Vol. 73, No. 4, pp. 392-416.©2007 Council for Exceptional Children.

Exceptional Children

Co- Teaching in InclusiveClassrooms: A Metasynthesisof Qualitative ResearchTHOMAS E. SCRUGGS

MARGO A. MASTROPIERIGeorge Mason University

KIMBERLY A. MCDUFFIEClemson University

ABSTRACT:T: Thirty-two qualitative investigations of co-teaching in inclusive classrooms were in-cluded in a metasynthesis employing qualitative research integration techniques. It was concludedthat co-teachers generally supported co-teaching, although a number of important needs were iden-tified, including planning time, student skill level, and training; many of these needs were linkedto administrative support. The dominant co-teaching role was found to be "one teach, one assist,"in classrooms characterized by traditional instruction, even though this method is not highly rec-ommended in the literature. The special education teacher was often observed to play a subordinaterole. Techniques often recommended for special education teachers, such as peer mediation, strategyinstruction, mnemonics, and training of study skills, self-advocacy skills, and self-monitoring, wereinfrequently observed.

I n response to recent trends and legisla- Bauwens, Hourcade, and Friend (1989); Cook

tion promoting inclusive instruction and and Friend (1995); and Friend (2002) discussedaccess to the general education curricu- criteria needed for an effective co-teaching rela-lum, many schools have implemented tionship. A number of co-teaching variations"co-teaching" (Cook & Friend, 1995) as have been identified (see also Friend & Cook,

a means for promoting effective instruction in in- 2003; Walther-Thomas, Korinek, McLaughlin, &elusive classrooms. Implemented to provide sup- wiHiams, 2000). These include:port for increasing the inclusion of students withdisabilities, co-teaching usually consists of one * One teach, one assist (or, "drift"), where onegeneral education teacher paired with one special teacher (usually, the general educationeducation teacher in an inclusive classroom of teacher) assumes teaching responsibilities,general education and special education students and the special educatiori teacher provides(e.g., Mastropieri & Scruggs, 2006, chapter 2). iridividual support as rieeded (Walther-392 Summer 2007

Thomas et al., 2000, did not mention thisvariation).

• Station teaching, where various learning sta-tions are created, and the co-teachers provideindividual support at the different stations.

• Parallel teaching, where teachers teach thesame or similar content in different class-room groupings.

• Alternative teaching, where one teacher maytake a smaller group of stiidents to a differentlocation for a limited period of time for spe-cialized instruction.

• Team teaching (or interactive teaching),where both co-teachers share teaching re-sponsibilities equally and are equally involvedin leading instructional activities.

P R E V I O U S R E V I E W S OFC O - t EACH ING

Previous reviews of co-teaching have summarizedaccumulated literature and identified importaritvariables. Friend and Reising (1993) provided anoverview of the history of co-teaching. These au-thors concluded that research was limited andmostly anecdotal; however, available evidence sug-gested that teachers believed thai: co-teaching hada positive effect on student achievement

Welch, Brownell, and Sheridan (1999) pro-vided a broader review of team teaching andschool-based problem-solving teaihs. This reviewincluded 40 articles on team teaching, of whichmany were technical reports, anecdotal reports, orposition papers. They concluded that teachers re-port positive attitudes toward various forms of co-teaching; however, there was limited knowledgeabout student outcomes, and a lack of empiricalevidence supporting co-teaching.

Weiss and Brigham (2000) reviewed 23quantitative and qualitative studies of co-teach-ing, published between 1987 and 1999, includ-ing investigations of both elementary andsecondary settings. They reported that consider-able variability was apparent in co-taught classes.However, the special education teacher typicallywas responsible for modifying instruction, behav-ior management, and monitoring studentprogress; whereas the general education teacher

was responsible for the content of instruction.Some evidence was presented that the standard ofindividualized instruction may not be met for stu-dents with disabilities. Important components ofsuccessful co-teaching experiences identified fromthis research included the general educationteacher's attitude, sufficient planning time, volun-tary participation, mutual respect, administrativesupport, and a shared philosophy of instructionand behavior management. Weiss and Brighamalso concluded that efficacy research was insuffi-cient.

Murawsld and Swanson (2001) conducted ameta-analysis of quantitative efficacy research onco-teaching. Their comprehensive search proce-dures yielded only six research reports (three jour-nal articles and three ERIC documents), whichyielded an overall effect size (standardized meandifference) of .40, from dependent measures in-cluding academic achievement:, social outcomes,attitudes, absences, and referrals. They concludedthat available research yielded moderate effects,but that the overall data set was too small to drawfirm conclusions.

Dieker and Murawski (2003) discussed co-teaching at the secondary level. They emphasizedthe importance of teacher preparation, sufficientplanning time, mastery of content by special edu-cation teachers, and pointed to large class sizesand high-stakes testing as particular challenges toco-teaching success. They recommended proac-tive communication, varied instructional practices(e.g., classwide peer tutoring), teacher training,use of a variety of co-teaching models, voluntaryparticipation, common planning periods, andfiexibility.

Weiss (2004) reviewed and updated the con-clusions of Weiss and Brigham (2000), and theresearch conducted since that time. She con-cluded that most of the studies reviewed had oc-curred in settings considered to be successful, andthat most of these studies concluded that the per-sonalities or teaching styles of the teachers wereparticularly important. She also reported that therole of the special education teacher was not al-ways clearly specified, and that outcomes of co-teaching were typically reported using vague orsubjective language. Another important issueraised by Weiss was the limited amount of effi-cacy research.

Exceptional Children 3 9 3

A number of other articles made some refer-ence to the research literature, but focused pri-marily on suggestions for teachers implementingco-teaching based on previous research and theauthors' personal experiences. Murawski andDieker (2004) provided suggestions and strategiesfor co-teaching at the secondary level. They em-phasized the importance of administrative sup-port, establishing co-teacher roles, effectiveplanning, shared classroom management, and ap-propriate assessment. Keefe, Moore, and Duff(2004) recommended that secondary co-teachersdevelop awareness of themselves, their co-teacher,their students, as well as relevant content andstrategies. They reported that research to date re-vealed that secondary teachers lacked training andskills and have more negative attitudes about co-teaching. Gately and Gately (2001) focused onimportant components of the co-teaching rela-tionship, including communication, contentknowledge, planning, classroom management,and assessment. Vaughn, Schumm, and Arguelles(1997) discussed common co-teaching issues,based on conversations with teachers. These issuesincluded "ownership" of students, classroom man-agement, space, communication, and planningtime.

Previous reviews and other relevant literaturehave generally concluded that efficacy research islimited. However, a number of variables of poten-tial significance have been identified, includingco-teacher compatibility, administrative supports,planning time, teacher training, and flexibility.

QUALITATIVE RESEARCH

Based on these previous reviews, it can be con-cluded that available efficacy data are generallypositive, but limited. In addition to importantquestions of eflicacy, however, a number of otherrelevant questions can be asked about the practiceof co-teaching. Based on considerations from pre-vious literature, these questions include the fol-lowing:

• How is co-teaching being implemented?• What are perceptions of teachers?• What problems are encountered?• What are the benefits perceived to be?

• What factors are needed to ensure success ofco-teaching?

Investigations addressing these questions aretypically qualitative in nature. Qualitative re-search is generally appropriate for describing andproviding insights about attitudes, perceptions,interactions, classroom structure, and behaviors,relevant to co-teaching. Qualitative research alsohas increased enormously in special education re-search over recent decades (Brantlinger, Jimenez,Klingner, Pugach, & Richardson, 2005; Pugach,2001; Scruggs, Mastropieri, & McDuffie, 2006).To date, a considerable amount of qualitative re-search has been conducted in the area of co-teach-ing. However, at present the research base consistsmostly of individual investigations with little pre-vious attempt to summarize or synthesize find-ings. This investigation, therefore, was intendedto systematically summarize and integrate thefindings of all available qualitative research re-ports into one integrative review. As such, it wasintended to shed light on the practice of co-teach-ing from the perspectives of relevant research. Inorder to do so, it was necessary to identify andimplement appropriate techniques for synthesis ofqualitative research.

RESEARCH SYNTHESIS

Research synthesis is an attempt to integrate sys-tematically a large body of related research litera-ture. The procedure was first applied toquantitative group-experimental research data,and referred to as meta-analysis (Glass, McGaw,& Smith, 1979). Since that time, literally thou-sands of meta-analytic investigations have beencompleted, and many of these have been appliedto special education (Forness, 2001). In additionto meta-analyses of group-experimental research,quantitative research synthesis techniques havebeen applied to single-subject research (Scruggs,Mastropieri, & Casto, 1987; Swanson & Sachse-Lee, 2000) and survey research (Scruggs & Mas-tropieri, 1996). Qualitative research synthesis hasbeen previously conducted, mostly in the healthsciences (Campbell et al., 2003; Paterson,Thorne, Canam, & Jillings, 2001), and some-times referred to as "meta-ethnography" (Noblit& Hare, 1988); "metasynthesis" (Sandelowski,

Summer 2007

Docherty, & Emden, 1997); or "metastudy" (Pa-terson et al.). Although some focused synthesiswork has been conducted in the area of educa-tional leadership and desegregation (Noblit &Hare, 1988), to date, no true integrative review ofqualitative special education research using re-search synthesis techniques has been identified.

The appropriateness and merits of qualitativemetasynthesis have been previously discussed inthe literature (see Sandelowski et al., 1997;Scruggs et al., 2006). It has been argued that thenature of qualitative research seems antithetical tosynthesis, or "summing up" (Light & Pillemer,1984), and that the original research may be dis-torted or endangered by this process. It could beargued, in fact, that it is exactly this idiographicelement that contrasts so sharply with quantita-tive studies, which offer general conclusions aboutthe behavior or performance of groups, and areless relevant to individual cases. Another concernis that summarization of research including thediversity of methodologies employed under theumbrella of "qualitative" research—including casestudies, phenomenological studies, ethnographies,semi-structured interviews, and narratives—couldtrivialize differences among them and could beproblematic in practice (Sandelowski et al.).

These concerns, however, should also beweighed against the consequences of not summa-rizing qualitative research. One problem is thatqualitative researchers often have been isolatedfrom each other, working in a "cottage industry,"to produce "one shot research" (Estabrooks, Field,& Morse, 1994, p. 510). This has limited oppor-tunity for researchers to learn from each other,and has reduced findings into "little islands ofknowledge" (Glaser & Strauss, 1971, p. 181).Without developing the connectedness latentwithin and across qualitative research studies, thisimportant body of research may exert only a lim-ited impact on policy and practice.

C O N D U C T I N G Q U A L I T A T I V EM E T A S Y N T H E S I S

Unlike quantitative synthesis (meta-analysis) ofgroup experimental research reports, qualitativemetasynthesis is not concerned with summarizingor reducing findings to a common, standardized

metric, such as a mean effect size. Rather, the pur-pose is to integrate themes and insights gainedfrom individual qualitative research into a higher-order synthesis that promotes broad understand-ings of the entire body of research, while stillrespecting the integrity of the individual reports.

Several researchers have proposed and em-ployed methods for systematically integratingqualitative research (see Scruggs et al., 2006, for adiscussion). For instance, Noblit and Hare (1988)described several ways qualitative research synthe-sis could be accomplished, including (a) "recipro-cal translation," involving recursive reading andanalysis, and comparison of metaphors used indifferent studies; (b) "refutational" meta-ethnog-raphy, investigating why researchers come to dif-ferent conclusions, such as Freeman's (1983)refutation of Margaret Mead's (1928) Coming ofAge in Samoa; and (c) "line-of-argument" synthe-sis, where studies are translated into one another,the result being a more parsimonious but encom-passing understanding of the phenomenon beingstudied. Noblit and Hate provided an example ofsuch a synthesis using five studies on racial deseg-regation. Schofield (1990) conceived of qualita-tive metasynthesis as the creation of cross-casegeneralizations based on generalizations madefrom, and about, individual cases (see also Miles& Huberman, 1994; Ragin, 1987).

Qualitative research synthesis in the healthsciences, generally using the models of Noblit andHate (1988), have been reported by Beck (2001),Campbell et al. (2003), and Jensen and Allen(1994). In the field of education, Gersten andBaker (2000) conducted a "multi-vocal synthesis"of instructional techniques for English languagelearners. This synthesis incorporated many of theanalytic principles discussed by Noblit and Hareand included intervention studies with experi-mental designs, descriptive studies of instructionalpractices, and an uncommon third source, inputfrom professional work groups.

In the present investigation, we determinedto tteat each identified research report as an indi-vidual "informant," and create a metasynthesisacross all individual research reports, using proce-dures familiar to qualitative researchers. In thisway, each author(s) is/are allowed to present origi-nal data and conclusions based on these data.That information is then integrated with the find-

Exceptional Children 3 9 S

ings of other researchers, in much the same way aqualitative researcher might use data from multi-ple informants to draw conclusions.

Considering the complexity of synthesizing alarge number of original research reports, eachcontaining its own individual data sources, weemployed NVivo software for entering text andother information, coding and categorizing quali-tative data, and assisting with organization ofqualitative data into general themes. Also lcnownas QSR NUD*IST Vivo (Fraser, 1999), NVivowas developed by Qualitative Solutions and Re-search Priority of Australia for use in qualitativeresearch procedures. NVivo was thought to beparticularly helpful in this investigation, becauseit allows a large amount of textual data to bestored and coded, and because it allows the re-searcher to reflect critically on the analysis as itunfolds, while storing individual insights thatmay be progressively refmed as more informationis added (see also Paterson et al., 2001).

In the present investigation, we determinedto treat each identified research report as anindividual "informant," and create ametasynthesis across all individual researchreports, using procedures familiar toqualitative researchers.

METHOD

SELECTION CRITERIA

This investigation gains understanding about thepractice and processes of co-teaching by synthe-sizing available qualitative research reports. Stud-ies that were included for this synthesis employedqualitative research methods as a primarymethodology, although studies were included ifthey also employed quantitative methods. Quan-titative surveys of co-teachers in which some addi-tional verbal responses were solicited (throughopen-ended or direct questions) were notincluded; however, substantive qualitative inter-views conducted subsequent to a quantitative sur-vey, and analyzed using qualitative methods, wereincluded. Studies that specifically focused on one

or more students with disabilities in an inclusiveclass, without specific reference to co-teaching asa primary research question, were not included(e.g., Zigmond, 1995; Zigmond & Baker, 1994).Reports included in this investigation had beenreported in journals, dissertations, and master'sresearch reports. Dissertations and theses were in-cluded if they met quality standards employed inthis synthesis, as discussed in a following section.

SEARCH PROCEDURES

Search procedures included the search of elec-tronic databases, including PsychlNFO, ERIC,Dissertation Abstracts, and Digital Dissertations.Descriptors employed in the searches included co-teaching, inclusion, mainstreaming, and coopera-tive teaching. We also employed wildcard versionsas well as multiple versions of these terms, for ex-ample, include, inclusive, included, mainstream,co-teach, coteach. An ancestry search of each ref-erence list was also employed, in order to identifyrelevant research that had been cited by authors ofidentified research. A descendant search of citedresearch, using the Social Sciences Citation Indexidentified reports that had cited relevant research.Finally, a hand search of relevant journals (anyjournals devoted to special education practice, forexample. Exceptional Children, Journal of SpecialEducation, Learning Disabilities Research & Prac-tice, Remedial and Special Education) was con-ducted to identify articles that may have beenoverlooked from the previous procedures.

We did not set any deliberate time limits inthe search. However, among the earliest referenceswas a paper by Bauwens et al. (1989), which citedno previous research (ongoing field test data werementioned). The first formal qualitative studies ofco-teaching as it is presently known appearedaround the mid-1990s, according to our searchprocedures. (A small number of reports did ap-pear before this time, but these did not meet ourquality criteria.)

DATA ANALYSIS

Once all relevant research reports were obtained,they were coded for a number of setting anddemographic variables, including geographicalregion; grade level; urban/rural/suburban setting;predominant co-teaching model; number of

3 9 6 Summer 2007

participants (including administrators, special ed-ucation and general education teachers, students,and other participants); type of disabilities repre-sented among the participants; socioeconomicstatus of the school; and subject(s) being taught.In addition, we coded selection criteria (e.g., rep-resentative, nonsystematic, known to investigator,considered outstanding), and whether or not co-teachers were volunteers. At least two codersagreed on all coding decisions.

Next, all research reports were converted codigital format and saved as separate documents.This was accomplished through retrieval from on-line versions of journals, and PDF files obtainedthrough Digital Dissertations. When necessary,reports were retyped and saved in electronic for-mat. Each report was saved as a separate docu-ment in NVivo.

All reports were read at least once before weimplemented coding procedures; during this pro-cess we took notes and wrote comments, andhighlighted significant text. We then imple-mented a process of open coding (see, e.g.,Creswell, 2006) to identify and code all seeminglyrelevant and consequential considerations. Thiswas an inclusive, recursive process, in which wecontinuously revisited previous coding decisionsto determine whether coding was being imple-mented systematically and consistently. Somecoding categories that appeared initially to be sig-nificant were found to be less well represented inthe literature as a whole. For example, we had ex-pected "appropriate curriculum" (i.e., accessible toall students, and appropriate for diverse learningneeds) would be considered an important compo-nent of successful co-teaching, yet reference tothis variable was made in only three reports. Wewere also surprised to note only a few oblique ref-erences to differentiated instruction, although thereasons for this became more clear over time.Likewise, we created coding categories for the in-fluence of prior experience, influence of high-stakes testing, class size, and teacher turnover;only ultimately to determine that these issueswere raised only rarely. Why these issues, and oth-ers, were only infrequently raised, however, was initself an important issue to be considered in thecontext of other data. Grade level at first seemedto us to be a variable of significance; however, an

overlapping and perhaps more significant variablewas seen to be content knowledge.

Overall, free coding of all studies resulted in69 categories ("free nodes" in NVivo), represent-ing many different facets of the co-teaching pro-cess. After this, a recursive process of categoryanalysis, contextual analysis, and identified rela-tionships among categories was implementedamong at least two coders. After discussion, appli-cation, and revision, we created four superordi-nate categories, each with at least 12 of ouroriginal category codes included:

• Expressed benefits of co-teaching.• Expressed needs for success in co-teaching.• Special and general education teacher roles in

co-teaching.• How instruction is delivered in co-taught

classes.

Although some overlap was noted, the originalcodes seemed to fit relatively easily within thesecategories. Subsequent analysis focused on axialcoding, where relationships between and amongcodes (within and across superordinate categories)were identified (Strauss & Corbin, 1998). For ex-ample, one of the most commonly mentionedcategories was planning and planning time for co-teaching; however, this category was very fre-quently mentioned (although not exclusively) inthe context of administrative support. Althoughmost investigations reported on professional ben-efits to co-teachers, this issue was mediated con-siderably by the issue of personal compatibility.

Data analysis procedures employed in this in-vestigation were largely inductive. The process ofanalytic induction "involves scanning the data forcategories of phenomena and for relationshipsamong such categories, developing working ty-pologies and hypotheses upon an examination ofinitial cases, then modifying and refining them onthe basis of subsequent cases" (LeCompte &Preissle, 1993, p. 254). Obtained data from theoriginal research reports were assimilated andevaluated in a recursive fashion, in order to de-velop hypotheses about the practices and perspec-tives associated with co-teaching. Similar toqualitative data analysis of original data, dis-crepant cases and negative cases were used to fur-ther understanding and refine hypothetical

Exceptional Children

constructs. Observations and themes from origi-nal research were subjected to the constant com-parative method, in which incidents, categories,and constructs were subjected to overlapping andrecursive comparisons (LeCompte & Preissle).For example, the paucity of data attesting to dif-ferentiated instruction, peer mediation, or strat-egy instruction in co-taught classrooms could atfirst appear puzzling, but was supported by otherdata attesting to the general education teacher'stypically dominant role in the co-taught class-room, coupled with the general educationteacher's typical affinity for whole class, homoge-neous instruction. As discussed in later sections,such practices placed significant limitations onco-teaching practice.

In this investigation, we avoided an actuarialapproach to data analysis. That is, rather thancounting instances of reported or observed phe-nomena and providing specific totals, means, orpercentages, we evaluated phenomena with re-spect to recurrence, corroboration, and presenceot absence of disconfirming instances in same orother research repotts (and how disconfirming in-stances, when observed, were explained). By thesemeans, we hoped to arrive at conclusions basedon procedures that were faithful to the data analy-ses employed in the original investigations.

STUDY QUALITY

One important consideration in research synthe-sis is the quality of the investigations being in-cluded. In making these determinations on thestudy level, we employed quality considerationsreferred to as "credibility or trustwotthiness" byBrantlinger et al. (2005). We were careful to en-dorse the caution of Btantlinger et al. against"using credibility measures as a checklist in a rigidor unrefiective way" (pp. 200-201); rather, weconsidered all these measures simultaneouslyalong with each study, employing such considera-tions as triangulation, disconfirming evidence,prolonged field engagement, detailed description,member checks and peer debriefing. We also con-sidered "quality indicators" as represented byBrantlinger et al. (Figure 3, p. 202) regarding sys-tematic and appropriate collection and represen-tation of data. We included all reports that met aminimum standard of quality, although some

variability was noted. It should further be consid-ered that all studies included had also been foundto be acceptable by some form of peer review,whether an editorial boatd, dissertation ot thesiscommittee.

In addition, we considered the credibility ofspecific data within individuai research reports.Two different forms of data were considered. Oneconsisted of original data (e.g., observations, in-terview transcripts, ot documentary evidence) col-lected from participants. The second form of dataconsisted of specific and general conclusionsdrawn by the researchers regarding co-teaching,based on the original data collected. For the pri-mary data reported by the authots of the researchreports, we considered carefully the quality indi-cators represented by Brantlinger et al. (2005).That is, for any participant comments reproducedin this synthesis, we ensured that, for example,the participant was appropriate, the question wasreasonable, and the comments were transcribedappropriately. Fot any researcher conclusion re-ported in this synthesis, we determined that theconclusion reflected appropriate credibility mea-sures (see Brantlinget et al.); that is, that datawere systematically collected and recorded, multi-ple informants and/or data sources were obtained,disconfirming evidence was considered, and theconclusion was reasonable and appropriate basedon the data collected.

RESU UTS

OVERALL CHARACTERISTICS OFTHE DATA SET

Using the search procedures and selection criteriastandards previously described, 32 original reportsof qualitative research on co-teaching were identi-fied (see Table 1). These reports involved as par-ticipants 454 co-teachers, 42 administrators, 142students, 26 patents, and 5 support personnel.These co-teachers were working in geographicallydiverse schools, representing states in the North-east, Mid-Atlantic, Southeast, Midwest, South-west, and West coast of the United States; inCanada; and in Austtalia.

As well as geographical representation, iden-tified studies represented a range of grade levels:

3 9 8 Summer 2007

J U

c

atio

a

E0

1o

C

achi

.E -c

list

u

tion

a

0

list

list

u u

list

JSuu

.5 -a1 I8 -

doc

, ob

s

c a

doc

sqo '

c

doc

sqo '

ini

doc

s, in

t

o

doc

sqo '

c

doc

, obs

c, o

bsc

doc

sqo '

c

u

, bio

a

doc

sqo '

in

c, ob

s

c c

t, ob

sin

doc

t, ob

s

c

t, ob

s

c

doc

t, ob

s

c

Io

•f3

3 "S

Gan

sub;

E EIt PA

J3 -D3

rt -tx>3

2 I"I -Q

3

ou

-! « - 4i

•3 "3 O

:S g- ? 3 c

CN u C (/] m

2SE)

;

• - ^

(3G

Etc

hrs

p; 2

adm

:; 3

inte

ofG

Tpa

irslS

EG

E,, 5

SE)

u

rs (6

G1

tchi

adm

; 12

O

pair

(dd

ition

al

i3at

2 pa

r* ofG

T7S

E;;

pair

(GE

,

SJ

GE,

3st

(3

ofG

Tpa

irs

1 SE

tchr

o

pairs

o

pair

(

supp

ort

CN

9SE;

GE,

20 s

t;

S"Ci

tchr

s0

pan

0 pa

il

-T3

•aaCO v^ —

ONONON

OOCN

ONONON

OO

QJ ^3 V)S 3 "S

cQ pa U

S 2 o

U 5 Q

O So 2

O3

OO

X X 3

oo

B-a

S u

o a.00 —'

(Noo

I ^Exceptional Children 3 9 9

s

uJm4h

1b§s

Pred

o

assis

(

•II

-a

5 E4-1 nl u «-<

n

o

o-o

o .£,c c

o

1 Z

-fi3

-s s" 00 I^

S3 " 6 b .,

"a I -s -3 Srv. D-. Q- D-,

ooo(N

cOUI

ouN

Ric

e

CNCN

Rosa

,

CNCN

13

-d

Sale

n

CNCNCN*-*

c

Tarr

a

^ 8 ^-T3 - a -o

o

o

o

^ -a

1 '!(u 1 3 So

PJ

w -a

w UH y

" fN

aa.

W

oo

o CNCN

OCOoo

o oiS i.j iS '^

.S .E .S -§

3 C

E' EU U to

-o E-a

(NOoCN

a. ^

oo

so(N

o

N

3 "

CXI

8 2 2

-ao

oo

1^ bo X

4 0 0 Summer 2007

15 involved primary, preschool, or elementaryschool classrooms; 14 studied junior high, middleschool, or high school classrooms; whereas 3investigated both elementary and secondary class-rooms. These schools and classrooms also repre-sented a range of locations, including 8 urban, 9subiirban, 4 rural schools, and 5 representing acombination of locations (6 were not reported).Ten of the reports specifically targeted "outstand-ing" examples of co-teaching for investigation;others were described as more typical of the co-teaching experience. Results are presented withrespect to the four superordinate coding cate-gories previously discussed.

BENEFITS OE CO-TEACHING

Benefits to Teachers. Teachers generally re-ported that they had benefited professionallyfrom co-teaching experiences. For example,Austin (2001), in his semistructured interviews of12 New Jersey co-teachers in K-12, agreed withmany other researchers in his fmding that generaleducation teachers

generally considered co-teaching to havecontributed positively to their professionaldevelopment: Speciai education co-teacherscited an increase in content knowledge, andgeneral education co-teachers noted the ben-efits to their skill in classrooni managementand curriculum adaptation, (p. 250)

In her qualitative investigation of three co-teach-ers in an integrated Grade 2/3 classroom, Bessette(1999) interviewed the general education teacher,who reported

"Having Mary as the special educationteacher show me what she knows, could onlymake me a better teacher. And, I feel that'sgoing to be the sarrie with Kelly, too—shehas lots of new ideas, and I've done nothingbut learn, and change, and grow." (p. 110)

Many other investigations specifically reportedsimilar professional benefits to co-teachers (e.g.,Buckley, 2005; Carlson, 1996; Curtin, 1998;Luckner, 1999; Rice & Zigmond, 2000; Salendet al., 1997; Tarrant, 1999; Thompson, 2001;Trent, 1998). For instance, one of the elementarygrade general education co-teachers from theFrisk (2004) investigation reported, " 'I learned so

much this year from my partner. I learned how toadapt lessons for each student; she really taughtme so much'" (p. 98). This perceived value, how-ever, appeared to be predicated on the two teach-ers being personally compatible. The need forcompatibility, discussed in a following section,was mentioned very often, frequently within thesame report, and several instances were providedwhere lack of compatibility undermined the effec-tiveness of the co-taught classroom (e.g.. Frisk;Norris, 1997).

Benefits to Students Without Disabilities.Teachers sometimes noted increased cooperationamong their students in co-taught, inclusiveclasses. Salend et al. (1997) quoted a general edu-cation kindergarten teacher who reported,

"Norma fell ofFher chair today and Robertimmediately asked. Are you OK?' in a con-cerned, caring way. Lee then got up to helpher pick up her crayons—it was wonderful."(p. 8)

Teachers sometimes noted increasedcooperation among their students inco-taught, inclusive classes.

Many other investigations supported theseconclusions, and provided evidence for academicbenefits, particularly through extra teacher atten-tion (e.g., Luckner, 1999; Pugach & Wesson,1995; Rice & Zigmond, 2000; Yoder 2000). Forexample, an elementary-level general educationstudent in the Drietz (2003) investigation re-ported, "'You can ask them [special educationteachers] a question, and they are there to helpyou"' (p. 30). Also in that investigation, however,a special education student reported, " 'Sometimesother people are asking for help when you needhelp more'" (p. 30). Co-teachers in a number ofinvestigations reported on the positive effects ofco-teacher collaboration as a social model for stu-dents (e.g., Carlson, 1996; Frisk, 2004; Hardy,2001; Haziett, 2D01; Trent, 1998). Across all in-vestigations, social benefits to students withoutdisabilities were discussed more frequently thanacademic benefits.

Benefits to Students With Disabilities. Reportsof benefits to students with disabilities were com-

Exceptional Children

mon in these investigations. Teachers in theWalther-Thomas (1997) investigation of 25 ele-mentary and middle schools reported that only afew students failed to succeed in co-taught classes:One special education teacher described a studentwho "'was truly amazed to Fmd that he could doOK in here . . . When he realized all of this, hewas willing to work harder than he ever had inthe self-contained classes'" (p. 399). Teachers inseveral investigations noted the benefit of expo-sure to peer models for appropriate behavior (e.g.,Carlson, 1996; Vesay, 2004; Ward, 2003; Yoder,2000).

One commonly expressed benefit of co-teaching was said to be the additional attentionreceived by students with disabilities. For exam-ple, Norris (1997) interviewed a general educa-tion middle school teacher, who responded

"The best thing about co-teaching is havinganother person in the classroom . . . knowingthat there are targeted students In the class-room who need extta help and having eithetthe co-teachei' or myself address those whilethe other teacher is doing something else."(pp. 84-5)

Five of six interviewed sixth-grade special andgeneral education students in the Drietz (2003)investigation mentioned the positive benefits ofextra attention. One student reported, "'I likethat there are two people to help out, and youdon't have to wait so long to get your question an-swered'" (p. 28). The sixth student, however, feltthat the extra classroom noise generated was dis-tracting. A student with hearing impairments in acombined first/second-grade class reported, '"It'sa good class for me because I learn more stuff"'(Luckner, 1999, p. 27). Pugach and Wesson(1995) interviewed 9 fifth-grade students in co-taught classes and concluded, "The students weinterviewed felt as if their academic and socialneeds-were being met better than had they beenin classes instructed by a single teacher" (p. 291).Dieker (2001) interviewed 54 secondary level stu-dents with and without disabilities and reportedthat all students reported benefiting from the co-taught class, except for one student labeled emo-tionally disturbed who reported, "'You can't getaway with anything'" (p. 19).

Student Skill Level. In spite of the substantialnumber of reports of student benefits, a numberof participants stated strongly their concern thatstudents included in co-tatight classes have a min-imum academic and behavioral skill level. Thiswas a very common qualification, appearing inmore than 20 of the 32 studies reviewed; discon-firmations, in the form of uiiqualified acceptanceof all students in co-taught classes, were notnoted. For example, Thompson (2001) studied11 elementary-level co-teachers and reported,"The participants repeatedly cautioned aboiit ad-ministrators forcing teachers to co-teach and feltequally adamant about including students withdisabilities whose needs could riot he met in thegeneral education setting" (p. 129). Six sec-ondary-level special education co-teachers in theWeiss and Lloyd (2002) investigation thoughtthat some students with special needs "did not be-long in co-taught classes but were there becauseschool policy required them to participate inmainstream classes" (p. 65). Some of the teachersin the Walther-Thomas (1997) investigation "re-ported . . . 'horror stories' about poorly plannedclassrooms . . . some classrooms ended up heavilyWeighted with students who had learning and/orbehavior problems. Unfortunately, these ill-fatedclassrooms set teachers and students up for failureand frustration" (p. 403). Bessette (1999) de-scribed the case of a student in a combined sec-ond/third-grade class who disturbed the harmonyof the class. Similarly, a second-grade teacher inthe Hazlett (2001) investigation reported

"Nathan had many teachers. He was here (inthe classroom) all day long . . . but he was sofrustrated and angry. He had tantrums he-cause he wanted to do what the other kidswere doing. He assaulted another child in theclassroom and after that he assaulted the TSSstaff, who was just a behavior person just forhim. . . . He had two people for just onechild!" (p. 107)

This teacher "emphatically denied that all chil-dten benefit from being in an inclusive class-room" (p. 107). Difficult students who threatenedco-teaching efforts were reported by many otherresearchers, (e.g., Carlson, 1996; Feldman, 1998;Frisk, 2004; Pugach & Wesson, 1995; Ward,2003). However, approaches for dealing with

4 O 2 Summer 2007

these students varied. That is, in the Carlson in-vestigation, one elementary co-teaching pair wasable to cope with problem students, but otherpairs were not. Feldman observed a secondarystudent who exhibited more than 75% off-taskbehavior. The special education teacher "alternatesbetween monitoring this student closely and ig-noring him altogether. [The general educationteacher] is essentially uninvolved with this stu-dent, appearing to view him primarily as [the spe-cial education teacher's] concern" (p. 80).Considering such cases, the general report of thebenefits to students with disabilities in co-taughtclasses must be tempered with teachers' concernthat students meet minimum skill expectations.

EXPRESSED NEEDS OF CO-TEACHERS

Administrative Support. In addition to re-ported benefits, teachers also expressed a numberof needs that in their view must be met for co-teaching to be successful. Primary among theseneeds was administrative support. For example,one teacher in Thompson's (2001) investigationof 11 elementary-level co-teachers spoke for thegroup in reporting, "'Administrative support—that would be number one. Number two—pick-ing the right teacher'" (p. 129). Salend et al.(1997) studied co-teachers in a kindergarten class-room and reported, "the support of the principalalso was instrumental in the success of the teach-ers' collaboration" (p. 8). Similarly, Chris andKelly, fifi:h/sixth-grade co-teachers studied in theCarlson (1996) investigation, "made it clear thatthe support of the principal was crucial" (p. 64).In Frisk's (2004) study of five elementary-level co-teaching dyads, who were in strong agreement onthis issue, one third-grade teacher reported, "'thedyad must be committed but . . . local and dis-trict school administration must also be commit-ted to supporting our inclusion model'" (p. 96).Other researchers supported this finding (e.g.,Curtin, 1998; Morocco & Aguilar, 2002; Norris,1997; Thompson; Vesay, 2004; Yoder, 2000). Nodisconfirming evidence—that administrative sup-port was not necessary—was identified. Adminis-trative support was seen to be linked to a numberof additional issues, discussed in the followingsections.

Volunteerism. Many teachers maintained thatit was necessary that co-teachers volunteer toteach together. Thompson (2001) reported thatall of the participating elementary teachers"strongly advocated for voluntary participation"(p. 129). Carlson (1996) reported that the ele-mentary-level behavior resource teacher, Amanda,"stated that it was critical 'that the impetus for theteam comes from the two individuals involved,that it's not imposed by administration'" (p. 154).The principal agreed that "'co-teaching cannot beforced. Rather, it is a way of doing things that thetwo teachers must choose, though it can be sug-gested. In other words, teachers have to pick theirco-teaching partners'" (p. 45). In Trent's (1998)study of four high school co-teachers, he re-ported,

Christine [the general education social stud-ies teacher] believed that the transfer ofKatherine [the special education teacher] wasa prime example of how teachers' opinionswere disregarded when planning co-teachingarrangements. Neither teacher had had a sayin this change and, unfortunately, Kather-ine's co-teaching experience with the U.S.Government teacher was not successful,(p. 510)

Describing an unsuccessful preschool co-teachingpair, Rosa (1996) commented, "the arrangementseemed doomed for a number of reasons. First,and possibly most important, the principal hadcome to Elaine and practically forced her to takeFrances because nobody else wanted her" (pp.137-138). Vesay (2004) studied three pairs ofearly childhood education co-teachers, and con-cluded, "the effect on their collaboration is: posi-tive when both teachers make a voluntarycommitment to initiating the partnership" (p.152).

Teachers' accounts of the necessity of volun-tary co-teaching were frequently reported (see alsoBuckley, 2005; Curtin, 1998; Frisk, 2004; Ha-zlett, 2001; Norris, 1997). However, Ward (2003)found a different opinion expressed in focusgroups of middle school teachers. One teacher re-ported,

"There are people in my building—this re-ally bothers me—that have the 'Free fromSpecial Ed' pass. I didn't know they [admin-

Exceptional Children 4 O 3

istrators] give those out, but some people inmy building have one and don't have anyspecial ed students because they exhibit qual-ities in the classroom that are not becomingto collaboration, so the special educator doesNOT want to place students in thoserooms." (p. 110)

Another teacher in this same investigation re-niarked,

"You have to say it is mandatory because Idon't think you ever want a policy that cer-tain teams or teachers can't have certain kids.Everyone should be doing something in theirown small way showing that they are movingalong that continuum." (p. Ill)

Howevet, the teachets in this investigation whofelt that co-teaching should be mandated also feltit should be phased in over a petiod of yeats andaccompanied by sufficient training and support.

Planning Time. A frequently noted issue wasthe importance of planning time, noted in nearlyall of the investigations. Yoder (2000) reportedthat "Ann [a junior high special educationteacher] noted in her journal, as well as repeatedlyduring the interview process that joint prepara-tion times are necessary, particularly during thefirst year of a co-taught class" (p. 104). In a studyof a secondary co-taught biology classroom,Curtin (1998) reported, "the special educationteacher felt the barrier to co-teaching was a lackof planning time for collaboration with the regu-lar education teacher" (p. 101). Dieker (2001)studied secondary-level co-teaching teams andconcluded, "the teams talked regularly about thestruggle to find adequate time to plan" (p. 20).These teachers reported having an average of 45.5min per week (often interrupted by other factors),but felt they needed nearly three times thatamount. In the Hazlett (2001) investigation, allco-teaching partners received 40-min scheduledplanning time per week. However, even this levelof planning time seemed insufficient, for teachersalso felt the need to meet on an ongoing basis, atlunchtime, in the morning, at recess, or at the endof the day. Vesay (2004) reported in her study ofthree preschool co-teaching teams, "In responseto a question of what makes their collaborativeteam successful Connie stated, 'For us it's sacred

planning time which we haven't had for twoyear[s]'"(p. 112).

Teachets frequently framed planning time inthe context of administrative support; for exam-ple, Austin (2001) interviewed co-teachers whoreported that they were satisfied with their pre-sent co-teaching assignment "but not with thelevel of support received from the school, notingthat they needed more planning time" (p. 251).Several other researchers discussed the importanceof administrative responsibility in facilitatingplanning, including Buckley (2005), Curtin(1998), Norris (1997), Ward (2003), and Yoder(2000).

Training. A very common theme across manyinvestigations was the need for teacher trainingfor co-teaching. In Vesay's (2004) study, onepreschool co-teacher, Connie, felt unprepared forcollaborative teaching. She admitted, "'Oh, abso-lutely! I was frightened, I had no background. Atrach[eostomy] scared me. A feeding tube fright-ened me, I was afraid I'd hurt somebody. I was!'"(p. 112).

In other instances, teachers expressed a needfor training to promote learning of more flexiblethinking (Buckley, 2005); strategies, and practicalskill development (Curtin, 1998); different co-teaching models (Feidman, 1998); use of technol-ogy (Luckner, 1999); characteristics of disabilities(Norris, 1997); collaborative consultation skills(Rice & Zigmond, 2000); group interpersonalskills (Rosa, 1996); and communicating more ef-fectively (Walther-Thomas, 1997). Most of theseinvestigations provided several examples of train-ing needs.

There were few disconfitmations of these ex-amples; however, although other teachers dis-agreed, one teacher from one of the fiveco-teaching pairs in the Frisk (2004) investigationreported,

"I think if someone is really interested in col-laboration the only way to really figure outhow to work with someone and how to in-teract is to do it. I never attended a work-shop in how to do inclusion or how tocollaborate." (p. 100)

A teacher in the Hazlett (2001) investigationfound that a district inservice '"wasn't very infor-mative since it didn't tell us how"' (p. 83). In spite

4 O 4 Summer 2007

of such instances, however, most teachers, whenasked, emphasized the importance of training.

Compatibility. Teachers were generally veryemphatic about the need for co-teachers to becompatible. Rice and Zigmond (2000) studied 17secondary co-teachers in Pennsylvania and Aus-tralia and concluded, "Several of the teachers . . .rated personal compatibility between partners asthe most critical variable for co-teaching success"(p. 194). Similarly, an elementary-level generaleducation teacher interviewed by Thompson(2001) commented when asked about co-teach-ing,

"I'd say, 'You should do it [co-teaching]. It'sawesome,' you know. But make sure that it'swith somebody that you get along with andthat you have the same, you know, ideasahout teaching and are equally motivated."(p. 128)

Similar cautions were observed in a numberof reports, including those by Buckley (2005),Luckner (1999), Norris (1997), Rice and Zig-mond (2000), Thompson (2001), and Westberg(2001).

A negative opinion, reported by the generaleducation middle school teacher in the Norris(1997) investigation, underlined the importanceof compatibility:

"If I had known that I would have to defendthe way I have always helieved in teaching,I would not have agreed to co-teach. . . .I have not been teaching for 30 years forsomeone else to tell me how to teach. . . .I am furious." (p. 107)

Frisk (2004) interviewed a general education first-grade teacher in an unsuccessful relationship whoattributed the dissolution of the partnership tothe special education teacher's

"inflexibility and personal issues." She thinksthe downfall in their collaboration occurredbecause "Julie spent a lot of time on thecomputer doing personal things. When youhave a lot of kids in the room with differentneeds you can't be doing it all by yourself."(p. 86)

Similar issues, frequently mentioned, includedmutual trust and respect (e.g., Curtin, 1998;Feldman, 1998; Frisk, 2004; Norris, 1997), and

appropriate attitudes (e.g., Buckley, 2005; Carl-son, 1996; Dieker, 2001; Rice & Zigmond,2000; Ward, 2003; Yoder, 2000).

Marriage. Many investigations included somereference to co-teaching as a marriage, that is, re-quiring effort, flexibility, and compromise for suc-cess. For example, Luckner (1999) reported, "Inmany ways, a co-teaching partnership can be con-sidered a professional marriage...it entails dealingwith a series of complex issues and emotions" (p.30). One of the Crade 5/6 co-teachers studied byCarlson (1996) reported

[Maureen] compared the co-teaching processto marriage, saying, "If you're not willing tohend then I don't think it would work." Kate[agreed], "It's like a marriage because youcompromise and you're getting different out-looks. You don't want to be a clone of oneanother." (p. 137)

Rice and Zigmond (2000) reported, "The teach-ers . . . described co-teaching as an unusuallyclose partnership or, what one termed, 'a profes-sional marriage,' which, 'like [a normal] marriage,you have to work at"' (p. 194). In discussing thefailures of a co-teaching relationship, Mastropieriet al. (2005) concluded, "It was difficult to deter-mine precisely what caused the erosion of the col-laborative relationship, but as the vice-principalreported, 'Forced marriages often fail'" (p. 265).Other references to co-teaching as a marriagewere reported by Bessette (1999), Buckley(2005), Curtin (1998), Frisk (2004), Moroccoand Aguilar (2002), Norris (1997), and Rosa(1996). The consistency of this metaphor pro-vides evidence for conformity of thought acrossstudies (cf Noblit & Hare, 1988).

TEACHER ROLES

Models of Co-Teaching. By a considerablemargin, the most prominent model of co-teach-ing reported in these investigations was some ver-sion of "one teach, one assist." For example,Westberg (2001) studied nine elementary co-teaching pairs and reported,

by far, the most prevalent teaching configu-ration observed was one teaching, one assist-ing. The general education teacher was mostfrequently the lead teacher, while the specialeducation teacher usually moved about the

Exceptional Children 4 O S

classroom and interacted as necessary withindividual students, although not necessarilyclassified students, (p. 70)

In some cases, the lead teaching duties alternatedbetween special and general education teacher(e.g., the high school co-teaching pair in theCurtin, 1998, investigation, or the fourth-gradeco-teachers in the Mastropieri et al., 2005, inves-tigation), but these cases were a decided minority.For instance, although Morocco and Aguilar(2002) observed team teaching in an eighth-grademath class, Magiera, Smith, Zigmond, andGebauer (2005) observed co-taught math classesin eight high schools and concluded.

The most common role . . . was monitoringof independent practice. . . . The other rolemost common to the special educationteacher was assisting students as the mathe-matics teacher maintained the role of pri-mary instructor. Cook and Friend (1996)described this as an appropriate role in thebeginning stages of co-teaching. . . . Teachersparticipating in our study, however, had co-taught for 3 to 5 years but had not gone be-yond this initial stage of co-teaching, (pp.20-21)

Antia (1999) observed five co-teachers in elemen-tary classes containing students with hearing im-pairments, and concluded

Although [the special educators] were re-sponsible for some direct teaching, they werealso responsible for assisting classroom teach-ers to make curricular adaptations and forplanning cooperatively with them. Thus,their major role appeared to be providingservices to classroom teachers rather than tothe children, (p. 213)

In an observational study of 14 high schools, Zig-mond and Matta (2004) concluded, "Our dataset indicates that the SET [special educationteacher] seldom took (or was permitted to take)the lead in instruction" (p. 63). Rice and Zig-mond (2000) concurred from their study of 17secondary teachers:

In all of our interviews and classroom obser-vations we did not find a model of co-teach-ing that fully met the criteria we set: a sharedteaching space with a diverse student group,shared responsibility for planning and for in-

struction, and substantive teaching by bothco-teaching partners, (p. 196)

One teach, one assist was the most commonmodel of co-teaching among the 16 co-teachers infour elementary schools studied by Haziett(2001). Haziett described the comments of a spe-cial education co-teacher in a developmentalkindergarten:

"Bertha and I use one teaching, one assisting.I go around the whole room for correctness.I think it's easier for her to be on the floor allthe time (because) she likes to be in control.Sometimes when I teach, she will interject. Inever interject when she's teaching becauseI'm not comfortable with her. (Besides) shehits all the basics when she teaches." (p. 101)

Other models of co-teaching were noted inthese investigations, although to a much lesser ex-tent. Vesay (2004) noted the use of parallel teach-ing in a preschool setting, although Haziett(2001) interviewed one teacher who reported,"'We tried the parallel (teaching), and it just didnot work out because two of the teachers havereal strong voices and each group was being verydistracted'" (p. 104). Anotber teacher reported," 'The kids with the most support have troubles inmath and language . . . so in those situations, theway we have it set up is parallel teaching eventhough we may not be in the same room'" (p.91). Other models reported by teachers or ob-served by researchers include team teaching(Curtin, 1998; Feldman, 1998; Mastropieri et al.,2005; Weiss & Lloyd, 2002); alternate teaching(Curtin); and station teaching (Tarrant, 1999).

It is interesting to note that some co-teachingmodels involved special education teachers andgeneral education teachers teaching in differentclassroom settings. For instance, Curtin (1998)reported the following observation of high schoolco-teachers:

These teachers were employing the co-teach-ing strategy called alternate teaching inwhich instruction is provided to studentsusing different approaches to a smaller groupof students. I was surprised that these twoteachers decided to separate the class, (p. 72)

Mastropieri et al. (2005) described a situation inwhich two co-teachers were in conflict: "In effect,the teachers determined that one way to reconcile

4 0 6 Summer 2007

serious problems in a co-teaching situation was todivide the class in two" (p. 265). Weiss and Lloyd(2002) reported observing co-teachers teaching inseparate classrooms:

Fot example, concetning his middle schoolsocial studies class, Jim said, "There were toomany disruptive hehaviots going on, andnone of the students [was] benefiting fromit. And the easiest fix I could come up withwas to split them, and we did. So he has 12students and I have 12 students." (pp.64-65)

Subordinate Roles and Content Knowledge. Inmahy instances the special education teacher as-sumed, or was seen to assume, a subordinate role(e.g., Antia, 1999; Buckley, 2005; Hazlett, 2001;Magiera et al., 2005; Mastfopieri et al., 2005;Norris, 1997; Pugach & Wesson, 1995; Rice &Zigmond, 2000; Zigmond & Matta, 2004). Forinstance, Norris wrote of a middle school specialeducation teacher, "identifying with the role inthe regular classroom as one of an assistant withless than equal status and an inability to success-fully meet the needs of students, became frus-trated in the co-teaching process " (p. 72). Indiscussing three teams of high school world his-tory co-teachers, Mastropieri et al. concluded, "Itwas rare to observe special educators deliveringinstruction to the entire class" (p. 265). Notesfrom observations of one of these teams revealed

"This team of teachers interacted as a bossand an assistant when working with the stu-dents. The general education teacher as-sumed conttol of all aspects of the classroomat all times. . . . Throughout this time pe-tiod, the special educator sat in the toom andoccasionally went atound to individual stu-dents to see if they needed any assistance."(observation notes; p. 266)

A special education teacher in the Antia investiga-tion reported, "Tm an aide sometimes, I'm an in-terpreter sometimes, and sometimes I'm ateacher'" (p. 211). A preschool special educationteacher repotted to Rosa (1996) "'She [the gen-eral education teacher] cettaihly allowed me todevelop all the behavior management programsthat were going on and things like that. I don'tthink she felt that I was taking over there, either'"

(p. 84), using language indicative of a subordinaterole.

In many cases, the subordinate role of thespecial education teacher appeared to reflect therelatively greater content knowledge of the gen-eral education teachet (e.g., Feldman, 1998; Mas-tropieri et al., 2005; Morocco & Aguilar, 2002;Pugach & Wesson, 1995; Rice & Zigmond,2000; Rosa, 1996). Fot instance, Weiss and Lloyd(2002) reported,

teachets said that the content atea of the classfotced them to take certain toles. Fot exam-ple, Greta said, "I don't feel confident insome classes to be a team" . . . and Esthersaid, "Do you think I would have the audac-ity to go in the geometry class and say I wasa collahotative teachet?" (p. 65)

In repotting on a high school English teacher anda special education co-teachet. Rice and Zigmond(2000) commented.

The two teachets described theit practice as"an enmeshing of out abilities" . . . but theywete cleatly not equal pattnets in the instruc-tion. In most cases, this disparity in toles wasexplained as necessaty because the special ed-ucation teachet lacked content knowledge,(p. 195)

However, expertise in content knowledge onthe part of the special education teachet could beassociated with a higher degree of shared responsi-bility (e.g., Pugach & Wesson, 1995; Rice & Zig-mond, 2000). Yoder (2000) obsetved a highschool Ametican literature class in which both co-teachers shared most teaching responsibilities eq-uitably. The general education teacher reported,"'I think Carmine [the special education teacher]and I mesh well because we complement eachother, and she also has the English background. Ithink that's a very string contributing part of it'"(p. 187).

Special education teachers in mote subordi-nate roles were not confined to secondary gi-adelevels. Instances of special education teachers as-suming subordinate roles in elementaty-levelclasstooms wete teported by, for example, Antia(1999), Hazlett (2001), Rosa (1996), and Salendet al. (1997), suggesting that secondary contentknowledge is not the only determinant of teacherroles. However, the lower status of special educa-

Exceptional Children 4 O 7

tion co-teachers, common at all grade levels,seemed hiore consistent in secondary levels, par-ticularly in those classes with rhore specializedcontent knowledge.

Subordinate Role and "7Mr^" Teachers alsoidentified turf issues that may have contributed tothe relatively subordinate role of the special edu-cation teacher. As stated by a successful middleschool special educatioh co-teacher in the Mo-rocco and Agiiilar (2002) investigation, '"We'reentering their environment and we have to be theones to go one step above and beyond'" (p. 332).A high school special education teacher reported,

'"Anytime you walk into another teacher'sclassroom there's going to be some type ofnegotiation that needs to occur for both ofyou in terms of jiist territory and what'sasked of you. And that's a tough thing to ne-gotiate.'" (Yodet, 2000, p. 150)

An administrator in the Norris (1997) inves-tigation stated, " 'We as an industry are very terri-torial. It is really difficult for teachers to worktogether, change, and accept new ideas"'(p. l45).Wood (1998) reported in her study of six elemen-tary co-teachers

When special education teachers attemptedto transplant their special education tech-niques or materials that were consideredatypical in the general education environ-ment . . . the general education teachers ad-mitted that they would sometimes reactterritorially, snubbing their suggestions, (pp.190-191)

One fifth-grade teacher in that investigationstated,

"[The special education teacher] tried to tellme how she wanted the discipline to run.And she brought in a chart and said, 'Nowwheh [Jeanie] does this, you put a star here.When [Tim] does this, you put a circle here.'And I said, 'Well, OK.' But, I never did itbecause that's not the way the discipline inthe class runs." (p. 191)

Buckley (2005) provided a middle schoolspecial education teacher's description of the diffi-culties of fitting into the general educationteacher's classroom:

"I mean, if you're talking, I try to let you fm-ish whatever you're doing. And then I'll con-tribute. I try not to bump in. Well, she toldme I was barging in on her. So it was like,'You will please not talk in my classroom.'And I was like, well, maybe I've got to beable to say something. It got to the pointthat I was raising rhy hand to talk. I thought,if this isn't stupid. But yeah. She really hadtrouble with somebody else in there." (p.174)

Buckley concluded, "The regular education teach-ers saw theniselves as the leader of their class-rooms" (p. 179). Although most teachers valuedthe special education teacher, "all of them alsosaid that they wanted things done their way andwanted to maintain control" (p. 179). Describinghovv she established her co-teaching relationshipwith a special education teacher, one general edu-cation teacher said,

"Okay, well first I would be in charge.[Laughs] And I would let her first observeme. And then I would invite her to perhapstry a couple of lessons and see how she does.And then perhaps now we're establishing abetter rapport with each othet and now I ambeginning to trust her, to trust her to teachin the way I am expecting the children to betaught, allow her to gradually take over somelessons." (p. 179)

Although ownership or turf issues vvere com-nion, they were not found in every classroom. Forexample. Frisk (2004) reported on the commentsof a general educatioh co-teacher of a third/fourth-grade class concerning co-teaching:

Faith thinks that it's successful because there'sno competition of egos in the room. "Wehave no problem if Erica takes one of mylessons and modifies it ot whatever... I don'tfeel that I must have total control of theroom, and I don't think she does, either . . .we complement one another." (p. G<S)

Note even in this example, however. Faith's use ofthe possessive in referring to "one of my lessons"(emphasis added).

Summer 2007

INSTRUCTIONAL DELIVERY INCO-TAUGHT CLASSES

General Education Teacher. There was veryconsiderable agreement that general education co-teachers favored strategies that could be applied tothe class as a whole. Antia (1999) studied pri-mary-grade co-teachers and reported, "teacherswere most ready to make adaptations that theyperceived as benefiting the entire class, for exarn-ple, visual strategies" (p. 213). Buckley (2005)concluded from her study of middle school socialstudies co-teachers, "Regular education teacherstend to plan globally rather than focusing on in-dividuals. Therefore, a strategy suggested by a spe-cial education teacher may possibly be providedfor the whole class" (p. 176). Hardy (2001) stud-ied a high school biology co-teaching pair and re-ported that some adaptations employed by thegeneral education teacher included advance orga-nizers, individual teaching, pacing, and classroomsupports such as weekly schedules and seating as-signments. However, Hardy observed:

a discrepancy was noted in the teachers'awareness of the necessity for specialized in-struction. . . . The teachers used whole-classactivities 100% of the time . . . students withdisabilities in the co-taught classrooms fol-lowed the same sequence of activities andused the same materials as peers, (p. 185)

Similarly, Feidman (1998) reported of his sec-ondary co-teachers, "Co-teachers are not likely toprepare individual lesson plans to accommodatestudents with LD . . . most of the accommoda-tions appear to be designed at the whole classlevel" (p. 89). Mastropieri et al. (2005) observed"little differentiation of instruction to address in-dividual needs" (p. 266) in co-taught high schoolworld history classrooms. Magiera et al. (2005)reported that teachers typically employed awhole-class lecture and independent seat work ap-proach in 10 co-taught high school mathematicsclasses.

On some occasions, the general educationteacher's reliance on traditional methods was asource of frustration to the special educationteacher. The middle school special educationteacher in the Norris (1997) investigation re-ported, "I'm not happy with^ instructional meth-ods that don't address the needs of students. Why

teach from a textbook that some students can'tread?" (p. 69). On the other hand, general educa-tion teachers frequently did not see a distinctionin the way they should address individual stu-dents. A general education fifth-grade teacher inthe Pugach and Wesson (1995) investigation re-marked, "'Personally, I haven't seen any magicmiracle on how to reach these [students]. It's thesame thing in regular ed. And thete is nothingdifferent that we don't do if we had the time'" (p.291). In some instances, general education teach-ers reported that they needed to help studentswith disabilities prepare for "the real world"(Buckley, 2005, p. 182), and this to some extentmay explain their observed reluctance to individ-ualize.

Special Education Teacher. One special educa-tion teacher in an eighth-grade math class "as-sumed a full range of instructional roles—madethe transition, introduced and explained the ac-tivity, provided instruction, and gave studentsfeedback on answers provided by the groups"(Morocco & Aguiiar, 2002, p. 336). However,such cases were rare. More typically, special edu-cation teachers generally provided the role of sup-porting the traditional role of the generaleducation teacher. Curricular adaptations of ahigh school special education co-teacher (Trent,1998, p. 506) included "developing outline sheetsor rnodified study guides for the textbook chap-ters (e.g., fiU-in-the-blank worksheets indicatingpage numbers where answers could be found)."Although these activities were generally seen to behelpful, they seemed quite different from the in-structional practices usually attributed to specialeducation teachers (e.g., Mastropieri & Scruggs,2006). Feidman (1998) reported, "The primarystrategy to accommodate LD students in this [sec-ondary] classroom takes the form of [the specialeducation teacher] providing temporary assistancevia answering a question, redirecting off task be-havior, or prompting attention" (p. 97). Curtin(1998) observed a secondary science class and re-ported, "The special education teacher stood atthe mailbox and made sure each student placedtheir folder in the ptoper slot" (p. 79). Hardy(2001) observed a special education teacher in ahigh school biology class and reported, "occasion-ally, during the lecture Janet would interject acomment to the class. At one time she said.

Exceptional Children 4O9

'Remember when we talked about what enzymesdid?'" (p. 166). In a first-grade class, the generaleducation teacher led the class in a song, whilethe special education teacher "moved about theroom organizing the chairs and picking up mate-rials that were out of place from the previous ac-tivity" (Rosa, 1996, p. 84).

The tasks of the special education teacherseemed to reflect limitations imposed on thewhole-class instruction that was commonly em-ployed in general education classrooms. Magieraet al. (2005) concluded, "Because whole-class in-struction continued to be the norm, special edu-cation teachers had few opportunities to offerindividual instruction" (p. 22). Some special edu-cation teachers served as models. " 'The first year Iwas a model for the students. Often, if [the sub-ject teacher] is lecturing, I would do the notes onthe overhead [projector] to model note-taking'"(Rice & Zigmond, 2000, p. 195). Zigmond andMatta (2004) observed 41 secondary co-teachingpairs and concluded.

The second teacher was a nice addition, anoccasional relief for the GET [general educa-tion teacher], and more attention to studentswhen class is organized for small group(team) or independent seatwork. But none ofwhat we saw would make it more likely .thatthe students with disabilities in the classwould master the material. We did not hearthe SETs [special education teachers] chimein with carefully worded elaborative explana-tions. . . . We virtually never saw the SETprovide explicit strategic instruction to facili-tate learning or memory of the content ma-terial, (p. 73)

Special Education Teacher and Behavior Man-agement. Frequently, it was assumed that the spe-cial educat ion teacher would assumeresponsibility for any problem behaviors that oc-curred in the classroom. Bessette (1999) reportedon a journal entry of a general education second/third-grade classroom teacher:

"Michael presents many challenges—the fearof the other students is real, and I will pledgeto keep them safe. Mary will restrain and re-move him while I continue with the rest ofthe class. It has taken its toll on all of us." (p.141)

Rosa (1996) reported that a first-grade special ed-ucation co-teacher "handled the problems thatcame up which might disrupt the activity" (p.84). Feldman (1998) observed of a secondary co-teaching pair: "[The general education teacher]actually presents the lesson information while [thespecial education teacher] stands off to one sideand focuses most of her attention on monitoringthe behavior of three of the seven LD students"(p. 80). The special education teacher as behaviormanager was described in a number of other in-vestigations (e.g., Buckley, 2005; Rice & Zig-mond, 2000; Trent, 1998; Yoder, 2000).

Peer Mediation. In some individual cases,peer mediation in the form of cooperative learn-ing or peer tutoring was employed very produc-tively in co-taught classrooms (e.g., Carlson,1996; Mastropieri et al., 2005; Pugach & Wes-son, 1995). For example, Tarrant (1999) reportedon the use of partner reading and partner spellingactivities in a mixed-grade elementary classroom.Overall, however, peer mediation was observed farless than might be expected in these investiga-tions. Feldman (1998, p. 96) observed that

GE/SE 1 are the only co-teaching pair in thepresent study to consistently utilize peer me-diated instructional strategies . . . . This iscurious in light of the relative popularity ofcooperative learning (Slavin, 1990) and em-pirical support for various forms of peer me-diated instruction (Utley, Mortweet, &Greenwood, 1997).

Referring to peer tutoring, a high school biologyteacher reported to Hardy (2001), "'I am goingto tell you however that right now there is notthat much of that going on.' Janet [the special ed-ucation teacher] adamantly disagreed and ex-pressed, 'Yes, there is.' According to field notes,peer tutoring was never observed" (p. 181).

Norris (1997) observed a middle school co-taught class and concluded, "Students rarely in-teracted with each other once instruction beganwithout permission and were reprimanded fordoing so" (p. 132).

Other techniques that might have been ex-pected were only rarely observed, such as princi-ples of effective instruction (Mastropieri et al.,2005; Tarrant, 1999; Westberg, 2001); differenti-ated instruction (Tarrant; Yoder, 2000); appropri-

4 1 O Summer 2007

ate curriculum (Dieker, 2001; Mastropieri et al.);mnemonic instruction (Mastropieri et al.;Walther-Thomas, 1997; Yoder); effective studentgrouping (Pugach &C Wesson, 1995; Tarrant); orstrategy instruction (Mastropieri et al.; Tarrant;Walther-Thomas). Beyond these exceptions, peermediation and other potentially helpful inclusiontechniques appeared to be greatly underemployed.

D I S C U S S I O N

We employed metasynthesis methodology in thisreview in order to enable us to examine issues andfmdings within and across studies with more pre-cision and to summarize the research on the levelof individual data rather than on the level of theresearch report. Although many of the present re-sults may have been obtained from a more tradi-tional study-by-study review, we believe ourmethodology allowed us to look across demo-graphic variables (e.g., location, grade level)within and across studies, to aggregate data at thelevel of the individual case rather than individualstudy, and to focus on within- as well as across-study variation. Use of NVivo software allowed usto compare a large number of issues witbin andacross studies and to examine more carefully dis-confirmations (or their lack) and interactionswith other issues. We believe that some of ourfindings—for example, grade level and contentknowledge, planning time and administrativesupport, or benefits tempered by concerns aboutstudent skill levels—were more easily uncoveredthrough this methodology.

Several general conclusions can be drawnfrom the results of this investigation. First, ad-ministrators, teachers, and students perceive themodel of co-teaching to be generally beneficial, togeneral education and to (at least some) specialeducation students in both social and academicdomains, and to the professional development ofteachers. Second, teachers have identified a num-ber of conditions needed for co-teaching to suc-ceed, including sufficient planning time,compatibility of co-teachers, training, and appro-priate student skill level. Many of these concernswere linked to the more general issue of adminis-trative support. Third, the predominant co-teach-ing model reported in these investigations is "one

teach, one assist," with the special educationteacher often playing a subordinate role deter-mined, in part, by content knowledge, teacher"turf," and the greater numbers of general educa-tion students in the co-taught classroom. Fourth,general education teachers typically employ wholeclass, teacher-led instruction with little individu-alization, whereas special education teachers func-tion largely as assistants in support of specialeducation students and other students in need,within the existing classroom context. Althoughseveral exceptions were noted for each of thesegeneral findings, overall the consistency of con-clusions—drawn across studies ranging widely ingrade level, subject matter, geographical location,specific setting, and student characteristics—wasremarkable. In short, it appears that the concernsabout co-teaching raised years ago by Boudah,Schumacher, and Deshler (1997) were prescient.

[TJeachers have identified a numberof conditions neededfi)r co-teaching tosucceed, including sufficient planningtime, compatibility of co-teachers, training,and appropriate student skill level

On a positive note, it can be concluded thatteachers and administrators were satisfied overall,or in some cases very enthusiastic, with co-teach-ing as presently practiced, and that most objec-tions raised in these investigations reflectedspecific co-teaching circumstances (e.g., compati-bility of co-teachers, administrative support)rather than the practice of co-teaching itself.These conclusions, however, must be tempered bythe fact that participants in nearly one third ofthe investigations were selected as being outstand-ing examples of co-teaching, and that most of theremaining teachers had volunteered for (or at leastnot objected to) these assignments. And, as Bakerand Zigmond (1995) noted, it is difficult to im-plement a policy based on volunteerism.

Examined critically, however, the practice ofco-teaching as described in these investigationscan hardly be said to resemble the truly collabora-tive models described by, for example. Cook andFriend (1995) or Walther-Thomas et al. (2000).

Exceptional Children

If the qualitative research to date represents gen-eral practice, it can be stated that the ideal of truecollaboration between two equal partners—fo-cused on curriculum needs, innovative practice,and appropriate individualization—has largelynot been met. Classroom instructional practiceshave not changed substantially in response to co-teaching. Classroom instruction has generallycontinued as whole class and lecture driven, andspecial education co-teachers have generally at-tempted to fit within this model to deliver assis-tance to students in need. Practices known to beeffective and frequently recommended—such aspeer mediation, strategy instruction, mnemonics,study skills training, organizational skills training,hands-on curriculum materials, test-taking skillstraining, comprehension training, self-advocacyskills training, self-monitoring, or even generalprinciples of effective instruction (e.g., Mas-tropieri & Scruggs, 2006)—^were only rarely ob-served. As a consequence, the co-teaching modelof instruction is apparently being employed farless effectively than is possible. As noted earlier byZigmond and Baker (1994) of inclusion classes,students with special needs are receiving goodgeneral education instruction, with assistance—but are they receiving a special education? Resultsof the present synthesis suggest they are not.

If the qualitative research to date representsgeneral practice, it can be stated that theideal of true collaboration between twoequal partners—-focused on curriculumneeds, innovative practice, and appropriateindividualization—has largely not beenmet.

The present results can be linked to moregeneral characterizations of teacher collaboration.Hargreaves (1994; see also Hargreaves, 2003) sug-gested that teacher collaboration can lead to in-creased confidence, which can lead in turn tomore experimentation and risk-taking, and ulti-mately continuous improvement. However, gen-uine collaboration must be spontaneous,voluntary, unpredictable, and oriented toward de-velopment. In contrast, co-teaching as described

in the present qualitative studies contains manyfeatures of what Hargreaves (1994) referred to as"contrived collaboration." Although many of thepairings described in these investigations werevoluntary (and some were experimental and inno-vative), overall they were regulated by administra-tors (often imperfectly), fixed in time and space,and predictable. Such a collaboration "divertsteacher's efforts and energies into simulated com-pliance with administrative demands that are in-flexible and inappropriate to the settings in whichthey work" (Hargreaves, 1994, p. 208).

There is a further issue. For such collabora-tion to be effective, the individuals in each pairshould be on an equal footing (unless it is mutu-ally understood that one of the pair is clearly ad-vanced in, for example, experience, expertise, orprofessional judgment, as in mentoring pairings).In co-teaching, however, the general educationteacher—because of her ownership of the class-room, the curriculum, the content, and most ofthe students—is very often in the dominant role,regardless of experience, expertise, or judgment.Therefore, the overall tilt of the classroom is typi-cally in the direction of the general educationteacher, where whole-class, teacher-led instructionis the rule, and the special education teacher ap-plies assistance only within the context of the ex-isting classroom structure. That this role issometimes mediated by a high level of contentknowledge on the part of the special educationteacher suggests that the special education teachermay be more accepted only to the extent to whichshe resembles the general education teacher. Inthese circumstances, a truly collaborative relation-ship—in the words of Rice and Zigmond (2000),"a shared teaching space with a diverse studentgroup, shared responsibility for planning and forinstruction, and substantive teaching by both co-teaching partners" (p. 196)—is very unlikely todevelop.

The participants of the investigations repre-sented in this metasynthesis cannot be character-ized as a random sample, and to that extent, therelationship between the present observations andthe population as a whole is unknown. In somecases (e.g., Hazlett, 2001), co-teachers declined toparticipate, citing problems in the co-teaching re-lationship they did not wish to discuss. Further,10 of the 32 investigations had been identified as

4 1 2 Summer 2007

outstanding examples of successful co-teaching(and none specifically selected as a negative exam-ple). It seems likely, then, that the studies in-cluded represent a mote favorable picture ofco-teaching than exists in general.

Neither, however, should the generalizabilityofthe present findings be discounted entirely. Therepotts included in this metasynthesis teptesenteda substantial number of teachers and administra-tors, in a wide variety of settings and situations.Nevertheless, we were struck by the remarkableconsistency ofthe findings. For example, in 21 ofthe investigations, general education teachersmaintained that a minimal student skill level wasan important criterion for successful inclusion;none of the other investigations specifically con-tradicted this conclusion. Very similar positionswere voiced on the impottance (and often, thechallenges) of planning time in 30 of the investi-gations; there were no disconfitming reports.Twenty-five of the investigations characterized thespecial education teacher's role as an assistant orin some way subordinate (although a smallernumber of investigations described differenttoles). The large size and diversity of the sampleand the consistency of many of the results arguestrongly that the present conclusions ate very sug-gestive of contemporary practice.

Future research could address tbe means bywhich individual schools are able to develop ttulycollaborative or genuine partnerships, and thespecific gains that can be realized by such prac-tices. Additionally, further efforts in tbe area ofqualitative research synthesis could help bring thevoices of individual students, teachers, and ad-ministrators into the domain of public discourseand help to strengthen the impact of qualitativetesearch. It is hoped that the present effort repre-sents one useful step in tbat direction.

REFERENCES

References marked with an asterisk indicate stud-ies included in the metasynthesis.*Antia, S. D. (1999). The toles of special educators andclasstootn teachets in an inclusive school. Journal ofDeaf Studies and Deaf Education, 4, 203-214.*Austin, V. L. (2001). Teachets' beliefs about co-teach-ing. Remedial and Special Education, 22, 245-255.

Baket, J. M., & Zigmond, N. (1995) The meaning andptactice of inclusion fot students with leatning disabili-ties: Themes and implications fot the five cases, journalofSpecial Education, 29, 163-180.Bauwens, J., Houtcade, J. J., & Ftiend, M. (1989). Co-opetative teaching: A model fot genetal and special ed-ucation integration. Remedial and Special Education,10(2), 17-22.

Beck, C. T (2001). Cating within nursing education: Ametasynthesis. Journal of Nursing Education, 40,101-109.*Bessette, H. J. (1999). A case study of genetal and spe-cial educatots' petspectives on collabotation and co-teaching within a language based integtated elementaryclasstoom. Dissertation Abstracts Intemational, 60 (09),3317A. (UMI No. AAI9946579)

Boudah, D. J., Schumachet, J. B., & Deshlet, D. D.(1997). CoUabotative insttuction: Is it an effective op-tion fot inclusion in secondaty classtooms? LearningDisability Quarterly, 20, 293-316.Btantlinget, E., Jimenez, R., Klingnet, J., Pugach, M.,& Richatdson, V. (2005). Qualitative studies in specialeducation. Exceptional Children, 71, 195-207.*Buckley, C. (2005). Establishing and maintaining col-labotative telationships between tegulat and special ed-ucation teachets in middle school social studiesinclusive classtooms. In T E. Sctuggs & M. A. Mas-ttopieti (Eds.), Cognition and learning in diverse settings:Vol 18. Advances in learning and behavioral disabilities(pp. 153-198). Oxfotd, UK: Elseviet.Campbell, R., Pound, P., Pope, C , Btitten, N., Pill, R.,Morgan, M., & Donovan, J. (2003). Evaluating meta-ethnogtaphy: A synthesis of qualitative teseatch on layexpetiences of diabetes and diabetes cate. Social Scienceand Medicine, 56, 671-684.*Catlson, L. D. (1996). Co-teaching fot integtation:An explotatoty study. Dissertation Abstracts Interna-tional, 57(08), 3389A. (UMI No. AAGNN11685)Cook, L., & Etiend, M. (1995). Co-teaching: Guide-lines fot creating effective practices. Eocus on Excep-tional Children, 25(3), 1-16.Cook, L., & Etiend, M. (1996). Co-teaching: Guide-lines fot creating effective practices. In E. L. Meyet, G.A. Vetgason, & R. J. Whelan (Eds.), Strategies for teach-ing exceptional children in inclusive settings (pp.309-330). Denvet, CO: Love.

Cteswell, J. W. (2006). Educational research: Planning,conducting, and evaluating quantitative and qualitativeresearch. Uppet Saddle Rivet, NJ: Ptentice-Hall.

Exceptional Children 4 1 3

*Curtin, J. P. (1998). A case study of co-teaching in aninclusive secondary classroom. Dissertation Abstracts In-ternational, 60 (01), 31A. (UMI No. AAG9917397)*Dieker, L. A. (2001). What are the characteristics of"effective" middle and high school co-taught teams forstudents with disabilities? Preventnig School Failure, 46,14-23.Dieker, L. A., & Murawski, W. W. (2003). Co-teachingat the secondary level: Unique trends, current trend,and suggestions for success. The High School Journal,86{A), 1-13.*Drietz, A. G. (2003). How students on individual ed-ucation plans perceive team teaching. Masters AbstractsInternational, 41 (06), 1580. (UMI No. AAI1414185)Estabrooks, C. A., Field, P. A., & Morse, J. M. (1994).Aggregating qualitative findings: An approach to theorydevelopment. Qualitative Health Research, 4, 503-511.*Feldman, R. K. (1998). A study of instructional plan-ning of secondary special and general educationco-teachers to accommodate learning disabled studentsin the general education classroom. DissertationAbstracts International, 58 (07), 4731 A. (UMI No.AAG9900508)Forness, S. R. (2001). Special education and relatedservices: What have we learned from meta-analysls? Ex-ceptionality, 9, 185-197.Fraser, D. (1999). Q_SR NUD*IST Vivo Reference guide.Melbourne, Australia: Qualitative Solutions and Re-search.Freeman, D. (1983). Margaret Mead and Samoa. Gam-bridge, MA: Harvard University Press.Friend, M. (2002). An interview with Dr. MarilynFriend. Intervention in School and Clinic, 37, m-lli.Friend, M., & Gook, L. (2003). Interactions: Collabora-tion skills for school professionals (4th ed.). New York:Longman.Friend, M., & Reising, M. (1993). Go-teaching: Anoverview of the past, a glimpse at the present, and con-siderations for the future. Preventing School Failure, 37,6-10.*Frisk, G. A. (2004). Teacher collaboration: Learningfrom inclusion dyad dialogues. Dissertation Abstracts In-ternational, 65 (6), 2158A. (UMI No. AAI3135903)Gately, S. E., & Gately, F J. (2001). Understanding co-teaching components. TEACHING Exceptional Chil-dren, 33(4), 40-47.Gersten, R., & Baker, S. (2000). What we know abouteffective instructional practices for English-languagelearners: A multi-vocal research synthesis. ExceptionalChildren, 66, 454-470.

Glaser, B. G., & Strauss, A. L. (1971). Status passage: Aformal theory. Ghicago: Aldine.Glass, G. V., McGaw, B., & Smith, M. L. (1979).Meta-analysis in social research. Thousand Oaks, GA:Sage.*Hardy, S. D. (2001). A qualitative study of the in-structional behaviors and practices of a dyad of educa-tors in self-contained and inclusive co-taught secondarybiology classrooms during a nine week science instruc-tion grading period. Dissertation Abstracts International,61 (12), 4731A. (UMI No. AAI3000278)Hargreaves, A. (1994). Changing teachers, changingtimes: Teachers' work and culture in the postmodern age.New York: Teachers Gollege Press.Hargreaves, A. (2003). Teaching in the knowledge society:Education in the age of insecurity. New York: TeachersGollege Press.*Hazlett, A. (2001). The co-teaching experience jointplanning and delivery in the inclusive classroom. Dis-sertation Abstracts International, 62 (12), 4064A. (UMINo. AAI3037002)Jensen, L. A., & Allen, M. N. (1994). A synthesis ofqualitative research on wellness-illness. QualitativeHealth Research, 4, 349-369.Keefe, E. B., Moore, V., & Duff, E (2004). The four"knows" of collaborative teaching. TEACHING Excep-tional Children, 3(^(5), 36-42.LeGompte, M. D., & Preissle, J. (1993). Ethnographyand qualitative design in educational research. New York:Academic Press.Light, R. J., & Pillemer, D. B. (1984). Summing up:The science of reviewing research. Cambridge, MA: Har-vard University Press.*Luckner, J. L. (1999). An examination of two co-teaching classrooms. American Annals of the Deaf, 144,24-34.*Magicra, K., Smith, G., Zigmond, N., & Gebauer, K.(2005). Benefits of co-teaching in secondary mathe-matics classes. TEACHING Exceptional Children,37{3), 20-24.Mastropieri, M. A., & Scruggs, T. E. (2006). The inclu-sive classroom: Strategies for effective instruction (3rd ed.).Upper Saddle River, NJ: Prentice-Hall.*Mastropieri, M. A., Scruggs, T. E., Graetz, J., Nor-land, J., Gardizi, W, & McDuffie, K. (2005). Gasestudies in co-teaching in the content areas: Successes,failures and challenges. Intervention in School andClinic, 40, 260-270.Mead, M. (1928). Coming of age in Samoa. New York:William Morrow.

4 1 4 Summer 2007

Miles, M. B., & Hubetman, A. M. (1994). Qualitativedata analysis: An expanded sourcebook (2nd ed.). Thou-sand Oaks, CA: Sage.*Motocco, C. C , & Aguilat, C. M. (2002). Coteach-ing fot content undetstanding: A schoolwide model.Journal of Educational and Psychological Consultation,13, 315-347.Mutawski, W. W., & Dieket, L. A. (2004). Tips andsttategies fot co-teaching at the secondaty level.TEACHING Exceptional Children, 36(5), 52-58.Mutawski, W. M., & Swanson, H. L. (2001). A meta-analysis of the co-teachitig teseatch: Whete ate thedata? Remedial and Special Education, 22, 258-267.Noblit, G. W., & Hate, R. D. (1988). Meta-ethnogra-phy: Synthesizing qualitative studies. Thousand Oaks,CA: Sage.*Nottis, D. M. (1997). Teachets' petceptions of co-teaching in an inclusive classtoom in a middle school:A look at genetal education and special educationteachets wotldng togethet with students with leatningdisabilities. Dissertation Abstracts International, 58 (06),2162A. (UMI No. AAG9735088)Patetson, B. L., Thotne, S. E., Canam, C, & Jillings,C. (2001). Meta-study of qualitative health research.Thousand Oaks, CA: Sage.Pugach, M. (2001). The stoties we choose to tell: Ful-filling the ptomise of qualitative teseatch fot special ed-ucation. Exceptional Children, 67, 439—453.*Pugach, M., & Wesson, C. (1995). Teachets' and stu-dents' views of team teaching of genetal education andleatning-disabled students in t\yo fifth-gtade classes.The Elementary School Journal, 95, 279-295.Ragin, C. C. (1987). The comparative method: Movingbeyond qualitative and quantitative Strategies. Betkeley:Univetsity of Califotnia Ptess.*Rice, D., & Zigmond, N. (2000). Co-teaching in sec-ondaty schools: Teachet tepotts of development in Aus-ttalian and Ametican classtooms. Learning DisabilitiesResearch and Practice, 15, 190-197.*Rosa, B. A. (1996). Cooperative teaching in the inclu-sive classtoom: A case study of teachets' petspectives.Dissertation Abstracts International, 57(7), 2968. (UMINO.AAG9638771)*Salend, S. J., Johansen, M., Mumpet, J., Chase, A. S.,Pike, K. M., & Dotney, J. A. (1997). Coopetativeteaching: The voices of two teachets. Remedial and Spe-cial Education, 18, 3—11.Sandelowski, M., Dochetty, S., & Emden, C. (1997).Qualitative meta-synthesis: Issues and techniques. Re-search in Nursing and Health, 20, 365-371.

Schofield, J. W. (1990). Increasing the genetalizabilityof qualitative teseatch. In E. W. Eisnet & A. Peshkini^As.), Qualitative inquiry in education: The continuingdebate (pp. 201-232). New Yotk: Teachets CollegePtess.Sctuggs, T. E., & Masttopieti, M. A. (1996). Teachetpetceptions of mainstteaming/indusion, 1958-1995: Ateseatch synthesis. Exceptional Children, 63, 59-74.Sctuggs, T. E., Masttopieti, M. A., & Casto, G. (1987).The quantitative synthesis of single subject teseatch:Methodology and validation. Remedial and Special Edu-cation, 8(2), 24-33.Sctuggs, T. E., Masttopieti, M. A., & McDuffie, K. A.(2006). Summatizing qualitative teseatch in special ed-ucation: Putposes and ptocedutes. In T. E. Sctuggs &M. A. Masttopieti (Eds.)'rAdvances in learning and be-havioral disabilities: Vol. 19. Applications of researchmethodology (pp. 325-346). Oxfotd, UK: Elseviet.Slavin, R. E. (1990). Cooperative learning: Theory, re-search, and practice. Uppet Saddle Rivet, NJ: Ptentice-Hall.Sttauss, A., & Cothin, J. (1998). Basics of qualitative re-search. Techniques and procedures for developing groundedtheory (2nd ed.). Newbuty Patk, CA: Sage.Swanson, H. L., & Sachse-Lee, C. (2000). A meta-analysis of single-subject-design intetvention teseatchfot students with LD. Journal of Learning Disabilities,33, 114-136.*Tattant, K. L. (1999). The coUabotative implementa-tion of an eatly litetacy cutticulum in a full-inclusionptimaty gtade classtoom: Co-teachets and studentswotking togethet to accomplish litetacy goals. Disserta-tion Abstracts International, 60 (10), 3599A. (UMI No.AA19948189)*Thompson, M. G. (2001). Captuting the phe-nomenon of sustained co-teaching: Petceptions of ele-mentaty school teachets. Dissertation AbstractsInternational, 63 (02), 560A. (UMI No. AAI3041129)*Ttent, S. C. (1998). False statts and othet dilemmas ofa secondaty genetal education coUabotative teachet: Acase study. Journal of Learning Disabilities, 31,303-513.Udey, C , Mottweet, S., & Gteenwood, C. (1997).Peet mediated insttuction and interventions. Focus onExceptional Children, 29(5), 1-23.Vaughn, S., Schumm, J. S., & Atguelles, M. E. (1997).The ABCDE's of co-teaching. TEACHING ExceptionalChildren, 30(2), 4-10.*Vesay, J. P. (2004). Linking petspectives and ptactice:Influence of eatly childhood and eatly childhood spe-cial educatots' petspectives of wotking collabotatively

Exceptional Children 4 1 5

in the integrated preschool classroom. Dissertation Ab-stracts International, 65 (03), 826A. (UMI No.AAI3126498)Waither-Thomas, C , Korinek, L., McLaughlin, V. L.,& Williams, B. T. (2000). Collaboration for inclusive ed-ucation: Developing successful programs. NeedhamHeights, MA: Allyn & Bacon.*Walther-Thomas, C. S. (1997). Co-teaching experi-ences: The benefits and problems that teachers andprincipals report over time. Journal of Learning Disabil-ities, 30, 395-408.*Ward, R. (2003). General educators' perceptions of ef-fective collaboration with special educators: A focusgroup study. Dissertation Abstracts International, 64(03), 861A. (UMI No. AAI3083896)Weiss, M. P. (2004). Co-teaching as science in theschoolhouse: More questions than answers. Journal ofLearning Disabilities, 37, 218-223.Weiss, M. P., & Brigham, F. J. (2000). Co-teaching andthe model of shared responsibility: What does the re-search support? In T. E. Scruggs & M. A. Mastropieri(Eds.), Advances in learning and behavioral disabilities:Vol. 14. Educational interventions (pp. 217-245). Ox-ford, UK: Elsevier.*Weiss, M. P., & Lloyd, J. L. (2002). Congruence be-tween roles and actions of secondary special educatorsin co-taught and special education settings. Journal ofSpecial Education, 36, 58-69.Welch, M., Brownell, K., & Sheridan, S. M. (1999).What's the score and game plan on teaming in schools?A review of the literature on team teaching and school-based problem-solving teams. Remedial and Special Ed-ucation, 20, 36-49.*Westberg, S. L. (2001). Implementing co-teaching as amodel of inclusion of students with mild learning dis-abilities in general education classrooms. DissertationAbstracts International, 62 (3), 977A. (UMI No.AAI3009383)*Wood, M. (1998). Whose job is it anyway? Educa-tional roles in inclusion. Exceptional Children, 64,181-195.*Yoder, D. I. (2000). Teachers' perceptions of elementsand competencies/characteristics that affect collabora-tive teaching at the secondary level. Dissertation Ab-stracts International, 61 (12), 4735A. (UMI No.AAI9996182)Zigmond, N. (1995). Inclusion in Pennsylvania: Edu-cational experiences of students with learning disabili-ties in one elementary school. Journal of SpecialEducation, 29, 124-132.

Zigmond, N., & Baker, J. M. (1994). Is the main-stream a more appropriate educational setting forRandy? A case study of one student with learning dis-abilities. Learning Disabilities Research dt' Practice, 9,108-117.*Zigmond, N., & Matta, D. (2004). Value added ofthe special education teacher on secondary school co-taught classes. In T. E. Scruggs & M. A. Mastropieri(Eds.), Advances in learning and behavioral disabilities:Vol 17. Research in secondary schools (pp. 55—76). Ox-ford, UK: Elsevier.

ABOUT THE AUTHORS

THOMAS E. SCRUGGS (CEC VA Federation),Professor of Special Education and Director,Ph.D. in Education Program; arid MARGO A .MASTROPIERI (CEC VA Federation), Professor,College of Education and Human Development,Ceorge Mason University, Fairfax, Virginia.KiMBERLY A. MCDUFFIE (CEC SC Federa-tion), Assistant Professor of Special Education,Eugene T. Moore School of Education, ClemsonUniversity, South Carolina.

Address correspondence to Thomas E. Scruggs,College of Education and Human Development,MSN 1D5, Ceorge Mason University, Fairfax,VA 22030 (e-mail: tscruggs@gmu.edu).

Manuscript received May 30, 2006; acceptedNovember 7, 2006.

INDEX OFADVERTISERS

Charles C Thomas, cover 2Chicago Public Schools, 474Council for Exceptional Children, 390, 434,487, 509Houghton Mififlin Education, cover 3MindPlay, Cover 3The Special Education Service Agency (SESA),510

4 1 6 Summer 2007