakers and administrators to implement practices to improve work environments for teachers. Retaining good teachers is a key to improve teacher and school quality nationwide.
UNPACKING THE EFFECTS: IDENTIFYING SCHOOL AND TEACHER FACTORS AND THEIR INFLUENCE ON TEACHERS’ INTENTIONS TO STAY OR LEAVE THE PROFESSION
Fall 2012 76 RESEARCH IN THE SCHOOLS
grants is to entice teachers with pay based upon their performance, or rather that of their students; the notion of paying for performance is most commonly called performance-based compensation. The current iteration of education reform, Our Future, Our Teachers: The Obama Administration’s Plan for Teacher Education Reform and Improvement, focuses largely on accountability measures (U.S. Department of Education, 2011). Characteristics of this plan, such as using performance based-measures to indicate improvement, as well as how federal funds will be appropriated to schools, like the NCLB Act before it, may do little to improve working conditions for teachers. This initiative remains focused on student achievement rather than on the environment in which teachers work and on efforts to retain those teachers who are effective.
Much discussion exists around student learning and student improvement, and student learning is certainly a crucial issue. However, the focus should be on the learning environment and teacher retention. Increased focus on accountability might not yield a working environment conducive to retaining teachers and fostering commitment to teaching (Feng, 2009). To address how these issues impact teachers’ intention to remain in the profession, we examined the data from the most recent complete Schools and Staffing Survey 2007-2008 (SASS) provided by the National Center for Education Statistics (NCES). These data were analyzed by using a nested design with Hierarchal Linear Modeling (HLM) to determine if contextual factors in the work environment of teachers play a significant role in teachers’ intentions to leave or to remain in the field. What we discovered is that environmental factors are strong when a teacher is deciding whether to remain or to leave the profession.
Thus, despite the current interest in teacher quality and supply, some clear gaps are evident in the published literature as indicated in the meta-analysis conducted by Borman and Dowling in 2008. Studies of teacher attrition seem to lack at least one of the following characteristics: use of a nationally representative sample or explicit linkages to a conceptual or theoretical framework within which to situate data analysis. Isolated in these works are the teacher individual characteristics from the context of the organizations in which they work, rather than taking them into consideration jointly. Singer and Willett (1994) proposed a plan for the NCES that would allow a true longitudinal study to be conducted on teachers and the directions that their careers are going. However, until a dataset like that proposed by Singer and Willett is executed, it is necessary to use the best and most comprehensive data about teachers available, the SASS dataset.
Building on previous research, we explored the contextual factors in the work environment as well as teacher characteristics that influence teachers’ intentions to remain in or to leave the profession. Rather than focus on factors that policymakers and administrators cannot control or that might take a long time to address, we elected to focus on those factors in the teaching environment that administrators might influence in a relatively short period of time. Using the most recent complete data set from SASS by NCES regarding the teachers and their perceptions, we discovered that teacher influence on school, teacher perception of control, and teacher’s perceived support, are factors in teachers’ intentions to leave or to remain in the field. In the discussion that follows, we briefly review prior research used to structure our study and present the methods, data, and results. The final section includes a discussion of the implications of the results, limitations of the study, and possibilities for future research.
Review of the Relevant Literature
Frequent and perpetual teacher turnover is
costly. When teacher turnover is continuous, the financial consequences are potentially dire for the school, district, and nation. Additionally, teacher turnover has a negative impact on student learning and the environmental conditions at the school. According to the National Commission on Teaching and America’s Future (NCTAF) (2003), estimates are that it costs between $4,300 in a small rural district to almost $20,000 in a large urban district for each teacher not retained (NCTAF, 2007; Shakrani, 2008). Regarding teacher turnover, 84% is due to teachers transferring between schools or leaving the profession (Alliance for Excellent Education, 2006), and the annual national cost of teacher turnover in public schools is more than several billion dollars a year (NCTAF, 2007). Virtually everyone in the educational system, including students, parents, and taxpayers, are affected by teacher turnover. This turnover makes it difficult to measure intangible effects of teacher turnover on a school, such as loss of morale, cohesion, and instructional quality in schools (Boyd et al., 2011; Luther & Richman, 2009).
Leading researchers, such as Ingersoll, Cochran-Smith, and Darling-Hammond, have conducted studies on teacher retention and attrition and have documented current trends: often, when given the opportunity, many teachers choose to move to a more preferable school or leave the profession (Boyd, Lankford, Loeb, & Wyckoff, 2005; Hanushek & Rivkin, 2004; Ingersoll, 2001, 2002; Ingersoll & Smith, 2003; Scafidi, Sjoquist, & Stinebrickner, 2007). Keigher (2010) indicated an increase from 6.6% attrition in the mid-1990s to
AMY L. SEDIVY-BENTON AND CARRIE J. BODEN-MCGILL
Fall 2012 77 RESEARCH IN THE SCHOOLS
8% in 2008-2009 (O’Donovan, 2011; Snyder & Dillow, 2011). Individuals who chose teaching as a career often leave in the first years of teaching (Ingersoll, 2001), and teachers with more education (e.g., a master’s degree) are more likely to leave the profession than are their less-educated counterparts (Grissmer & Kirby, 1997; Scafidi et al., 2007). The problem is not a shortage of teachers; rather, the issue is keeping teachers in the field (Darling-Hammond, 2003; Ingersoll & Smith, 2004). According to the U.S. Department of Education’s examination of departures, in 2003-2004, 56% of those teachers who left teaching cited job dissatisfaction as the reason and indicated a desire to find an entirely new career (U.S. Department of Education, National Center for Educational Statistics, 2004). Results were that in this country, almost four million individuals could teach based on their training but choose not to do so (Curran, Abrahams, & Manual, 2000). The environment that these teachers work in could potentially explain this phenomenon. According to Farber (2012):
We have an anti-teacher climate that has only worsened since 2010…The situation has only gotten worse, with layoffs, pay cuts, anti-union sentiment, program cuts and strict mandates that are part of federal education laws. If we are to make any reform or new initiative work in education, we have to create schools that are supportive, humane, dynamic, and creative. (para 4) Several researchers identified working
conditions to be a significant factor in teachers’ decisions to leave the field (Bronfenbrenner, 1976; Donaldson & Johnson, 2010; Hanushek & Rivkin, 2004). A myriad of factors contributing to teachers’ decisions to leave include too little support for teachers, student discipline problems, low funding for schools and low pay for teachers, lack of teacher influence, low respect for the profession, teacher self-efficacy, job stress, staff relations, effective school leadership, and safety (Darling-Hammond, 2003; Hirsch & Emerick, 2007; Ingersoll, 2007; Klassen & Chiu, 2011; Kopkowski, 2008; Loeb, Darling-Hammond, & Luczak, 2005). Working conditions include administrative support, integrity of the classroom, resources for teaching, and the amount of input that teachers feel they have when making decisions (Ingersoll, 2001, 2002). Workplace conditions also include items such as class size, student achievement, discipline problems, and parental involvement (Bronfenbrenner, 1976). Administrative support, in particular, can greatly affect the rate of teacher turnover in a school setting (Greiner & Smith, 2009; McLaughlin, Pfiefer, Swanson-Owens, & Yee, 1986). Ferriter and Norton (2004) pointed out that “administrators have a great deal of influence over school climate
and teacher efficacy,” (p. 19) and the most effective administrators are supportive. Darling-Hammond (2003) added that school leadership may have a magnetic effect, and teachers will search for and stay in the work environments where they are supported and are most able to excel. Conversely, McLaughlin et al. (1986) established that often teachers feel that they have inadequate opportunities for personal success, and a lack of administrative support exists. This lack of support may negatively influence a teacher’s intention to stay in or leave the teaching profession.
Little and McLaughlin (1993) examined a mathematics collaborative as well as other subjects and identified factors that primarily focus on support and collaboration within the school with teachers and their peers. Based upon interviews and surveys, they noted that teachers work better with support, and they need to feel that they fit into the organization. Similarly, Kalleberg and Mastekaasa (1998) discovered that employees who felt to be “part” of the organization were more likely to remain. Not only does the feeling of involvement influence the teacher’s career decision, it can also have positive impact on student achievement. Teachers who are more vested and involved with their schools tend to pass this investment on to the students (Huberman, 1989, 1995). When teachers feel a part of the process, they are often more willing to stay.
Researchers also have provided evidence that student characteristics can influence teachers’ decisions (Boe, Bobbitt, Cook, Barkanic, & Maislin, 1998; Fowler, 2004; Grissmer & Kirby, 1997; Sclan, 1993). Indicated in these works was that teachers who leave the profession are working in schools that have high proportions of low-achieving, low-income students. Loeb et al. (2005) indicated that a school’s racial make-up as well as low-income student population could predict teacher turnover.
National and local personnel policies can contribute to teachers’ decisions. In 2001, the implementation of the NCLB Act held states accountable for student outcomes. No Child Left Behind became a federal law, reauthorized in 2010, with revisions to adjust for shortcomings in the original plan. However, the focus remains on accountability. In 2009, the Obama Administration initiated Race to the Top; this $4.35 billion fund is in place to serve as an initiative for states to use innovative ways to show significant improvements in student outcomes. In addition to the race to the top, the Elementary and Secondary Education Act (ESEA) also offered school districts Teacher Incentive Fund (TIF) grants for implementing performance based programs for high need schools (U.S. Department of Education, 2009; 2012). With the implementation of NCLB, the number of alternatively certified teachers
UNPACKING THE EFFECTS: IDENTIFYING SCHOOL AND TEACHER FACTORS AND THEIR INFLUENCE ON TEACHERS’ INTENTIONS TO STAY OR LEAVE THE PROFESSION
Fall 2012 78 RESEARCH IN THE SCHOOLS
entering the classroom has risen. The NCLB requirement of highly qualified indicates that teachers must hold a degree, pass a required state- mandated exam, as well as teach in their subject area (U.S. Department of Education, 2005). Retention Strategies With the current climate of the 2012 Presidential administration, an intense focus has been placed on the premise that money will be a significant factor in the improvement of student achievement. The thinking seems to be that with TIF incentives, teachers will stay in the profession longer. However, alluded to in the literature was that incentive pay does not influence the decision to stay in the profession. Rather, salary is significant only when it is not attached to any type of student achievement. The monetary factor seems to be more important during the beginning of their teaching career (Brewer, 1996; Darling-Hammond, 2003; Hanushek, Kain, & Rivkin, 1999), and researchers indicated that when teachers are paid a lower salary, they are more likely to move on to opportunities that provide more from a monetary standpoint (Fresko & Alhija, 2009; Hung & Smith, 2008). In addition to researchers documenting the effects of teacher salary, authors of other studies primarily focus on monetary policies and methods for teacher retention (Clotfelter, Glennie, Ladd & Vigdor, 2006).
A second, less-than-effective retention strategy has been to offer bonuses to teachers to work in poorly performing schools. In South Carolina, districts offered teachers an $18,000 bonus to work in the weakest schools. Only 20% of the teachers who were offered the bonus lasted for the length of the contact (Kopkowski, 2008). Similarly, in Massachusetts, districts offered teachers a $20,000 bonus over a 4-year period. Most teachers decided that the net extra income was not worth moving from a preferred school to a less desirable one (Kopkowski, 2008). According to the Education Commission of the States, other retention strategies include scholarship programs, loan forgiveness programs, tax credit or mortgage assistance, and stipends (Glennie, Coble, & Allen, 2004). Unfortunately, few of the programs have been effective in retaining teachers (Hammer, Hughes, McClure, Reeves, & Salgado, 2005). It appears in the case of teacher retention, environmental factors outweigh the incentive of extra income. Conceptual Model
The conceptual model of teacher retention and attrition used in this research is grounded in Bronfenbrenner’s (1979) Ecological Theory. The model follows common characteristics that have been associated with teachers leaving the field at both the teacher and organizational levels.
Bronfenbrenner suggested that individuals are not only affected by their immediate surroundings, but also their mesosystem, connections with their microsystem such as other teachers working in school, and the exosystem in which the teacher does not function directly, such as the federal government. In the past, Darling-Hammond (2000, 2003) and others have primarily focused only on the contribution of individual teacher characteristics and how they contribute to attrition. In turn, little information exists about how teachers and their workplaces interact and contribute to their decision to leave the classroom. The organizational factors or school characteristics need to be considered because they encompass the school level at which the teacher works, as does locale, poverty level, and control and influence that teachers have within the school.
Teachers in various districts might perform similar work; however, they do this work in different workplaces. These schools vary in both location and student demographics, and the schools also differ in what types of management or administration that are present. Based upon this framework and indication that schools are different from one another, examined in this research were both the teachers and their environments. By studying the teachers and their environments, we considered how the differences of teachers within a school, the environments across schools, and individual and environmental factors interact to affect a teacher’s intention to remain in the profession. Research Question
The question guiding this research was: Which individual and environmental factors most heavily influence teachers’ intentions to remain in or leave the profession? We addressed this question by employing HLM and allowing for the separation of individual and environmental factors.
Method
Research Design
To examine the factors in the work environment that influence teachers’ intentions to remain in or to leave the profession, we examined the most recent dataset from 2007-2008 SASS by the NCES. This comprehensive dataset contains information regarding teachers and their perceptions. We employed a hierarchical linear model in which two levels of data were prepared: level-1 contained individual information about the teacher, characteristics, demographics and their perceptions of the work environment and level-2 contained information about the characteristics of where they work, such as autonomy and control. This nested design allows for the variance between and within school to be extracted. The
AMY L. SEDIVY-BENTON AND CARRIE J. BODEN-MCGILL
Fall 2012 79 RESEARCH IN THE SCHOOLS
combination of these variables described the variation between teachers and their choices. Teachers’ intention to remain in the profession was used as the dependent variable; the continuous nature of this variable allows a range to be established and allowed for better understanding of the impact of the independent variables. Equations presented represent hypothesized relationships between the predictive variables, such as contextual work variables–influence, support, and control–on the outcome of a teacher’s intention to leave the profession. Data
The Schools and Staffing Survey restricted dataset was secured from the NCES to allow us to conduct this analysis. This dataset is the largest, most extensive survey of K-12 school districts, schools, teachers, and administrators in the United States today. Each cycle of SASS administers survey questionnaires to a random sample of approximately 53,000 teachers, 12,000 principals, and 4,500 districts, representing all types of teachers, schools, and districts in all 50 states. These data, nationally representative at the school level, are based on a complex survey design that was administered by the NCES to national probability samples. Schools are the primary sampling unit in SASS. The public school sampling frame was based on Common Core of Data, a file of information collected annually by NCES from all state education agencies and believed to be the most complete public school listing available at the time of sample selection. Criteria for inclusion in the survey at the school level were set at least three teachers per school in order to protect the participants. During the 2007-2008 school year, close to 51,000 public school teachers completed the survey. These data contain a substantive subsection targeting those individuals who are involved with any school-level aspect of education, such as principals and teachers. The dataset includes training experience, mentoring programs, school structure, and subject taught for participants. The teaching-focused sections of the survey include questions about employment history, job changes, and school climate. For the purposes of this study, only those teachers who indicated they would either remain or leave the profession were kept in the sample. Teachers who were undecided or did not provide sufficient data were removed. To adhere to the SASS definition of a teacher, this sample includes any individual who was a full-time K-12 teacher. Part-time, long- and short-term substitutes, student teachers, teacher aids, and non-teaching employees were not included. Based on these criteria, a sample of 20,324 public school teachers were identified in this dataset and included in this analysis.
Teachers’ perception of workplace: Created variables. The SASS survey contains several items that are directed at the amount of involvement that a teacher has within the school, how much involvement a teacher perceives s/he has within the school, and the teacher’s perception of support. Items that addressed this issue were plentiful, and 20 items addressed how much control, support, or influence teachers had within their schools. These specific items were removed from the survey because they indicated how the teachers viewed their workplace environments. It would be inappropriate to leave all of these items as individual variables as they would provide inaccurate results due to multicollinearity. The items that are directed at these issues were analyzed via a factor analysis with a Primary Components Factoring. This method was used because it reflects both the common and unique variance of the variables. A Varimax rotation was used because results make it as easy as possible to identify each variable with a single factor (Gorsuch, 1983). The factor scores, as presented in Table 1, then were taken from the factor analysis and used to create a composite variable for teacher influence within the school as perceived by the teachers and a composite variable influence and support as perceived by the teachers.
This factor analysis on the teacher survey variables yielded three factors. Questions that were reported by the teacher regarding influence clustered together under Factor 1, which was renamed teacher influence on school. Factor 2 displayed factor coefficients of items pertaining to teachers’ perception of control and was renamed teacher perception of control. For Factor 3, items regarding the teachers’ perception of management and support clustered together, and this factor was renamed teacher’s perceived support. The factor scores from these items then were used to create a composite variable for each of the factors, which in turn were used in the analysis. These newly created composite variables were tested for score reliability. The newly created variables produced high reliability scores, including α=.84 for teacher influence on school, α=.76 for teacher perception of control, and α=.76 for teacher’s perceived support. Because these newly created variables provided high reliability scores, these variables were used in the analysis. Data Analysis
A hierarchical linear regression model was used on the SASS data to examine teachers’ intentions to leave the profession. Individual characteristics that were included in level-1 as control variables were: gender, race/ethnicity, new to teaching, highest level of education, and salary. To explore the variables that contribute to a teacher’s intention to leave the profession due to
UNPACKING THE EFFECTS: IDENTIFYING SCHOOL AND TEACHER FACTORS AND THEIR INFLUENCE ON TEACHERS’ INTENTIONS TO STAY OR LEAVE THE PROFESSION
Fall 2012 80 RESEARCH IN THE SCHOOLS
contextual factors or workplace conditions, level-2 variables were included. Level-2 variables consist of: school level, locale, poverty level, violence in the school, teacher’s perception for and control in the classroom, teacher’s perception of support in
the school, teacher’s perception of influence at school, and the perception of control that teachers feel they have in their working environment.
Table 1 Rotated Component Factor Matrix for School Influence/Teacher Perceptions Component
Factor 1 Teacher Influence at
School
Factor 2 Teacher Perception of
Control
Factor 3 Teacher Perceived
Support Influence-Performance Standards .625 .429 .320
Influence-Curriculum .596 .419 .066
Influence-School budget .531 .325 .058
Influence-Discipline .686 .487 .182
Influence-Teacher Hiring .568 .268 .087
Influence-Teacher Evaluation .571 .054 .207
Influence-Prof Dev Content .605 .105 .115
Control-Selecting Materials .111 .734 .328
Control-Selecting Content .095 .664 .296
Control-Selecting Techniques .158 .5657 .142
Control-Evaluating and Grading .286 .739 .144
Control -Disciplining Students .113 .642 .238
Control-Homework to be Assigned .086 .570 .201
Teacher- Staff Expectations .240 .394 .731
Teacher-Admin Supportive and Encouraging
.403 .143 .867
Teacher-Principal Support .184 .073 .723
Note: Rotated Factor Matrix Extraction Method: Principal Component Factoring. Three factors extracted. Five iterations required.
AMY L. SEDIVY-BENTON AND CARRIE J. BODEN-MCGILL
Fall 2012 81 RESEARCH IN THE SCHOOLS
Results Hierarchical Regression
To determine what factors contribute to teachers leaving the profession, a three-step hierarchical model was tested. The first block of variables included demographic information about the teachers such as age, gender, ethnicity, degree level, education, and years in the profession. The second block included information about their workplace, such as wage. These variables were tested at level-1 to determine statistical significance. Upon adding the third block school-level characteristics were entered at level-2. Variables in the third block included mean responses from teachers and principals on perceptions of school violence, influence on decision-making, control over the environment, effectiveness of school management, support, salary, school-level (elementary or secondary), locale, and poverty level (measured by number of students receiving free and reduced lunch). Teacher Turnover
To address if teachers will decide to leave the profession or not, the outcome variable addresses the question of how long these teachers intend to remain in teaching. When asked if they would remain in teaching, their possible response options were: 1= As long as I am able, 2 = Until I am eligible for retirement, 3 = Will probably continue unless something better comes along, 4 = Definitely plan to leave teaching as soon as I can, and 5 = Undecided at this time. Response number 5 did not provide a definitive decision; thus, teachers with this response were removed from the data. Hierarchical linear models were constructed and tested to determine the effects individual and school characteristics have on teachers leaving the profession. The sample used in these analyses comprised 20,794 full-time teachers and 5,964 schools.
The first model created in HLM contains no level-1 or level-2 predictors, commonly called the unconditional model or a one-way analysis of variance (ANOVA) with random effects because it separates variability into within-school and between-school components (Raudenbush & Byrk, 2002). This preliminary step provides information regarding whether enough variance exists to indicate the presence of statistically significant differences among schools in teachers’ perceptions and their intentions on leaving the profession. The equations for the one-way ANOVA model are:
(2.1) Level 1 Yij = β0j + rij Level 2 β0j = γ00 + u0j
In this equation, Yij is the outcome for teacher i in school j; β0j is the intercept or school mean teacher response for whether they intend to leave the teaching profession; rij is the teacher-level residual or the difference between the response of teacher i from the average response of school j; γ00 is the grand mean response for the outcome across all schools; and u0j is the level-2 residual or the difference between the mean response of school j and the grand mean (Raudenbush & Byrk, 2002). Results from this unconditional model are presented in Table 2.
For the fixed effects, the weighted least squares estimate for the grand mean of teachers choosing to leave the profession was = 1.79. The grand mean indicated, that on average, teachers responded that they intended on remaining in the teaching profession. The within-school variance was σ2 = 0.63, and the between-school variance was τ00 = 0.02. Because τ00 was statistically significant from zero, the average teacher’s intention of leaving or remaining in the profession differed from the different school means. Based on the information in Table 3, this null model shows that the estimate for between-school differences on the outcome is statistically different from zero (χ2 = 6584.13, p < 0.001). Thus, schools differed in their average teacher response for their intention to leave the profession.
The plausible range of values for the average teacher intention of leaving the profession, at the 95% confidence interval, was 1.51 to 2.08. On the scale, a response of “1” represented the teacher’s intention to stay in the profession as long as possible, whereas a response of “4” represented the teacher’s intention to leave the field as soon as possible. Consequently, on average, the teachers indicated that they would remain in the profession.
The interclass correlation represents the proportion of variance in the outcome that is between schools (Raudenbush & Byrk, 2002). The equation is:
(3.1)
(3.2) This analysis produced an intraclass
correlation coefficient of .032, which indicated that 3% of the variance in teacher’s responses was between-schools and 97% was within-schools. From this outcome, it can be inferred that much more within school variation was present than between school variation on a teacher’s intention to leave the profession.
ρ = τ 0 0 τ 0 0 + σ 2
ρ = . 0 2 1 . 0 2 1 + . 6 2 4
= . 0 3 2
UNPACKING THE EFFECTS: IDENTIFYING SCHOOL AND TEACHER FACTORS AND THEIR INFLUENCE ON TEACHERS’ INTENTIONS TO STAY OR LEAVE THE PROFESSION
Fall 2012 82 RESEARCH IN THE SCHOOLS
Table 2 Results from the One-Way ANOVA Model Fixed Effect Coefficient SE Average school mean, γ00 1.797 .005 Random Effect
Variance Component df χ2 p-value
School mean, u0j
.021 5963 6584.13 < .0001
Level-1 effect, rij .623 Table 3 Effects of Personal Characteristics on a Teachers Decision to Remain in the Profession
Model 0 Model 1 Model 2
Constant 1.797 (p < .0001) 1.776 (p < .0001) 1.816 (p < .0001)
Teacher Demographics
Male 0.101* 0.104*
New teacher 0.109* 0.139*
Black 0.004 0.008
American Indian -0.028 -0.022
Asian -0.009 -0.020
Hispanic -0.025 -0.034
Teacher Education
Alt Cert -0.018
Degree 0.019
Earnings -0.003*
Variance Component
σ2
.623 τ2
.021
σ2
.622 τ2
.019
σ2
.605 τ2
.028
* statistically significant at p < .05
AMY L. SEDIVY-BENTON AND CARRIE J. BODEN-MCGILL
Fall 2012 83 RESEARCH IN THE SCHOOLS
Due to the statistical significant variation both
within and between schools, variables were entered at level-1 and level-2. The null model showed a large percentage of the variation lies within the schools. To determine which teacher variables contribute statistically significantly to a teacher’s intention to leave the profession, teacher variables were entered at level-1. Proceeding with the analysis, variables were added in blocks starting with personal characteristics of the teachers and then education of the teachers. Variables that were not statistically significant were deleted from the model. This resulting model was:
(4.1) Level-1 Yij = β0j + β1jX1ij+ β2jX2ij +…..+βQjXQij+rij (4.2) Level-2 βqj=γq0+γq1W1j+γq2W2j+ ….γqSqWSqj+uqj
In Model 1, teacher characteristics were
added to the unconditional model to determine if a decrease was present in the variance, σ2, at level-1, the teacher level. The proportion of variance explained at level-1 was found by using the following formula:
(5.1) (5.2) = .04
Table 3 presents the results of Model 1.
Gender and teaching for less than 3 years accounted for 4.7% of the variance in teachers’ intentions of leaving the profession at the teacher level, leaving 95.3% of the within-school variance as unexplained. This model also suggests that males’ intention of leaving the profession was greater than females’ intentions, and the intentions of staying in the profession of new teachers were greater than were those of teachers who had been in the profession for longer than 3 years.
In Model 2, teacher characteristics also were combined with teacher education. Male teachers had a greater intention of leaving the profession than did females, and new teachers were less likely to remain in the profession than were teachers who had been teaching for longer than 3 years. This finding was consistent with Darling-Hammond (2000), who concluded that beginning teachers leave the profession at a higher rate than do those teachers who have been teaching for more than 5 years. Another finding was that teachers who earned a higher salary had greater intentions of remaining in the profession. By including these
additional teacher education characteristics to the model, the variance accounted for within-school was
(6.1) (6.2) = .028
The addition of teacher education
characteristics accounted for an additional 2.8% of the within-school variance when compared to the null model, leaving 97.2% of the within-school variation as unexplained.
Upon examining both of these models, we discovered that the race/ethnicity of a teacher was not statistically significant and thus did not impact their intentions to leave the profession. Alternative certification also was not statistically significant in this model, suggesting that teachers’ intentions to leave the profession are not influenced by their route of certification. This finding regarding alternative certification is inconsistent with the literature in which teachers who do not pursue traditional routes of certification leave the profession at a higher rate than do those teachers who receive their certification via a traditional route such as a 5-year preparation program (Darling-Hammond, 2001). Overall, each step of the model of within-school variation explained slightly more of the variance with Model 3 explaining approximately 3% more variation than the unconditional model did on its own.
All variables were fixed in level-2 except for earnings because of the lack of between school variation and the fact that these variables were dummy coded as 0 or 1. Due to the lack of significant between-school differences, teacher earnings were kept as a fixed effect in the full model.
Full Hierarchal Linear Model
The personal and educational variables that had the strongest effect on a teacher’s intention of leaving the profession at level-1 were selected to provide a preliminary model. Workplace conditions and school context variables were entered at level-2. An exploratory model was implemented that examined the variability of the teachers’ decision to leave the profession not only within the school but also between the schools. The personal and educational variables that were statistically significant in the prior models were kept at level-1, and workplace conditions and school context variables were entered at level-2. These two models are placed together to give a full HLM.
ˆ σ 2 A N O V A − ˆ σ 2 r a n d o m c o e f f i c i e n t ˆ σ 2 A N O V A
ˆ σ 2 A N O V A − ˆ σ 2 r a n d o m c o e f f i c i e n t ˆ σ 2 A N O V A
UNPACKING THE EFFECTS: IDENTIFYING SCHOOL AND TEACHER FACTORS AND THEIR INFLUENCE ON TEACHERS’ INTENTIONS TO STAY OR LEAVE THE PROFESSION
Fall 2012 84 RESEARCH IN THE SCHOOLS
The Male and New to teaching variables were entered at level-1 uncentered. The teachers’ income that they earn at the school (earnings) was group mean centered. In this full model, only the intercept of the outcome variable has random effects. Earnings were entered with random effects but it was determined to be not statistically significant. The remaining variables at level-1 were dummy coded as 0 or 1 and thus were treated as fixed.
To explain the school-level variations in the intercept slope, level-2 school characteristics were entered. These variables were selected based upon past study findings, such as Smith and Ingersoll (2004), who examined organizations and characteristics of those organizations that lead to a teacher’s decision to leave the profession. Free and reduced lunch (lunch) was entered grand mean centered; school level (level-2) was entered uncentered and Influence in school (INFSCH_1), support of management (support), student delinquency (SDLQ_T_1), salary (SALARY), influence on school (influence), and control in school (control) all were entered grand centered. The final model was as follows:
(7.1) Level-1 Model Y = β0 + β1(Male) + β2(NEWTCH) + β3(earnings) + r (7.2) Level-2 Model β0 = γ00 + γ01(Lunch) + γ02(School level) +
γ03(Student Delinquency)+γ04(Support) + γ05(School Influence) + γ06(Salary) + γ07(Influence) + γ08(Control) + u0
β1 = γ10 β2 = γ20 β3 = γ30 β4 = γ40 β5 = γ50
Table 4 contains the following teacher-level variables that were determined to be statistically significant predictors: gender, and if they were a new teacher or not. For school-level variables predicting the variation in school means on teachers’ intention to leave the profession, the school level (elementary, secondary), poverty of the school, reported teacher influence in school, mean perception of school support, mean student delinquency, mean salary, mean teacher control, and mean teacher influence statistically significantly predicted the teacher-level intercept. The proportion of variance that is explained by the model for the intercept term was determined using the formula suggested by Raudenbush and Byrk (2002):
(8.1) Proportion Variation Explained in
βq =
(8.2) Proportion Variation Explained in βo = = .42
The proportion of variance explained, in β0
was 43%. Forty three percent of the original random variation in the intercept was explained by the teacher’s perception of the school, influence, control, support, the student delinquency in the school, poverty of the school, school level, and the salary offered. These results suggest that the mean intention for all teachers in schools was related to school-level predictors. Summary of Results
It is difficult to pinpoint one factor that influences a teacher’s decision to remain in the profession; however, a few key variables present themselves as contributors to this decision. Although holding everything else constant, the following conclusions were made: Based on the analysis, the teachers in the sample seemed to have greater intentions of remaining in the profession than to leave. More than 80% of the teachers included in the sample responded that they would remain the profession for as long as they were able, or until they were eligible for retirement. Gender was a statistically significant variable (β = .084), indicating that females were more likely to intend to remain in the profession; new teachers, those teachers who have been in the field for 3 years or less, also indicated they intended to remain in the profession, as did elementary school teachers. The level of the school was statistically significant with a coefficient of β = -.074, indicating that elementary school teachers had greater intentions of remaining in the profession. In addition, salary (β = -.006) was a statistically significant indicator regarding whether or not a teacher had intentions of remaining in the profession. The more that teachers were paid, the more were their intentions to remain in the profession increased. The data also indicated that schools with a high student population that received free and reduced lunch had teachers with lower intentions of leaving the profession (β = .056).
It is worthwhile to recognize the variables that are listed above. It is equally important to recognize that all of these statistically significant variables were either those characteristics over which school administrators have no control, such as whether the teacher is new to the teaching field, or the poverty level of the school, or they are factors that will take a significant amount of
AMY L. SEDIVY-BENTON AND CARRIE J. BODEN-MCGILL
Fall 2012 85 RESEARCH IN THE SCHOOLS
planning and time to initiate a change, such as salary. However, the variables that are of interest are those items that the district and school administrators have both control over and can initiate change in a short amount of time, because that is what is desired in the current climate of education. The factors over which both school and district administrators have direct control are factors that exist in the context in which they work. When examining these contextual factors, as identified by the created variables that address teachers’ perceptions of support, influence, and
control at the school and in their classrooms, we determined that when teachers believed that they had a supportive environment in which to work, they had greater intentions of remaining in the profession (β = -.013). Similarly, when teachers’ perception of influence that they had within their schools was high (β = -.007), they were more likely to remain in the profession, and teachers who believed they had control in their schools (β = -.056) had greater intentions of remaining in the profession.
Table 4 Results from the Intercepts- and Slopes-as-Outcomes Model
Fixed Effect Coefficient SE Model for School Means, β0 Intercept, γ00 1.847 0.024 Percent of Free Lunch, γ01 -0.056* 0.025 School Level, γ02 -0.074* 0.013 Mid-size City Local, γ03 -0.022 0.027 Fringe of Large City Local, γ04 -0.029 0.026 Fringe of Mid-size City Local, γ05 -0.016 0.028 Large Town, γ06 -0.053 0.049 Small Town, γ07 -0.018 0.026 Rural, γ08 -0.010 0.026 Student Delinquency, γ09 0.018* 0.003 Salary, γ10 -0.006* 0.001 Influence in School, γ11 -0.007* 0.002 Support by Management, γ012 0.013* 0.002 Influence by Teachers, γ013 -0.022* 0.004 Control by Teachers, γ014 -0.056* 0.006 Model for Male – Choice Slope, β1 Intercept, γ10 0.084* 0.013 Model for New Teacher – Choice Slope, β2 Intercept, γ20 0.145* 0.022 Model for Earnings – Choice Slope, β3 Intercept, γ30 -0.003* .000 Intercept, γ50 -0.044* .016
Random Effect Variance Component df χ2 p-value School Mean, u0j 0.012 5950 6209.001 0.010
Level-1 Effect, rij 0.619
Discussion
These findings provide functional results when discussing potential reasons why teachers might choose to leave the profession. Not only do these results provide some basic information to policymakers, but this information is helpful to school administrators as well. When hiring, principals might more specifically consider their work environments and how well their candidates will fit within these environments to avoid potential turnover. The consideration of the work
environment can assist in their budget planning and how their funds may be used in the future. These significant contextual factors are factors over which school administrators have control, and they can initiate immediate change without having to go through many administrative channels and wait for approval to implement measures that might produce change. For example, principals in a high poverty school might want to hire new teachers and then give those teachers support and influence within the school environment. This additional level of support might not seem like a
UNPACKING THE EFFECTS: IDENTIFYING SCHOOL AND TEACHER FACTORS AND THEIR INFLUENCE ON TEACHERS’ INTENTIONS TO STAY OR LEAVE THE PROFESSION
Fall 2012 86 RESEARCH IN THE SCHOOLS
simple task, but the notion of control and support can go a long way in providing a stable teaching force in the school environment (Huang & Waxman, 2009; Johnson, 2006; Sclan, 1993).
One finding that is inconsistent with the literature is that of poverty. Cited in previous works (e.g., Ingersoll, 2001), authors have implicated poverty level of the school as a contributing factor to teachers choosing to leave the profession. In this study, the opposite effect was present; teachers in schools with a high level of poverty were more likely to stay. These inconsistencies might be because the schools that fit the poverty profile are more likely to be in urban areas. It is plausible that the teachers who work at these schools do not have the mobility that a teacher at a low poverty suburban school might have. The relationships between the teachers’ personal characteristics and the characteristics within the school context and how they interact seem to provide the most information regarding why a teacher intends to remain or to leave the profession.
Implications
One final question that remains is: what practical and policy implications can be taken from this study that would reduce the amount of turnover in teachers? The results from this analysis indicate that contributing factors associated with individual teacher information (e.g., age, gender, education) are predictive of teachers’ intentions, and these factors cannot be manipulated. However, relationships between where the teachers worked and their intention of leaving the field were influenced by the experiences that they had within the school.
The focus of these findings indicates that attention should be paid to the attitudes and the perceptions of teachers’ working environments, how teachers perceive their role in the school when it comes to influence and control, and the support that teachers receive in their teaching roles. This work establishes that the amount of control and influence that teachers have within their environments can affect their decision to stay or to leave the field (Mastekaasa, 1998; McLaughlin et al., 1986). Our research also supports the argument that schools potentially can reduce teachers leaving the profession by allowing them maximum involvement in school decisions as well as by providing them with some control over their classrooms and curricula. These findings also are relevant to federal mandates that are currently on the rise, because these mandates primarily focus on student scores and financial rewards rather than focusing on those factors that will keep good teachers in the field, such as their workplace environment.
Limitations and Suggestions for Future Research
In our study, we made an effort to create a
research design that allowed us to capture the complex dynamics regarding reasons that teachers leave the profession. However, several limitations were present. First, it is important to note that throughout this study, teacher intentions of leaving the profession rather than the actual act of leaving the profession were discussed. Thus, it should be stated that intentions of leaving the profession might not necessarily be inclusive of those teachers who actually leave the profession, and direct inference is not warranted. Second, because differences were present among the schools in which these teachers teach and this influences their level of commitment to the field, it would be worthwhile to investigate what these schools are doing differently. The variables in this study could be more precisely defined and measured.
Conclusion
The focus on the status of education in the
United States is not something that will be going to the wayside anytime soon. The NCLB Act and Race to the Top (U.S. Department of Education, 2012) focus more on monetary rewards, and less focus is on those environments in which teachers work. Meaningful and lasting educational reform is a lengthy process, and many factors that lead to teachers leaving the field are out of the immediate control of policymakers and school administrators, who struggle with abysmal state budgets, policy implementation lag, and other constraints. It is unlikely that factors such as facilities and pay will be improved in the near future. Consequently, rather than focusing solely on student learning, policymakers and administrators also need to consider contextual factors to improve teachers’ work environments. Creating an environment that supports teachers and allows a commitment to teaching is crucial to student success (Feng, 2009).
With tight budgets developing and the exorbitant cost of teacher turnover, this research provides policymakers with a better understanding of the environmental factors contributing to the loss of teachers. Armed with the understanding that teacher influence on school, perception of control, and perceived support are vital factors to retention, policymakers and administrators may develop practices to increase teacher retention, improve teacher quality and, thus, improve the quality of our nation’s schools The lead editors for this article were John R. Slate and Anthony J. Onwuegbuzie.
AMY L. SEDIVY-BENTON AND CARRIE J. BODEN-MCGILL
Fall 2012 87 RESEARCH IN THE SCHOOLS
References Alliance for Excellent Education. (2006). Tapping
the potential: Retaining and developing high-quality new teachers. Washington, DC: Author. Retrieved from www.all4ed.org/publications/
Aud, S., Hussar, W., Johnson, F., Kena, G., Roth, E., Manning, E., Wang, X., & Zhang, J. (2012). The Condition of Education 2012 (NCES 2012-045). U.S. Department of Education, National Center for Education Statistics. Washington, DC. Retrieved from http://nces.ed.gov/pubsearch
Boe, E., Bobbitt, S., Cook, L., Barkanic, G., & Maislin, G. (1998). Teacher turnover in eight cognate areas: National trends and predictors. Philadelphia, PA: University of Pennsylvania, Center for Research and Evaluation in Social Policy.
Borman, G. D., & Dowling, N. M. (2008). Teacher attrition and retention: A meta-analytic and narrative review of the research. Review of Educational Research, 78, 367-409. doi:10.3102/0034654308321455
Boyd, D., Grossman, P., Ing, M., Lankford, H., Loeb, S., & Wyckoff, J. (2011). The influence of school administrators on teacher retention decisions. American Educational Research Journal, 48, 303-333. doi:10.3102/0002831210380788
Boyd, D., Lankford, H., Loeb, S., & Wyckoff, J. (2005). Explaining the short careers of high-achieving teachers in school with low-performing students. The American Economic Review, 95, 166-171. doi:10.1257/000282805774669628
Brewer, D. J. (1996). Career paths and quit decisions: Evidence from teaching. Journal of Labor Economics, 14, 313-339. doi:10.1086/209813
Bronfenbrenner, U. (1976). The experimental ecology of education. Educational Researcher, 5(9), 5-15. doi:10.2307/1174755
Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press.
Clotfelter, C. T. Glennie, E. , Ladd, H. F.& Vigdor, J. L. (2006) Would higher salaries keep teachers in high-poverty schools? Evidence from a policy intervention in North Carolina. (NBER Working Paper No. 12285). Cambridge, MA National Bureau of Economic Research.
Curran, B., Abrahams, C., & Manual, J. (2000, January 25). Teacher supply and demand: Is there a shortage? Washington, DC: National Governor’s Association, Center
for Best Practices, Education Policy Studies Division.
Darling-Hammond, L. (2000). Futures of teaching in American education. Journal of Educational Change, 1, 353-373.
Darling-Hammond, L. (2001). Standard setting in teaching: Changes in licensing, certification, and assessment. In V. Richardson (Ed.), Handbook of research on teaching (4th ed., pp. 751-776). Washington, DC: American Educational Research Association.
Darling-Hammond, L. (2003). Keeping good teachers: Why it matters what leaders can do. Educational Leadership, 60(8), 6-13.
Donaldson, M., & Johnson, S. M. (2010). The price of misassignment: The role of teaching assignments in Teach for America teachers’ exit from low-income schools and the teaching profession. Educational Evaluation and Policy Analysis, 32, 299-323. doi:10.3102/0162373710367680
Farber, K. (2012). My view: Six ways to retain great teachers. Retrieved from http://schoolsofthought.blogs.cnn.com/2012/06/19/my-view-six-ways-to-retain-great-teachers/?iphoneemail
Feng, L. (2009). Opportunity wages, classroom characteristics, and teacher mobility. Southern Economic Journal, 75, 1165-1190.
Ferriter, W., & Norton, J. (2004). Creating a culture of excellence. Threshold, 2(1), 18-21.
Fowler, W. J., Jr. (Ed.). (2004). Developments in school finance, 2003: Fiscal proceeding from the Annual State Data Conference of July 2003. (Report No. NCES 2004-325). Washington, DC: U.S. Department of Education, National Center for Education Statistics.
Fresko, B., & Alhija, F. (2009). When intentions and reality clash: Inherent implementation difficulties of an induction program for new teachers. Teaching and Teacher Education, 25, 278-284. doi:10.1016/j.tate.2008.12.001
Glennie, E., Coble, C. R., & Allen, M. (2004, November). Teacher perceptions of the work environment in hard-to-staff schools. Denver, CO: Education Commission of the States. Retrieved from http://www.ecs.org/clearinghouse/55/87/5587.doc
Greiner, C. S., & Smith, B. (2009). Analyses of selected specific variables and teacher attrition. Education, 129, 579-583.
Grissmer, D., & Kirby, S. N. (1997). Teacher turnover and teacher quality. Teachers College Record, 99(1), 45-56.
UNPACKING THE EFFECTS: IDENTIFYING SCHOOL AND TEACHER FACTORS AND THEIR INFLUENCE ON TEACHERS’ INTENTIONS TO STAY OR LEAVE THE PROFESSION
Fall 2012 88 RESEARCH IN THE SCHOOLS
Gorsuch, R. L. (1983). Factor analysis. Hillsdale, NJ: Lawrence Erlbaum.
Hammer, P. C., Hughes, G., McClure, C., Reeves, C., & Salgado, D. (2005). Rural teacher recruitment and retention practices: A review of the research literature. National Survey of Rural Superintendents and Case Studies of Programs in Virginia. Edvantia, VA: Appalachia Educational Laboratory.
Hanushek, E. A., & Rivkin, S. G. (2004). How to improve the supply of high quality teachers. Brookings Papers on Educational Policy 2004 (pp. 7-44). Washington, DC: Brookings Institution Press.
Hanushek, E. A., Kain, J. F., & Rivkin, S. G. (1999). Do higher salaries buy better teachers? (Working paper #7082). Cambridge, MA: National Bureau of Economic Research.
Hirsch, E., & Emerick, S. (2007). Teacher working conditions are student learning conditions: A report on the 2006 North Carolina teacher working conditions survey. Center for Teaching Quality. Retrieved from http://www.eric.ed.gov/PDFS/ED498770.pdf
Huberman, M. (1989). The professional life cycle of teachers. Teachers’ College Record, 91, 31-57.
Huberman, M. (1995). Networks that alter teaching. Teachers and Teaching, 1, 193-221. doi:10.1080/1354060950010204
Huang, S., & Waxman, H. (2009). The association of school environment to student teachers’ satisfaction and teaching. Teaching and Teacher Education, 25, 235-243. doi:10.1016/j.tate.2008.07.015
Hung, L., & Smith, C. (2008). Why do teachers leave the teaching profession?: Can alternative certificate programs eliminate the U.S. teacher shortage issue? Saarbrücken, Germany: VDM Publishing.
Ingersoll, R. M. (2001). Teacher turnover and teacher shortages: An organizational analysis. American Educational Research Journal, 38, 499-534. http://dx.doi.org/10.3102/00028312038003499
Ingersoll, R. M. (2002). Out of field teaching, educational inequity, and the organization of schools: An exploratory analysis. Seattle, WA: Center for the Study of Teaching and Policy, University of Washington.
Ingersoll, R. M. (2007). The science and mathematics teacher shortage: Fact and myth. NSTA Reports, 18(9), 6-7.
Ingersoll, R. M., & Smith T. (2004). Do teacher induction and mentoring matter? NAASP Bulletin, 88(638), 28-40. http://dx.doi.org/10.1177/019263650408863803
Ingersoll, R. M., & Smith, T. M. (2003). The wrong solution to the teacher shortage. Educational Leadership, 60(8), 30-33.
Johnson, S. M. (2006, Summer). Why new teachers stay? American Federation of Teachers, 9-45.
Kalleberg, A., & Mastekaasa, A. (1998). Organizational size, layoffs and quits in Norway. Social Forces, 76, 1243-1273. http://dx.doi.org/10.2307/3005834
Keigher, A. (2010). Teacher attrition and mobility: Results from the 2008-09 Teacher Follow-up Survey. (NCES 2010 – 353). U.S. Department of Education. Washington, DC: National Center for Education Statistics. Retrieved from http://nces.ed.gov/pubsearch
Klassen, R. M., & Chiu, M. M. (2011). The occupational commitment and intention to quit of practicing and pre-service teachers: Influence of self-efficacy, job stress, and teaching context. Contemporary Educational Psychology, 36, 114-129. doi:10.1016/j.cedpsych.2011.01.002
Kopkowski, C. (2008). Why they leave. NEA Today, 26(7), 21-25.
Little, J. W., & McLaughlin, M. W. (1993). Teachers’ work: Individuals, colleagues and contexts. New York, NY: Teachers College Press.
Loeb, S., Darling-Hammond, L., & Luczak J. (2005). How teaching conditions predict teacher turnover in California schools. Peabody Journal of Education, 80(3), 44-70. doi:10.1207/s15327930pje8003_4
Luther, V., & Richman, L. J. (2009). Teacher attrition: Listening to teachers to find a solution. Academic Leadership, 7(4), 29.
McLaughlin, M. W., Pfiefer, S., Swanson-Owens, D., & Yee, S. (1986). Why teachers won’t teach. Phi Delta Kappan, 67, 420-426.
Mastekaasa, A. (1998). Organizational size, layoffs, and quits in Norway. Social Forces, 76(4), 1243-1273.
National Commission on Excellence in Education, (1983). A Nation at Risk: The imperative for educational reform. Washington, D.C.: U.S. Department of Education.
National Commission on Teaching and America’s Future. (2003). No dream denied: A pledge to America’s children. New York, NY: Author.
National Commission on Teaching and America’s Future. (2007). The cost of teacher
AMY L. SEDIVY-BENTON AND CARRIE J. BODEN-MCGILL
Fall 2012 89 RESEARCH IN THE SCHOOLS
turnover in five school districts: A pilot study. New York, NY: Author.
O’Donovan, E. (2011). Is there a teacher shortage on the horizon? District Administration, 47(3), 74-76.
Organisation for Economic Co-operation and Development. (2012). Education at a glance 2012: OECD Indicators. Retrieved from http://dx.doi:.org/10.1787/eag-20120en
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.) Newbury Park, CA: Sage.
Scafidi, B., Sjoquist, D., & Stinebrickner, T. (2007). Race, poverty, and teacher mobility. Economics of Education Review, 26, 145-159. doi:10.1016/j.econedurev.2005.08.006
Schleicher, A. (2010). The case for 21st century learning. Retrieved from http://www.oecd.org/general/thecasefor21st-centurylearning.htm
Sclan, E. (1993). The effect of perceived workplace conditions on beginning teacher’s work commitment, career choice commitment, and planned retention. Ann Arbor, MI: University Microfilms International.
Shakrani, S. (2008). Teacher turnover: Costly crisis, solvable problem. Education Policy Center, Michigan State University. Retrieved from http://0search.ebscohost.com.iiiserver.ualr.edu/login.aspx?direct=true&db=eric&AN=ED502130&site=ehost-live&scope=site
Singer, J. D., & Willett, J. B. (1994). Modeling duration and the timing of events: Using survival analysis to long-term follow-up Studies. In S. L. Friedman & H. C. Heywood (Eds.), Developmental follow-up: Concepts, genres, domains, and methods (pp. 315-330). Orlando, FL: Academic Press.
Smith, T. M., & Ingersoll, R. M. (2004). What are the effects of induction and mentoring on beginning teacher turnover? American Educational Research Journal, 41, 681-714. http://dx.doi.org/10.3102/00028312041003681
Snyder, T. D., & Dillow, S. A. (2011). Digest of Education Statistics 2010 (NCES 2011-015). Washington, DC: National Center for Education Statistics, Institute of Education Sciences, U.S. Department of Education.
U.S. Department of Education, National Center for Education Statistics. (2004). The Condition of Education 2004 (NCES 2004–077). Washington, DC: U.S.
Government Printing Office. U.S. Department of Education. (2005). New No
Child Left Behind flexibility: Highly qualified teachers. Retrieved from http://www2.ed.gov/nclb/methods/teachers/hqtflexibility.html
U.S. Department of Education. (2009). Executive summary Race to The Top program. Retrieved from http://www2.ed.gov/programs/racetothetop/index.html
U.S. Department of Education. (2010). ESEA reauthorization: A blueprint for reform. Retrieved from http://www2.ed.gov/policy/elsec/leg/blueprint/index.html
U.S. Department of Education. (2011). Our future, our teachers: The Obama Administration’s plan for teacher education reform and improvement. Retrieved from http://www.ed.gov/sites/default/files/our-future-our-teachers.pdf
U.S. Department of Education. (2012). Teacher incentive fund. Retrieved from http://www2.ed.gov/programs/teacherincentive/index.html
MID-SOUTH EDUCATIONAL RESEARCH ASSOCIATION
The Mid-South Educational Research Association (MSERA) was founded in order to encourage quality
educational research in the Mid-South and to promote the application of quality educational research in
schools. Members of MSERA share interests in educational research, development, and evaluation.
While most members are from institutions of higher education, many others represent state departments
of education, public and private agencies, and public school systems. Graduate students comprise a
significant portion of the membership. A majority of MSERA members are from the six states
represented by the organization, but others are from many other states and several foreign countries. The
MSERA is the largest regional educational research organization in the country.
The organization provides several services for its members. The annual meeting, held every November,
offers many formal and informal opportunities for professional development through special training
courses, sharing of research findings, and programmatic interests with colleagues. Members receive a
subscription to RESEARCH IN THE SCHOOLS and the Mid-South Educational Researcher. The
MSERA also provides recognition and cash rewards for its outstanding paper, an outstanding
dissertation, and professional service.
For access to the index of articles published in Research in the Schools past issues, please
visit our website at:
http://msera.org/rits.htm
In addition, editorials are presented at the website for download and classroom use.
RESEARCH IN THE SCHOOLS Mid-South Educational Research Association
College of Education
Sam Houston State University
Box 2119
Huntsville, Texas 77341-2119
BULK RATE U.S. POSTAGE
PAID PERMIT NO. 3029
BIRMINGHAM, AL