problematic preferences? a mixed method examination of principals' preferences for teacher...

http://eaq.sagepub.com/Quarterly

Educational Administration

http://eaq.sagepub.com/content/49/1/52The online version of this article can be found at:

DOI: 10.1177/0013161X12451025

July 2012 2013 49: 52 originally published online 3Educational Administration Quarterly

Mimi EngelPrincipals' Preferences for Teacher Characteristics in Chicago

Problematic Preferences? A Mixed Method Examination of

Published by:

http://www.sagepublications.com

On behalf of:

University Council for Educational Administration

at: can be foundEducational Administration QuarterlyAdditional services and information for

http://eaq.sagepub.com/cgi/alertsEmail Alerts:

http://eaq.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

at UNIV OF WINDSOR on November 3, 2014eaq.sagepub.comDownloaded from at UNIV OF WINDSOR on November 3, 2014eaq.sagepub.comDownloaded from

http://eaq.sagepub.com/

http://eaq.sagepub.com/content/49/1/52

http://www.sagepublications.com

http://www.ucea.org

http://eaq.sagepub.com/cgi/alerts

http://eaq.sagepub.com/subscriptions

http://www.sagepub.com/journalsReprints.nav

http://www.sagepub.com/journalsPermissions.nav



What is This?

- Jul 3, 2012OnlineFirst Version of Record

- Jan 15, 2013Version of Record >>

at UNIV OF WINDSOR on November 3, 2014eaq.sagepub.comDownloaded from at UNIV OF WINDSOR on November 3, 2014eaq.sagepub.comDownloaded from

http://eaq.sagepub.com/content/49/1/52.full.pdf

http://eaq.sagepub.com/content/early/2012/06/29/0013161X12451025.full.pdf

http://online.sagepub.com/site/sphelp/vorhelp.xhtml



Educational Administration Quarterly49(1) 52 -91

© The University Council forEducational Administration 2013

Reprints and permission: http://www. sagepub.com/journalsPermissions.nav

DOI: 10.1177/0013161X12451025http://eaq.sagepub.com

451025 EAQ49110.1177/0013161X12451025EngelEducational Administration Quarterly© The University Council forEducational Administration 2013Reprints and permission: http://www. sagepub.com/journalsPermissions.nav

1Vanderbilt University, Nashville, TN, USA

Corresponding Author:Mimi Engel, Department of Leadership, Policy, & Organizations, #414 Peabody College, Vanderbilt University, Nashville, TN 37203-5721, USA Email: [email protected]

Problematic Preferences? A Mixed Method Examination of Principals’ Preferences for Teacher Characteristics in Chicago

Mimi Engel1

Abstract

Purpose: Relatively little is known about how principals make decisions about teacher hiring. This article uses mixed methods to examine what char-acteristics principals look for in teachers. Research Methods: Data were gathered using a mixed method approach, including in-depth interviews with a representative sample of 31 principals as well as an online survey of 368 principals in the Chicago Public Schools. Data analysis techniques included both qualitative analysis to look for patterns and themes and exploratory regression analysis of surveys to examine variation in preferences across school and principal characteristics. Findings: Results indicate that princi-pals focus on behaviors and skills rather than qualifications. Principals report looking for teachers who care about students, have content knowledge, are willing to go beyond contractual obligations, and have classroom management skills. Principals, in general, talk extensively about caring, classroom manage-ment, and willingness to “give extra,” while most say little about content knowledge or teaching skills. Preferences vary substantially across low- and high-achieving schools. Implications: Whether the skills and behaviors that

Article

at UNIV OF WINDSOR on November 3, 2014eaq.sagepub.comDownloaded from


Engel 53

principals report focusing on during the teacher-hiring process are those that will benefit their students most is an important question, particularly because even ineffective teachers are rarely dismissed from public schools.

Keywords

teacher hiring, teacher characteristics, principals’ preferences, urban schools, contextual variation

A lot of people say oh, hire for attitude because we can train for every-thing else. Not in teaching. You can’t get somebody to be an intellect if they’re not. You can’t get somebody to be a reader, keep up on cur-rent events . . . unless they’re already in that . . . Initially, we thought we could hire for attitude, and bring along someone’s intellect. Because we really want people to be able to manage kids’ behavior, and therefore be successful at teaching. But gosh, if you can manage kids’ behavior but then you’re not saying anything, you’re not doing anything, you don’t ask higher order questions, you don’t know how to ask higher order questions, you don’t really understand that . . . now . . . we’re like, okay, let’s look at our candidates, and how bloody smart are they? (Principal Ames, October 2, 2006)

Principal Ames,1 a Chicago principal, describes the qualities she looks for when hiring teachers for her school. She explains that when she first became a principal, her primary focus when hiring teachers was on classroom man-agement, but that over time her perspective has shifted and she now looks for content knowledge and teaching skills. Implicit in this principal’s description of what she looks for in teachers is that she is making a trade-off—focusing on teaching and learning over discipline and control—when selecting teachers for her school.

Selecting wisely from the available supply of teachers is one of the most crucial dimensions of a principal’s job. Poor selection decisions can have detrimental results, particularly because in many districts it can be hard to dismiss even probationary teachers. Even when there are few barriers to fir-ing teachers, teachers are rarely dismissed (Jacob, 2010).

The notion that teachers make a large contribution to the academic achieve-ment of their students is widely accepted by the general public and has been empirically validated by education researchers (e.g., Aaronson, Barrow, & Sander, 2007; Nye, Konstantopoulos, & Hedges, 2004; Rockoff, 2004). But, while we know that teachers are important, our ability to predict, based on



54 Educational Administration Quarterly 49(1)

qualifications and credentials, whether an individual teacher will excel in the classroom is limited. Furthermore, there is a small but growing body of evidence suggesting that principals themselves have a somewhat limited ability to identify the teachers in their schools who are successful in raising student achievement (e.g., Jacob & Lefgren, 2008). We also know relatively little about what characteristics principals and administrators look for in teachers during the hiring process (Harris, Rutledge, Ingle, & Thompson, 2010).

The analyses that follow examine principals’ hiring preferences using both qualitative interviews and surveys of Chicago Public Schools (CPS) princi-pals. Although studies of teacher labor markets have increased over the past decade, much of the focus has been on supply or how teachers make deci-sions about entering and exiting teaching, and where they choose to teach (Harris et al., 2010; Jacob, 2007). The analyses presented here focus instead on the demand side of the teacher labor market or how principals and districts make hiring decisions. Specifically, the study seeks to answer the following questions:

What qualities and characteristics do principals look for when hiring teachers?

Why do they look for these qualities?How do they ascertain whether teachers have the qualities they prefer?And, do principals’ preferences for teacher characteristics vary system-

atically across schools?

This study adds to research on the demand for teachers in general and to research on what characteristics principals look for in teachers. It is the first study to examine principals’ preferences in a large urban district, and the first to use a mixed method approach including both detailed interviews and a larger-scale survey. Interview results provide information about what princi-pals look for in candidates, why they report particular preferences, and what strategies they use to determine whether prospective teachers possess the qualities they seek. Survey results provide the opportunity to extend and con-firm qualitative information regarding how principals’ preferences vary by school composition and principal characteristics, finding distinct preferences among principals in particular contexts.

Prior ResearchUnderstanding principals’ preferences for teacher characteristics is impor-tant, particularly because teacher hiring is one of the most likely paths



Engel 55

through which we might expect school leaders to influence student achieve-ment. Using High School and Beyond, Brewer (1993) found that principals have an effect on student achievement through the role that they play in selecting teachers. Strauss (2003) found few direct links between school administrators and student achievement but speculated that administrators affect achievement through hiring decisions. More recently, analysis of Florida data on the association between principals’ skills and time use and school-level student achievement found that spending more time (Horng, Klasik, & Loeb, 2010) or being more skilled (Grissom & Loeb, 2011) in areas categorized as organizational management—including hiring personnel—as opposed to administrative tasks or instructional monitoring, was posi-tively associated with student test score outcomes (Grissom & Loeb, 2011; Horng et al., 2010).

Can Principals Identify Quality?An important question, then, is the extent to which principals are able to identify effective teachers when hiring. In a laboratory experiment examin-ing principals’ evaluations of videotaped teacher interviews, Emley and Ebmeier (1997) found that principals appeared to be able to identify strong and weak candidates.

Related research has examined whether principals are able to recognize the quality of the teachers in their own schools. Recent studies have explored whether principals can identify the relative effectiveness of their teachers by comparing principals’ assessments with value-added measures of teacher effectiveness (Harris & Sass, 2009; Jacob & Lefgren, 2008). Both studies found that principals’ subjective ratings were correlated with value-added. Jacob and Lefgren (2008) found that while principals seemed reasonably good at identifying teachers who produce the largest and smallest gains, they were unable to distinguish among teachers in the middle range of the distribution.

What Teacher Characteristics Are Related to Teacher Quality?Studies estimating the effect that teachers have on their students’ academic achievement have consistently found that teachers have a large effect on student achievement and that teacher effectiveness varies substantially (Aaronson et al., 2007; Hanushek, 1986, 1997; Hanushek, Kain, O’Brian, & Rivkin, 2005; Harris & Sass, 2008; Kane, Rockoff, & Staiger, 2008; Nye et al.,




2004; Rockoff, 2004). However, attempts to link observable characteristics to teacher effectiveness have shown relatively few traits to be predictive of teacher quality (Aaronson et al., 2007; Hanushek, 1986, 1997; Hanushek et al., 2005; Kane et al., 2008; Nye et al., 2004; Rockoff, 2004), finding little or no relationship between teachers’ years of education and teacher effective-ness (Hanushek, 1986, 1997; Hanushek et al., 2005; Nye et al., 2004). Findings with regards to teacher certification or licensure have been mixed, with some finding small positive effects (e.g., Clotfelter, Ladd, & Vigdor, 2006, 2007a, 2007b; Goldhaber & Brewer, 2000) and others finding no relationship (Hanushek et al., 2005; Kane et al., 2008).

The handful of teacher characteristics that have been more consistently linked to teacher effectiveness include teaching experience, with teachers generally found to be less effective in their first few years of teaching (Clotfelter et al., 2006, 2007a, 2007b; Hanushek et al., 2005; Harris & Sass, 2008; Nye et al., 2004; Rockoff, 2004). Teachers’ ability as measured by test scores, IQ scores, and the selectivity of the colleges and universities they attend has also been shown to have a small positive relationship with test score gains of their students (Clotfelter et al., 2007a, 2007b; Ehrenberg & Brewer, 1994, 1995; Hanushek, 1997). Furthermore, a growing number of studies have found links between teachers’ knowledge of content and peda-gogy and the achievement of their students (Croninger, Rice, Rathbun, & Nishio, 2007; Harris & Sass, 2008; Hill, Rowan, & Ball, 2005).

Thus, while we know that teachers are the most important school-related determinant of student learning (e.g., Aaronson et al., 2007; Rockoff, 2004), there are only a handful of teacher characteristics—experience, a teacher’s own ability as measured by test scores or college selectivity, and perhaps content and pedagogical knowledge—that we know consistently predict stu-dent achievement. It is interesting to consider both whether principals seem to value these traits and what other qualities principals focus on when hiring teachers to fill vacancies in their schools.

What Do Principals and Administrators Look for in Teachers?Relatively little is known about how principals and administrators make hiring decisions. However, understanding this process is important, particu-larly because studies have suggested that principals do not hire the “best” teachers (Ballou, 1996; Ballou & Podgursky, 1997; Pfuam & Abramson, 1990). For example, Ballou (1996) found that teachers who attended more selective colleges and universities were not more likely than job candidates



Engel 57

who attended less selective institutions to be hired to fill teaching vacancies. An analysis of the 1993-1994 Schools and Staffing survey by Baker and Cooper (2005) provided further evidence that most principals appear to put little emphasis on a candidate’s academic background, and that this may vary by principals’ own academic backgrounds. They found that principals who attended more selective universities themselves were more likely to employ teachers who attended selective universities and that this relation-ship was particularly pronounced in high-poverty schools.

Few studies have tried to identify what teacher characteristics principals and administrators prefer (Harris et al., 2010; Liu & Johnson, 2006). Statewide surveys of district-level administrators in New York (Balter & Duncombe, 2005a, 2005b, 2006) and Pennsylvania (Strauss, Bowes, Marks, & Plesko, 2000) examined administrators’ hiring preferences. They found that district officials emphasized a candidate’s undergraduate major, prior teaching expe-rience, and subject matter knowledge, but did not focus on the caliber of candidates’ academic institutions in the hiring process.

Harris et al. (2010) reviewed the literature on principals’ preferences. The authors noted that in addition to the sparse literature on the subject, other limitations included disparate methodological approaches and the fact that few characteristics were repeated across studies (all the studies asked prin-cipals to rank prespecified lists of characteristics). Despite these problems and limitations, the authors reported that previous research had found that principals reported preferring teachers who were enthusiastic and had strong communication skills. Less consistently, studies reported that principals pre-ferred teachers with particular teaching skills or knowledge. For example, surveying a random sample of Virginia high school principals, Dunton (2001) found that principals sought candidates who were enthusiastic, were good communicators, were student focused, had classroom management skills, and had knowledge of teaching strategies. Other research has sug-gested that principals and administrators prefer younger candidates (Young & Allison, 1982).

In their own study of principal preferences, Harris et al. (2010) inter-viewed 30 principals from a midsized Florida school district. Based on an analysis of open-ended responses, they found that principals most often reported strong teaching skills, caring, knowledge of subject, ability to work with others, experience, enthusiasm, and communication skills as the most important characteristics they considered in applicants for teaching positions.

In one of the only studies of principals’ preferences that compares princi-pals across urban and suburban districts, Papa and Baxter (2008) surveyed a




random sample of principals in New York State. The authors found that prin-cipals in urban schools rated the importance of a prospective teaching candi-date’s performance both in college and as a classroom teacher much lower than principals in suburban and rural schools.2

Data and MethodThe CPS was selected for this study because district administrators, spe-cifically the Department of Human Resources, and school principals were open to participating. With nearly 600 schools employing 25,000 teachers and serving more than 400,000 students, the CPS is an ideal district for studying the preferences of principals who work in a diverse context serving a large number of disadvantaged students. More than 90% of CPS students are non-White, and more than 85% come from low-income families. The CPS also provides an opportunity to study a population of principals and teachers that is highly diverse in terms of race/ethnicity, with more than two thirds of principals and half of teachers reported as being either African American or Hispanic.3

Despite the fact that the CPS serves a large population of disadvantaged students, the district still receives many applications, with approximately 10 applicants for each teaching position (Jacob, 2007). In 2006, nearly 20,000 individuals applied to teach in the CPS. More than half of the 2006 appli-cants to the district were from outside Chicago or out-of-state, and more than 40% attended undergraduate institutions with Barron’s ratings of very competitive or higher (Engel & Jacob, 2011).

Principals of CPS schools have autonomy from the central office in deci-sions related to teacher hiring. They are able to choose which candidates they will interview and make job offers to, and also independently choose when, where, and how to go about finding prospective teachers for their schools. Despite high levels of autonomy in the decision-making process, the district makes available an array of supports and resources to principals who are hiring. The CPS Department of Human Resources (HR) maintains a large database of job applicants that can be accessed by all principals. While principals choose the teachers for their schools, once their job offer has been accepted, HR is charged with processing new teachers and com-pleting the hiring process.

While principals are able to decide how they will find and hire teachers to fill vacancies in their schools, HR has developed a large number of recruit-ing events and services aimed at expanding the supply of teachers in the district. These include district attendance at college and university job fairs



Engel 59

throughout the Midwest, inviting qualified candidates to visit Chicago and leading them on tours of Chicago neighborhoods, and implementing multi-ple, large-scale job fairs attended by thousands of prospective teachers and hundreds of CPS principals and administrators. Below, data collection meth-ods for the qualitative interviews are described, followed by information on how the survey was fielded.

Interview DataSample. The principal sample was selected using two methods. Twenty-

six schools are part of a stratified random sample. Schools were stratified by region and level (elementary and secondary). The regional strata capture variation in both racial/ethnic and achievement composition of schools. I excluded schools that had no vacancies for the 2006-2007 academic year (the year when interviews were conducted) to increase the likelihood that all principals in the sample would be able to talk about their hiring experiences for the current academic year. Special education schools, alternative schools, and schools that had formed within the past year were also excluded. Addi-tionally, six schools were sampled purposively for piloting and to insure that charter schools as well as some of the district’s top performing schools would be represented. Two higher performing elementary schools were included as pilot schools for testing the interview protocol. Two charter schools and two top performing CPS schools (one elementary and one high school) were included as well. In two high schools, the interviewees were assistant principals instead of principals. In both cases, the assistant princi-pals had primary responsibility for hiring teachers.

Demographic characteristics for schools in the analytic samples and all CPS schools are shown in Table 1. The qualitative sample is very similar to the CPS population in terms of student achievement and principal gender (columns 1 and 2). The proportion of high schools is higher in the sample than in the population of CPS schools (34% vs. 15%) because high schools were purposefully oversampled to include a large enough number (n = 11) to allow the comparison of hiring practices in high schools and elementary schools. The sample includes 14 schools that are below the 2005 median CPS school-level achievement in reading, which is how low-achieving schools are defined in the analyses that follow. Furthermore, 10 sample prin-cipals are male, 16 are African American, 11 are White, and 4 are Hispanic.

Interviews. The author conducted semistructured qualitative interviews about the teacher hiring process with 31 principals during September and October 2006, (one principal of the original sample of 32 refused to participate).




Table 1. Demographic Characteristics for Qualitative Sample, All Chicago Public Schools (CPS), and Comparing Principals/Schools That Attended Summer 2006 Job Fairs and Registered Online (Thus Completing Surveys) and Those That Did Not

(1) (2) (3) (4)

Qualitative

Sample (n = 31)All Schools (N = 627)

Surveyed (n = 368)

Not Surveyed (n = 259)

Mean or Proportion

School demographic characteristics Proportion of CPS school

population0.05 1.0 0.59 0.41

High school 0.34 0.15 0.18 0.13 Below full sample median

achievement0.45 0.50 0.54 0.43a

School-level race Predominantly Black school 0.5 0.43 0.47 0.37a

Predominantly Hispanic school 0.06 0.15 0.13 0.17 Predominantly minority or racially

mixed school0.34 0.27 0.24 0.30

Integrated school 0.09 0.09 0.09 0.09 CPS region Region 1 0.29 0.19 0.17 0.22 Region 2 0.13 0.14 0.10 0.19a,b

Region 3 0.19 0.14 0.17 0.09a

Region 4 0.13 0.17 0.18 0.16 Region 5 0.06 0.13 0.13 0.13 Region 6 0.19 0.18 0.18 0.17Principal demographic characteristics Principal race African American 0.55 0.49 0.56 0.39a,b

White 0.23 0.27 0.26 0.29 Hispanic 0.13 0.14 0.12 0.18a

Male principal 0.32 0.30 0.29 0.31 Average principal age 55 54 52 56a,b

Principal education/ability MA 0.89 0.85 0.91 0.93 PhD 0.07 0.11 0.11 0.13 Mean BA Barron’s rating 2.08 2.16 2.21 2.07 Mean MA Barron’s rating 1.95 2.22 2.26 2.15 Failed at least one certification

exam0.06 0.06 0.07 0.04

Note. Actual sample size varies because of missing data for individual variables.a. Significant difference (p < .05) between surveyed (column 3) and not surveyed (column 4).b. Joint significance test examining if differences are larger than chance, given the number of comparisons (using a Bonferroni–Holm family-wise error rate), finds three statistically signifi-cant differences (p < .05)—African American principal, age, and Region 2.



Engel 61

In these interviews, which lasted from 40 to 90 minutes (1 hour, on average), principals were asked to describe what characteristics they look for when hir-ing teachers. Furthermore, principals were asked to explain how they deter-mine whether candidates have these characteristics and why they believe they are important. Interview questions were open-ended, and principals were encouraged to add information that they felt was important to their hiring pro-cess even if it had not been asked about directly. Interviews covered a range of topics related to teacher hiring (e.g., principals were also asked about the tim-ing of their teacher hires and how they go about finding prospective teachers). The portion of the interview protocol that included questions about principals’ preferences for teacher characteristics can be found in Appendix C.

Analysis. Principal interviews were transcribed verbatim, and full interview transcripts were read to look for themes in principals’ perceptions and experi-ences and to help describe the hiring process in general, as well as how it varied across principals in different types of schools (e.g., primary vs. sec-ondary, high vs. low achieving). Principals’ responses were sorted topically into broad categories representing both the subjects discussed during inter-views and the overarching themes common across interviews.

Principals described the characteristics they look for when hiring new teachers, and discussed whether and how they are able to tell if candidates have those qualities. Principals were also asked to discuss how they think about previous teaching experience and whether they prefer hiring experi-enced or inexperienced teachers.

Any conversations with principals about what they look for in teachers, what characteristics they prefer when hiring, and what qualities they believe make a good teacher were coded as “teacher characteristics.” Under the teacher characteristics heading, there were more than 15 subcodes for teacher qualities and characteristics mentioned multiple times by principals. Within the subcoded areas, various themes emerged that helped explain why princi-pals value a particular characteristic, or how they determine whether a candi-date has that quality.

Qualitative interviews provided detailed information about what princi-pals look for in teachers, why they prefer the qualities they do, and how they determine whether candidates have what they are looking for. Interview results also suggested the possibility that principals’ preferences for teacher characteristics vary systematically across schools. Surveys provided the opportunity to examine whether interview results regarding principals’ pref-erences and contextual variation across schools would generalize to a larger sample of CPS principals




Surveys

Sample. A total of 368 CPS principals were surveyed, and the response rate was 100% among principals of traditional CPS schools who registered for at least one of the Summer 2006 CPS job fairs online.4 The sample includes 59% of all CPS principals. Table 1 (columns 3 and 4) shows descriptive sta-tistics for the 368 principals who registered online for the 2006 summer job fairs, and the 259 principals who did not attend fairs or failed to register online. The table shows that the two groups are similar on a number of school- and individual-level demographic characteristics, although on aver-age those who attended fairs were principals at somewhat lower achieving schools that were more likely to serve predominately African American students.

Survey. The survey was administered electronically through the district’s intranet system from April to July 2006 as part of the online registration pro-cess for the three large-scale CPS job fairs held during the summer of 2006. It was designed in collaboration with administrators who oversee teacher recruitment efforts within the CPS HR department, as they have extensive experience with the teacher hiring process. Previous research on what teacher characteristics principals prefer was also used to develop the survey. As the survey was designed and administered in collaboration with HR and was embedded in the online registration system, principals completed it automati-cally while registering for the job fairs.

The survey asked principals to both rate and rank their preferences for teacher characteristics from a list of 15 teacher qualities. Respondents rated the importance of each characteristic on a 5-point Likert-type scale (rang-ing from not at all important to extremely important). They also ranked their top three characteristics from the list of 15. Items on the survey included characteristics such as the selectivity of the candidate’s under-graduate or graduate institution, years of education, years of teaching expe-rience, as well as items about the candidate’s classroom management skills, ability to improve student achievement, enthusiasm for teaching, and abil-ity to communicate. The full list of survey items is shown in Appendix D and in Table 3.

Analysis. Principals’ survey responses were linked to administrative data describing their demographic characteristics and educational backgrounds and then analyzed using ordered logistic regressions to explore the extent to which principal and school characteristics predict principals’ self-reported preferences.



Engel 63

Principals’ survey responses are modeled as a function of their demo-graphic characteristics as well as the characteristics of their schools and students:

Teach Prefi = a

i + β

1Principal

i+ β

2School

i + β

3Student

i + e

i,

where Teach Prefi is the survey response (to one of the 15 survey items) of

the ith principal; Principali is the collection of demographic characteristics

for principal i; Schooli represents characteristics of principal i’s school;

Studenti represents the average characteristics of students within school i;

and ei is a stochastic error term. This analysis examines how characteristics

such as school-level student achievement and principal educational back-ground predict principal responses to survey items. All independent variables are included in each model with the intent of isolating the relationship between school and principal characteristics and principals’ preferences.

The principal characteristics included as independent variables in explor-atory regressions are principal age, gender, race/ethnicity, selectivity of undergraduate and graduate institutions attended, and whether the principal has a record of having failed any certification exams. Selectivity of princi-pals’ undergraduate and graduate institutions is measured using ratings from the Barron’s Profiles of American Colleges. The measures included here are on an ordinal scale ranging from 1 to 5, with 5 representing the most competi-tive institutions.

School-level variables used to predict principal survey responses include the racial and achievement composition of the school, level (elementary vs. secondary), and region (the CPS is divided into six geographic regions). Missing data on all covariates are handled using a set of missing data dummy variables. See Table 1 for a list of principal and school demographic charac-teristics and descriptive statistics.

ResultsAs is noted above, qualitative interviews provided rich information about what principals look for in teachers, why they prefer the qualities they do, and how they ascertain whether an applicant has what they are looking for. Following the presentation of qualitative results, survey results extend and confirm qualitative analyses, examining principals’ preferences and contex-tual variation among a large sample of CPS principals. As qualitative data were analyzed first and guided analysis of survey results, qualitative results are presented first below, followed by survey findings.




Interviews: What Do Principals Look For?

When asked to describe what characteristics they look for when hiring teachers, principals in the study identified an array of skills and behaviors. The four most common qualities that principals mentioned were that teach-ers care about children, have content knowledge, are willing to give extra time or bring something extra to the classroom, and have classroom man-agement skills. Below, I describe these four qualities and explore what principals say about them in depth. I also examine differences in principals’ preferences across low- versus high-achieving schools and elementary ver-sus high schools. Table 2 shows the frequency with which principals men-tioned various characteristics overall and for subgroups by school level and achievement.

Interestingly, principals’ preferences vary systematically across low- and high-achieving schools. Among principals in low-achieving schools, the three characteristics mentioned most often were classroom management, car-ing, and willingness to do or give extra. Principals in high-achieving schools, on the other hand, were most likely to mention the importance of content knowledge, caring, and teaching skills. Appendix A shows qualitative responses for additional subgroups, including principal race, gender, and experience.

Caring about children. The modal response among principals, 61% of the sample (n = 19), was that it is essential that teachers understand, love, or care about children. A succinct characterization of this ideal was provided by two principals who, when explaining that caring is an essential quality in a suc-cessful teacher, mentioned the adage kids don’t care how much you know until they know how much you care. Principals who said they look for teach-ers who care about or understand children noted that they believe teachers lacking this quality would be unable to reach their students.

The most important quality to me, the first one is always do they care about kids? . . . Therefore my biggest thing is I try to find out do they care about kids? What else have they done in their life that interacts them with kids? Have they worked at a day care, coached Little League, worked at the Rec Center? What did you do in some of your work study jobs at school? Things like that, that keeps them involved with kids. Because they have to sincerely like children to be effective. Because you have to understand the adolescent, you have to under-stand the environment in which they live in, because it’s so diverse now. Our kids come with many different obstacles. And if you’re not



Engel 65

compassionate towards the roads that they travel, it’s going to be a problem. (Principal Porter, September 21, 2006)

This principal believes that a teacher needs to understand the challenges students face and have empathy for them in order to be effective. In addition to reviewing resumes and asking about prior experience with children, as Principal Porter describes above, principals noted that they also listened to whether and how candidates talked about children when assessing whether prospective teachers cared about children and were committed to working with them.

This emphasis on caring is, at least in part, a function of the student popu-lation that many urban schools serve. That is, some principals who reported that whether teachers care about children was a key selection criterion also suggested that this characteristic is particularly important in contexts where the student population is disadvantaged.

Well they have to have a certain amount of empathy for children, because here we’re 100% African American, low income. And they have to really love kids, and really realize that there’s more to a child than just what you see. They bring a lot of baggage with them when they come. And the way to get to these kids is not by yelling and screaming. It’s by talking quietly and trying to understand what’s really going on. Because they’re used to being yelled at all the time. So when you start yelling, they stop listening. But if you sit down and make eye contact and say okay, now tell me what’s really—I’m really concerned about you. Tell me what’s really going on. Then they start to open up, and then they understand that you’re a human being too. And that changes their behavior. (Principal Franklin, September 26, 2006)

This principal indicates that to be effective in her school, where the stu-dent body is particularly disadvantaged, teachers need to be sensitive to the environments from which the children come and need to be able to commu-nicate to those children that they care about them.

Principals identified three aspects of caring: understanding children in general as well as the circumstances of the specific population served by the school, having empathy for children, and enjoying working with children. Nearly three quarters (71%) of principals in low-achieving schools




mentioned this quality, compared with 53% of principals in high-achieving schools. African American principals were two times more likely than their White counterparts to discuss the importance of caring (Appendix A, 75% vs. 36%, respectively). This pattern suggests the possibility that principals of schools serving more disadvantaged student populations put more emphasis on caring.

Content knowledge and teaching skills. More than half (n = 17) of principals reported that content knowledge is an important characteristic for the teach-ers they hire. Interestingly, although content knowledge was mentioned often, principals rarely elaborated on it. One principal’s perspective on content knowledge might help to explain why, despite the fact that it was mentioned often, content knowledge was rarely discussed in detail.

Respondent: Extracurriculars, because content knowledge, that’s a given.

Interviewer: Content knowledge is a given?Respondent: Yeah, it’s a given.Interviewer: It’s a given that it’s important, or it’s a given that they

have it?Respondent: A given that they have it. I mean, they have the degree,

they’ve been to school, most people have Masters, you know, they have that, yeah. (Assistant Principal Wilson, September 13, 2006)

This assistant principal, who is in charge of hiring in mathematics and sci-ence for her high school, equates educational background with content knowledge and thus focuses on other criteria.

Of the 17 principals who mentioned the importance of content knowledge, only 7 elaborated in any way about its importance, and only 2 of these 7 (one quoted at the beginning of this article) discussed in detail why content knowl-edge matters. Responses among the 5 principals who briefly elaborated on the importance of content knowledge were limited.

What kind of knowledge they have, especially as we start hiring for upper grades, we’re looking to find people that have expertise in con-tent areas. Even if we’re not hiring them, say, for a science position or a math position, you know we have a large school, and things change. (Principal Healey, September 12, 2006)

Although this principal notes that content knowledge in the upper grades is important, she does so only after mentioning the importance of



Engel 67

the candidate being interested in and committed to her school and being willing to go beyond contractual obligations. There was a small differ-ence across elementary and high schools in principals’ preferences for content knowledge: 50% of elementary school principals indicated con-tent knowledge as a teacher characteristic they sought versus 64% of high school principals.

Although teaching skills/teaching style ranks fifth across principals in terms of how often it was mentioned as an important characteristic, it is interesting to consider teaching skills along with content knowledge, as these are both critical to the process of teaching and learning. Forty-two percent of principals mentioned that they look for candidates with good teaching skills or with a particular teaching style. While teaching was mentioned relatively often by sample principals, similar to content knowl-edge, most principals did not elaborate on why they care about teaching skills or how they recognize if a candidate has them. Among the 13 prin-cipals who mentioned the importance of teaching skills, 6 went beyond simply naming this characteristic, with 3 providing sentence or two long explanations, and the other 3 providing more information. An example of a principal who detailed the importance of teaching skills and student learning follows.

We may ask you, tell us about you as a teacher, what do you think about yourself as a teacher? What is your philosophy of education? And they tend to describe. Then we look at flexibility, and from their answers we know, we can tell if this person’s very flexible . . . Whether they understand that all students don’t learn the same way . . . Sometimes we go to theories of learning, for instance we use Vygotsky, Piaget quite often, and that’s cooperative learning, and versus vicarious learning, so we have them respond to that. And then we get an idea as to what their instructional practice is about. (Principal Jackson, September 12, 2006)

Unlike the majority of principals in the sample, including those who did mention or briefly discuss teaching skills, Principal Jackson describes, in detail, how he probes during interviews to glean information about a teach-er’s teaching style and philosophy.

Most principals who mentioned teaching skills or teaching style, however, did so in a cursory manner. This may be because instruction falls lower on the priority list of some principals.




The more obvious [characteristics sought] are the number of years, the skills, how they will resolve problems in the classroom. We ask a vari-ety of questions such as behavior. How do you establish a connection with your students? And that’s always tough for people when they interview, because they don’t know what we’re talking about. So we have to explain, you are the teacher: you’re a counselor, you’re a friend, you’re a mom, you’re a dad, you’re a social worker, and then you’re an educator. (Principal Reynolds, September 11, 2006)

During the conversation that preceded the above quote describing what she looks for in a teacher, this principal mentions passion for teaching, classroom management skills, content knowledge, the ability to give extra, and several other characteristics. While Principal Reynolds men-tions content knowledge, she does so only briefly. In fact, the passage above suggests that she believes that, for the population of students served by her school, caring and social support are the most critical aspects of a teacher’s job.

Giving extra. Principals also talked about looking for teachers who are will-ing to do work that goes beyond teaching when making hiring decisions. More than half of the sample (n = 16) mentioned that when interviewing teachers they look for candidates who indicate that they will provide services in addition to the classroom teaching during school hours to which they are contractually obligated. Giving extra includes being available to students before and after school and being involved in extracurricular or community activities. Principals identified several different means for determining whether candidates would give extra as teachers.

Then we also ask questions about schedule. We know that our—the contract for teachers calls for them to come in at 8:00, 8:30, 9:00—And in an ideal world, all you have to work is six hours. Well anyone who’s been in a classroom, you know you can’t do it in six hours. You know you’re going to have to take some work home. You know that in a school like [school name], for example, we have after school activities, we have school community activities, we have meetings—everything is in the evening. (Principal Reynolds, September 11, 2006)

This principal noted that she asks candidates about their schedules to determine whether they have the availability that she expects of teachers in her school.



Engel 69

As Table 2 indicates, there were large and statistically significant differ-ences across principals of low- and high-achieving schools in terms of the frequency with which they mentioned the importance of giving extra time. While 64% of principals in low-achieving schools mentioned giving extra, only 29% of principals from high-achieving schools did so.

Classroom management. Nearly half of principals (n = 14) mentioned classroom management as an important teacher quality. Principals who talked about the importance of classroom management skills often noted that without them, a teacher’s other characteristics were of little or no value. The following quotation exemplifies both the importance of classroom man-agement skills and shows that for some principals, these skills supersede other qualities that teachers might bring to the job.

First of all I’m looking for someone who knows how to manage a classroom, because many of the new hires that I have hired over the last 10 years, they’re good academically, some of them have been 4.0, you know, 4.0 and so forth, but when it comes to teaching, they cannot teach because they don’t have the skills to deal with classroom man-agement, and that’s key. And our kids—I’m just coming from a class-room now, someone I just hired. They are afraid to tell children to sit and learn. Those same set of children can go to the next classroom and you won’t even know it’s an issue. (Principal Clark, October 3, 2006)

It is interesting to contrast this quotation from Principal Clark with the one that begins this article. Principal Ames reports that she has learned to value content knowledge, teaching skills, and intellect above classroom manage-ment; believing that she can develop the later, but not the former. Principal Clark, on the other hand, indicates that without classroom management skills, she believes a teacher will not succeed. Thus, she prioritizes classroom man-agement over knowledge or intellect.

Principals reported several means for determining whether applicants had classroom management skills. One way that principals tried to shed light on candidates’ skills in this area was to ask candidates to respond to scenarios.

It’s hard to spot the classroom management. Sometimes they’ve memorized these good answers. So we tend to give them cases. The last part of the interview has a couple of cases, where we have actual, live, real things that have happened to us. Let’s say a student comes in drowsy, and the child looks intoxicated. How would you manage that? Because that happened to us, and the child . . . didn’t get enough sleep.




Table 2. Percentage of Sample Principals Who Reported (During Qualitative Interviews) That Particular Teacher Qualities Were Important to Them During the Hiring Process, Overall and by School Achievement and Level

(1) (2) (3) (4) (5)

Overall Percentage

(n = 31)High Achieving

(n = 17)Low Achieving

(n = 14)Elementary

(n = 20)High School

(n = 11)

Characteristic Care about/understand

children61 53 71 60 64

Content knowledge 55 59 50 50 64 Do/give extra 52 29 64* 55 45 Classroom management 45 24 71* 55 27 Teaching skills/teaching

style42 47 36 40 45

Passion/enthusiasm for teaching

39 35 43 40 36

Team player 26 35 14 20 36 Committed to

professional development

23 18 29 30 9

High expectations of students

16 18 14 15 18

Previous exposure to diversity

16 18 14 0 45*

Credentials 13 24 0* 15 9 Technological proficiency 10 6 14 10 9 Flexibility 10 18 0† 5 18 Good communicator 10 18 0† 0 27*Experience Prefers experienced

teachers13 6 21 20 0*

Considers experience of current faculty

32 41 21 30 36

Note. Only qualities and characteristics mentioned by three or more principals are included in table. High- and low-achieving schools are defined as being above or below the median (Chicago Public Schools) achievement in 2005. † and * indicate statistically significant differences between high- and low-achieving schools or between elementary and high schools.†p < .10.*p < .05.

So we need to ask the pertinent questions, call home, find out. And those were tough. (Principal Reynolds, September 11, 2006)

A number of principals reported similar methods for learning about can-didates’ classroom management skills during interviews. In addition to cases or scenarios, principals reported that they had applicants describe their



Engel 71

classroom management plans, and also asked them to describe their own successes and failures in this area.

The most dramatic difference between principals in high- and low-achieving schools in terms of what teacher characteristics they value is the relative importance of classroom management skills. As Table 2 shows, among princi-pals of low-achieving schools, classroom management skills ranked first, along with caring about children, with 71% of principals mentioning it. In contrast, only about a quarter (24%) of principals in high-achieving schools mentioned the importance of this characteristic, and it ranked seventh. This difference is statistically significant (p < .05). Elementary school principals were more likely to be concerned with classroom management skills than were high school principals (55% vs. 27%, respectively), although this differ-ence is not statistically significant.

Characteristics mentioned less often. A number of other characteristics were also mentioned multiple times by principals (see Table 2). Less than 40% of the sample mentioned any of the following characteristics: passion for teaching, ability to work well with others, and commitment to participating in profes-sional development. Other characteristics mentioned by three or more princi-pals included having high expectations of students, previous exposure to diversity and disadvantage, credentials (e.g., certification, attending a particu-lar college), technological proficiency, flexibility, and communication skills. Although the importance of credentials was mentioned by only four principals, all four were principals from high-achieving schools. These principals talked about the quality of the college or graduate school a candidate attended and also mentioned specific teacher training programs and schools of education that they prefer. Finally, the following characteristics were each mentioned twice by principals during interviews: confidence, fit with school, leadership skills, organizational skills, intelligence, and ability to work with parents.

What about experience? The vast majority of principals (87%) reported that they do not look specifically for candidates with prior teaching experience when hiring. While four principals said they prefer candidates with some expe-rience, all but one of them said that they do still hire new teachers. Interestingly, approximately one third (n = 11) of principals said they often prefer teachers with little to no experience. Among the reasons these principals gave for their preference for inexperienced teachers is that they are enthusiastic (n = 8), mold-able and/or have not yet developed bad habits (n = 8), that they bring in new ideas (n = 3), and are untenured and therefore easier to remove (n = 2).

The new teacher has this energy and level of excitement, and an ingé-nue quality that is appealing. And they’re going to change the world. And they’re going to fix every child that comes into their classroom.




And, you know, you know that that isn’t going to happen. And yet you appreciate all of that energy and excitement that they have, because that helps to revitalize the school as well. In fact, one of my grade levels right now is sitting with all old-timers. And it’s a grade level that needs to not have that. But, you know, it’s just the way—these teachers haven’t retired yet. And I hired one new teacher because she, hopefully, is going to add some of the energy that needs to come into this grade level . . . You love the experience and the expertise that the older teacher offers, and then you love the vitality that the new teacher has. (Principal Taylor, August 29, 2006)

Another principal describes the trainability that she perceives in newer teachers:

I have found out that sometimes your newer teachers are easier to train and to mold into fantastic teachers. I don’t know if it’s the age issue or if it’s the ability to be a little more flexible. We tend to get a little set in our own ways as we grow old. I know I have. But I see a special quality in younger teachers that I’ve been lucky enough to have a good balance in this school. So while knowledge is extremely important for me, I am willing to balance it if I see other qualities that are valuable for that grade level. (Principal Reynolds, September 11, 2006)

In addition to providing insight into this principal’s appreciation for newer teachers, in the above quotation the principal mentions “balance” in reference to her faculty. This was a relatively common theme among sam-ple principals when discussing teaching experience, although they did not mention it in relation to other teacher characteristics. When discussing how they think about experience when hiring new teachers, one third (n = 10) of the sample reported that they take the experience level of their current fac-ulty into consideration.

Interviewer: Do you tend to hire newer teachers who have not been in the field long? Do you look for more veteran teachers? Is it a mix? How do you think about that?

Respondent: When we look at it, we look at the department overall. If we find that a department needs a person with three, four, five years’ experience, we will tend to go with experience. If not, then we have no problem going with a person that’s fresh out of college, because we have good mentoring teachers. (Principal Jackson, September 12, 2006)



Engel 73

Interestingly, principals in high-achieving schools were twice as likely as those in low-achieving schools (41% vs. 21%) to report taking into account the composition of their current faculty, although this difference is not statis-tically significant.

SurveysDescriptive survey results. Enthusiasm for teaching, classroom management

skills, and the ability to provide a positive role model are the top three charac-teristics that the 368 surveyed CPS principals report looking for in teachers, based on the average importance rating and the proportion of principals who rated the characteristics a 5 or most important (Table 3, columns 1 and 4, respectively). However, only one of those three characteristics—classroom management skills—was among the top characteristics when the proportion of principals ranking a characteristic among the most important is used. In that case (columns 5-7) classroom management skills, the ability to create a fun and stimulating learning environment, ability to increase student achievement in ways that will not necessarily show up on standardized tests, and the extent to which a candidate’s philosophy of learning is a good fit with the school’s were principals’ top choices.

The characteristics that surveyed principals found least important included a candidate’s gender, distance from school, having an advanced degree in education, prior teaching experience, and the quality of the college or univer-sity attended. All these characteristics were at the bottom of the list in terms of both average importance rating and the proportion of principals ranking them most important.

Mirroring results from qualitative interviews, principals in low-achieving schools are statistically significantly more likely than principals in high-achieving schools to rate classroom management skills as important. Columns 2 and 3 (Table 3) show survey item means by whether a school was above or below the median CPS achievement. Principals in low-achieving schools are also significantly more likely to value a teachers’ ability to raise student test scores than their counterparts in high-achieving schools. On the other hand, principals in high-achieving schools are significantly more likely to indicate the importance of a candidate’s “ability to increase student achievement in ways that will not necessarily show up on standardized tests.”

Exploratory regressions. Table 4 presents results from ordered logit mod-els predicting principals’ responses to 4 of the 15 survey items. The out-come variables are principals’ average ratings for four survey items (results for all 15 survey items appear in Appendix B).5 Table 4 includes results for survey items about classroom management skills, ability to raise student



74

Tabl

e 3.

Des

crip

tive

Stat

istic

s fo

r th

e 15

Sur

vey

Item

s Fr

om O

nlin

e Pr

inci

pal S

urve

ys A

dmin

iste

red

Prio

r to

Eac

h Jo

b Fa

ir

(n =

368

). Pr

inci

pals

Bot

h R

ated

Eac

h It

em (

Like

rt-T

ype

Scal

e 1-

5) a

nd R

anke

d T

heir

Top

Thr

ee It

ems

(1)

(2)

(3)

(4)

(5)

(6)

(7)

Teac

her

Cha

ract

eris

tics

(Sur

vey

Item

s)

Ave

rage

Im

port

ance

Rat

ing

(1 =

leas

t; 5

= m

ost)

Ave

rage

Impo

rtan

ce

Rat

ing

Hig

h A

chie

ving

Ave

rage

Impo

rtan

ce

Rat

ing

Low

A

chie

ving

Prop

ortio

n R

atin

g It

em

a 5

Prop

ortio

n R

anki

ng It

em M

ost

Impo

rtan

t

Prop

ortio

n R

anki

ng It

em in

th

e To

p Tw

o

Prop

ortio

n R

anki

ng It

em in

th

e To

p T

hree

The

can

dida

te’s

enth

usia

sm fo

r te

achi

ng (

Que

stio

n 15

)4.

854.

844.

850.

860.

090.

180.

37

The

can

dida

te’s

clas

sroo

m

man

agem

ent

skill

s (Q

uest

ion

1)4.

804.

724.

86a

0.82

0.36

0.63

0.85

The

can

dida

te’s

abili

ty t

o pr

ovid

e a

posi

tive

role

mod

el fo

r st

uden

ts

(Que

stio

n 9)

4.69

4.68

4.66

0.72

0.00

0.02

0.16

The

ext

ent

to w

hich

the

can

dida

te’s

philo

soph

y of

lear

ning

is a

goo

d fit

w

ith t

he s

choo

l’s (

Que

stio

n 3)

4.61

4.56

4.63

0.65

0.12

0.20

0.40

The

can

dida

te’s

abili

ty t

o re

late

wel

l w

ith p

aren

ts (

Que

stio

n 2)

4.59

4.61

4.59

0.64

0.00

0.02

0.25

The

can

dida

te’s

abili

ty t

o ra

ise

stud

ent

test

sco

res

(Que

stio

n 4)

4.51

4.43

4.65

a0.

610.

050.

180.

35

The

can

dida

te’s

abili

ty t

o cr

eate

a

fun

and

stim

ulat

ing

clas

sroo

m

envi

ronm

ent

for

stud

ents

(Q

uest

ion

6)

4.45

4.38

4.49

0.58

0.24

0.43

0.74

The

can

dida

te’s

abili

ty t

o re

late

w

ell w

ith c

olle

ague

s (i.

e., o

ther

te

ache

rs a

nd a

dmin

istr

ator

s in

the

sc

hool

; Que

stio

n 8)

4.45

4.43

4.43

0.53

0.01

0.04

0.26

(con

tinue

d)



75

(1)

(2)

(3)

(4)

(5)

(6)

(7)

Teac

her

Cha

ract

eris

tics

(Sur

vey

Item

s)

Ave

rage

Im

port

ance

Rat

ing

(1 =

leas

t; 5

= m

ost)

Ave

rage

Impo

rtan

ce

Rat

ing

Hig

h A

chie

ving

Ave

rage

Impo

rtan

ce

Rat

ing

Low

A

chie

ving

Prop

ortio

n R

atin

g It

em

a 5

Prop

ortio

n R

anki

ng It

em M

ost

Impo

rtan

t

Prop

ortio

n R

anki

ng It

em in

th

e To

p Tw

o

Prop

ortio

n R

anki

ng It

em in

th

e To

p T

hree

The

can

dida

te’s

abili

ty t

o in

crea

se

stud

ent

achi

evem

ent

in w

ays

that

w

ill n

ot n

eces

sari

ly s

how

up

on

stan

dard

ized

tes

ts (

Que

stio

n 12

)

4.12

4.20

3.98

a0.

390.

120.

260.

50

The

qua

lity

of t

he c

olle

ge o

r gr

adua

te p

rogr

am t

he c

andi

date

at

tend

ed (

Que

stio

n 10

)

3.23

3.30

3.22

0.11

0.00

0.01

0.01

The

can

dida

te’s

prio

r te

achi

ng

expe

rien

ce (

Que

stio

n 7)

3.22

3.11

3.27

0.11

0.00

0.02

0.07

Whe

ther

the

can

dida

te h

as a

tr

aditi

onal

ver

sus

an a

ltern

ativ

e ce

rtifi

catio

n (Q

uest

ion

13)

2.74

2.61

2.84

0.08

0.00

0.00

0.01

Whe

ther

the

can

dida

te h

as a

MA

or

oth

er a

dvan

ced

degr

ee in

ed

ucat

ion

(Que

stio

n 11

)

2.59

2.63

2.53

0.04

0.01

0.01

0.02

How

far

the

cand

idat

e liv

es fr

om t

he

scho

ol (

Que

stio

n 14

)2.

042.

062.

080.

020.

000.

000.

00

The

can

dida

te’s

gend

er (

Que

stio

n 5)

1.56

1.39

1.71

a0.

030.

000.

000.

01

a. D

iffer

ence

bet

wee

n hi

gh-

and

low

-ach

ievi

ng s

choo

ls is

sta

tistic

ally

sig

nific

ant

(p <

.05)

.

Tabl

e 3.

(co

ntin

ued)




test scores, prior teaching experience, and quality of college or graduate program. Regression results are shown for these items either because they were noted as being particularly important during qualitative interviews (classroom management skills), because prior research has linked the characteristic to teacher value-added (prior teaching experience, quality of college or graduate program), or because they seem particularly useful for exploring variation across high- and low-achieving schools (ability to raise student test scores). In each model, the principal’s average score across the 14 other survey items was included as a control to adjust for individual response bias.

Results indicate that principals in low-achieving schools value classroom management skills more than principals in high-achieving schools, control-ling for other principal and school characteristics (Table 4, column 1). These results support findings regarding differences in the importance of classroom management across high- and low-achieving schools from qualitative inter-views (Table 2). Interestingly, both high school principals and principals who attended more selective undergraduate institutions were also significantly less likely to emphasize classroom management skills.

Principals in low-achieving schools put more emphasis on a teacher’s ability to improve student test scores than principals in high-achieving schools, while principals who attended more selective colleges rated this characteristic as less important, conditional on the other variables in the model (Table 4, column 2).

Finally, regression results indicate principals’ preferences for prior teaching experience, as well as for the quality of a candidate’s college or graduate institution, do not vary significantly by level or achievement, although principals who attended more selective institutions were signifi-cantly more likely to rate prior teaching experience as important. These characteristics were considered unimportant to the vast majority of princi-pals who were surveyed, ranking 10th and 11th out of 15 items (Table 3). Overall, Table 4 and Appendix B show that there is little systematic varia-tion in principals’ preferences for teacher characteristics across the vari-ables included in the models.

DiscussionThe current study expands on the work of Harris et al. (2010) and previous research on principals’ preferences for teacher characteristics, making several distinct contributions. It is the only study on this subject focused on a large urban district serving a predominately disadvantaged student population. It uses qualitative interviews as well as complementary results from a survey of



Engel 77

Table 4. Ordered Logistic Regressions Predicting Principal Responses to Key Survey Questions, Controlling for School and Principal Demographic Characteristics

(1) (2) (3) (4)

Classroom Management

Skills

Raise Student Test

Scores

Prior Teaching

Experience

Quality of College or Graduate

Program

Dependent Variable (Survey Items) Q1 Q4 Q7 Q10

School and principal characteristics

High school −0.922 −0.608 −0.404 0.457 (0.490) (0.392) (0.342) (0.336) Average achievement −0.023* −0.023* −0.001 0.001 (0.011) (0.009) (0.008) (0.008) Predominately Hispanic −0.594 0.122 −0.715 −0.103 (0.599) (0.469) (0.413) (0.410) Predominately minority -0.477 0.069 −0.532 −0.333 (0.501) (0.394) (0.325) (0.322) Racially mixed/integrated −0.045 0.361 0.120 0.677 (0.682) (0.547) (0.475) (0.485) Principal age in years <40 −0.758 −0.134 −0.014 0.387 (0.462) (0.394) (0.332) (0.330) 40-50 0.530 −0.261 −0.283 −0.012 (0.491) (0.331) (0.294) (0.291) Male principal −0.244 −0.239 0.171 0.067 (0.338) (0.263) (0.235) (0.235) Black principal −0.044 0.351 0.512 0.068 (0.472) (0.360) (0.325) (0.324) Hispanic principal 0.178 0.123 0.934* −0.407 (0.534) (0.409) (0.374) (0.385) Principal failed one or

more certification exams

−0.207 0.470 −0.140 −0.323

(0.285) (0.399) (0.228) (0.218) Barron’s rating, principal

undergraduate institution

−0.543** −0.351* 0.361** 0.090

(0.192) (0.144) (0.130) (0.129) Average score on 14

other principal items1.646** 1.784** 2.762** 2.686**

(0.406) (0.334) (0.302) (0.301)Observations 368 368 368 368

Mean score on this characteristic (descriptive)

4.8 4.51 3.22 3.23

Note. Robust standard errors are given in parentheses. Predominately Black schools, principals older than 50 years, White principals, and female are the omitted categories.*p < .05.**p < .01.




a large sample of principals in the CPS, the nation’s third largest school district. Qualitative results provide the first data on why principals believe particular characteristics are important and how they determine whether candidates have those characteristics. Surveys provide the opportunity to extend qualitative results by examining whether contextual variation found in interviews extends to a larger sample of CPS principals. Below, implications of the results described above are discussed with a particular focus on two areas. First, principals discuss characteristics they perceive as easily observed or ascertained in more depth than those that are likely much more difficult to detect. Second, principals’ preferences vary sub-stantially across different school contexts. This variation may have impor-tant ramifications.

Principals Emphasize Characteristics They Believe They Can DiscernWith district policy in place since 1996 followed by No Child Left Behind in 2001, high-stakes testing has been the norm in the CPS for well over a decade. The primary goal of improving student test scores should lend itself naturally to a focus on instruction. Thus, we might expect most principals to discuss the importance of a teacher’s content knowledge and teaching ability when talking about their hiring preferences. However, in-depth dis-cussion of preferences regarding instructional practice was the exception, not the rule. It is possible that principals’ minimal discussion of teaching skills and content knowledge reflects their limited capacity to judge these important qualities.

While principals spoke in detail about why they value caring about chil-dren, classroom management skills, willingness to go beyond contractual obligations, and how they identify whether a candidate has these qualities, they spoke far less about content knowledge and teaching skills. Indeed, teaching skills were not even mentioned by over half of principals. This may indicate, as Harris et al. (2010) suggest, that principals hire under the assump-tion that teaching credentials verify that a teacher has the content knowledge, teaching skills, and ability or intelligence to succeed in the classroom. In fact, one principal quoted above states this explicitly, noting that content knowledge was “a given” if a candidate had the requisite credentials. However, relying on credentials will likely give principals little information about the content knowledge or teaching skills of prospective teachers, given that value-added research indicates that certification and degree are not reliable indicators of teacher effectiveness (e.g., Hanushek et al., 2005; Kane et al., 2008; Nye et al., 2004).



Engel 79

During interviews, principals explained that they screen for classroom management skills by having applicants describe how they managed class-rooms in the past, discuss their classroom management plans, or respond to scenarios. Principals also described means for ascertaining whether prospec-tive teachers cared about or were good with children and explained how they determine whether a candidate is likely to go beyond basic contractual obliga-tions. However, most principals did not share analogous means for assessing content knowledge or teaching skills. It is possible that this is because princi-pals do not have means for determining whether a candidate will provide high-quality instruction.

The qualities that principals describe being able to discern include class-room management skills, caring, and willingness to give extra. It is interest-ing to consider the fact that these are affective traits or managerial skills that, perhaps, are relatively easy to observe. In contrast, the absence of descrip-tion of means for determining a candidate’s teaching ability or content knowledge may speak to the fact that principals have limited capacity to evaluate these skills when hiring and also may be less comfortable in terms of their own expertise with regard to classroom instruction.

Context Matters: Differences Across High- and Low-Achieving SchoolsThe current study documents systematic variation in principals’ preferences for teacher characteristics across different school contexts within a single district. Principals in low-achieving schools are more likely to report looking for teachers who care about students, have classroom management skills, and are willing to do or give extra. It seems likely that the strong preference for these characteristics among principals of low-achieving schools may be driven by principals’ perceptions of the unique circumstances and needs in urban schools serving a largely disadvantaged student population.

Both interview and survey results indicate that CPS principals from low-achieving schools place more value on teachers’ classroom management skills than principals of high-achieving schools. On the one hand, the need for classroom management skills may, indeed, be more pervasive in low-achieving schools. However, it is important to consider whether a focus on classroom management skills might overshadow other potentially important characteristics, such as content knowledge and teaching skills, and whether this could be detrimental to students. Previous research on principals’ preferences for teacher characteristics with a sample that was at least somewhat more advan-taged and higher achieving than the one explored here (Harris et al., 2010) did not find classroom management to be a skill that principals mentioned.




The lack of discussion of content knowledge and teaching skills appears to be somewhat more pronounced among principals in low-achieving schools. These principals were nearly 3 times more likely to mention the importance of class-room management skills, and more than 2 times as likely to mention willingness to go beyond contractual obligations, than their colleagues in higher achieving schools. Furthermore, principals in high-achieving schools were somewhat more likely to mention the importance of content knowledge and teaching skills and were more likely to mention the importance of credentials, such as selectivity of college or university (although even among principals from higher achieving schools, these characteristics were mentioned infrequently).

LimitationsThis study is not without limitations. While the fact that the principal survey was administered online as part of the job fair registration process resulted in a very high response rate, the mandatory nature of the survey could also have influenced principal responses. Respondents may have rushed through the sur-vey or tried to provide socially appropriate answers to HR administrators. While these are possibilities, results from qualitative interviews suggest that they are unlikely to have posed a major problem. Qualitative interview response rates were also very high (97%), despite the fact that principals were recruited and consented directly by the researcher. Furthermore, during qualitative interviews, principals appeared to be relaxed, open, and comfortable talking about their preferences for teacher characteristics, suggesting that this was not a topic that they considered to be controversial or about which they were defensive.

It is also important to note that the results presented here are from a study in a single school district. While these results may generalize to other large, urban districts serving students with similar backgrounds, and where princi-pals are charged with hiring the teachers for their schools, there are many contexts in which principals’ preferences are likely to differ substantially. This is a particularly important point to make note of, given that results pre-sented here indicate that principals’ preferences vary substantially and sys-tematically, even within a single district.

Finally, it is important to note the order in which the surveys and inter-views were administered. Online surveys were conducted in the spring and early summer of 2006 when principals were registering for job fairs. Principal interviews were conducted several months later in the fall. The fact that the interviews followed the survey meant that interviews were not used to develop survey items. This resulted in the omission of several teacher charac-teristics that were found to be important during interviews, including caring about children and willingness to give extra, from the survey.



Engel 81

Conclusion and Recommendations

The fact that sample principals provide extensive detail about how they determine whether a candidate cares about children or can manage a class-room, but say little or nothing about assessing content or pedagogical knowl-edge suggests that they may be failing to screen for these key determinants of a teacher’s ability to contribute to student learning. If principals assume that candidates can teach or lack the means for determining which teachers might have stronger instructional skills and therefore focus on classroom management or caring, they may be hiring teachers and learning after the fact that a new hire’s knowledge of content or pedagogy was weaker than expected. This is of particular concern considering the fact that teachers are rarely dismissed for poor performance (Jacob, 2010).

Thus, an important question is whether the trade-offs that principals may be making are, in fact, in their students’ best interest, particularly because these trade-offs may occur with more frequency in the most disadvantaged schools. Principal Ames, whose quote begins this article, tells us she has learned over time to focus on content knowledge and intelligence and is willing to accept candidates who need support or training to improve classroom management. Many sample principals, however, seem to focus on classroom management or caring, perhaps at the expense of content and pedagogical knowledge. While principals serving more disadvantaged students discussed the diverse array of student needs that teachers in their schools must be prepared to address, whether a focus on skills beyond content and pedagogical knowledge comes at a cost for their students is a question that should be explored in future research.

In addition to considering what principals do report focusing on, it is also worth noting that principals were consistent across qualitative interviews and surveys in terms of the characteristics they consider less important. The fact that principals are not focused on the small number of teacher characteristics that are positively associated with teacher effectiveness in some studies of teacher value-added—selectivity of a teacher’s college or university, certi-fication, and years of experience—may initially seem counterintuitive. However, principals’ low interest in these credentials and characteristics may, in fact, be very logical when considered in light of empirical evidence that consistently shows that credentials and experience explain only a very small portion—as little as 3% to 5%— of the variation in teacher effectiveness (e.g., Goldhaber, 2002; Nye et al., 2004; Rockoff, 2004). The behaviors, skills, and dispositions that principals are most likely to value may, in fact, be those that matter most for student outcomes. For example, evidence suggests that peda-gogical content knowledge is positively associated with teacher effectiveness (Hill, Rowan, & Ball, 2005).




We know little, however, about the relationship between teacher value-added and many of the skills and behaviors (e.g., classroom management, caring) that principals report valuing most. More research on the relationship between these behaviors and teacher effectiveness would further our understanding of the extent to which principals value skills and behaviors that directly foster student learning. An additional research question that is generated by the current study is whether the importance of teacher characteristics for student learning varies contextually. In other words, are good classroom management skills more highly correlated with teacher value-added in low-achieving schools than in high-achieving schools? Alternately, are principals who focus on instructional quality, which in turn, may lead to higher levels of student engagement, elimi-nating the need to focus on classroom management skills in isolation?

The results reported here have implications regarding the professional prac-tices of principals. On the one hand, the fact that principals do not appear to focus on credentials may, in fact, be a logical choice given the substantial body of research indicating credentials to be poor predictors of teacher effectiveness in the classroom. However, suggestive evidence presented above may also be cause for concern regarding the qualities that principals do say they value. The fact that principals do not articulate their means for determining whether teach-ers have content and pedagogical expertise, despite their acknowledgement of the importance of these skills, may be problematic. It is possible that principals focus on caring and classroom management skills because they lack the means to discern content and pedagogical knowledge when hiring. If this is the case, it is an important problem and one that should be explored further in future research. Given that principals articulate their strategies for determining whether candidates have other skills that they believe are important, perhaps principals’ limited ability to discern content and pedagogical knowledge could be improved through training and professional development.

Research that replicates and builds on the study described here both in other urban districts and in more advantaged/suburban districts could begin to inform a number of the questions that it generates. Even within the same district, principals from high- versus low-achieving schools report prefer-ences for teachers with different characteristics. Would a similar study in a more advantaged suburban Chicago-area district produce similar results? Do principals in different contexts (e.g., rural districts, more advantaged suburban districts) look for different things when they are hiring teachers? Whether principals in more advantaged districts would be more focused on candidates’ content and pedagogical knowledge than on whether they care or can manage a classroom is an open question. Exploratory studies in a diverse range of contexts will help us begin to understand this contextual variation and its implications for students and schools.



83

Perc

enta

ge o

f Sam

ple

Prin

cipal

s Who

Rep

orte

d (D

urin

g Q

ualit

ative

Inte

rvie

ws)

Tha

t Par

ticul

ar

Teac

her

Qua

litie

s Wer

e Im

port

ant t

o The

m D

urin

g th

e H

iring

Pro

cess

, Ove

rall

and

by P

rincip

al R

ace/

Ethn

icity

, Gen

der,

and Y

ears

of E

xper

ienc

e(1

)(2

)(3

)(4

)(5

)(6

)(7

)(8

)

Ov e

rall

Perc

enta

ge

(n =

31)

Blac

k Pr

inci

pal

(n =

16)

Whi

te

Prin

cipa

l (n

= 1

1)

His

pani

c Pr

inci

pal

(n =

4)

Fem

ale

Prin

cipa

l (n

= 2

1)

Mal

e Pr

inci

pal

(n =

10)

Prin

cipa

l at

Scho

ol <

5 Ye

ars

(n =

15)

Prin

cipa

l at

Scho

ol ≥

5 Ye

ars

(n =

16)

Cha

ract

eris

tic

C

are

abou

t/un

ders

tand

chi

ldre

n61

7536

a75

5770

5369

C

onte

nt k

now

ledg

e55

5055

7557

5033

75a

D

o/gi

ve e

xtra

5256

450

4840

4744

C

lass

room

man

agem

ent

4556

3625

5230

4050

Te

achi

ng s

kills

/tea

chin

g st

yle

4250

3625

4340

4738

Pa

ssio

n/en

thus

iasm

for

teac

hing

3931

5525

3840

4038

Te

am p

laye

r26

1336

5024

3040

13

Com

mitt

ed t

o pr

ofes

sion

al d

evel

opm

ent

2325

1825

2420

2719

H

igh

expe

ctat

ions

of s

tude

nts

1619

180

1910

1319

Pr

evio

us e

xpos

ure

to d

iver

sity

1619

180

1030

2013

C

rede

ntia

ls13

69

5019

0a13

13

Tech

nolo

gica

l pro

ficie

ncy

1013

90

1010

713

Fl

exib

ility

1013

90

520

713

G

ood

com

mun

icat

or10

027

a0

1010

713

Expe

rien

ce

Pr

efer

s ex

peri

ence

d te

ache

rs13

199

014

1013

13

Con

side

rs e

xper

ienc

e of

cur

rent

facu

lty32

1955

*25

3820

2044

Not

e. O

nly

qual

ities

and

cha

ract

eris

tics

that

wer

e m

entio

ned

by t

hree

or

mor

e pr

inci

pals

are

incl

uded

in t

he t

able

.a.

Indi

cate

s di

ffere

nces

bet

wee

n Bl

ack

and

Whi

te, m

ale

and

fem

ale,

exp

erie

nced

ver

sus

inex

peri

ence

d pr

inci

pals

wer

e st

atis

tical

ly s

igni

fican

t (p

< .0

5). B

ecau

se t

here

wer

e on

ly

four

His

pani

c pr

inci

pals

in t

he s

ampl

e, I

did

not

test

diff

eren

ces

betw

een

Blac

k an

d H

ispa

nic

or W

hite

and

His

pani

c pr

inci

pals

.

App

endi

x A



84

Ord

ered

Log

istic

Regr

essio

ns P

redi

ctin

g Pr

incip

al S

urve

y Re

spon

ses

Usin

g Sc

hool

and

Prin

cipal

D

emog

raph

ics a

s In

depe

nden

t Var

iabl

es

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

Scho

ol a

nd

Prin

cipa

l C

hara

cter

istic

sq1

q2q3

q4q5

q6q7

q8q9

q10

q11

q12

q13

q14

q15

Hig

h sc

hool

−0.

922

−0.

283

0.34

1−

0.60

8−

0.68

2−

0.03

9−

0.40

4−

0.30

40.

102

0.45

70.

240

0.78

8*0.

024

0.17

7−

0.16

0

(0.4

90)

(0.4

06)

(0.4

08)

(0.3

92)

(0.4

27)

(0.4

06)

(0.3

42)

(0.3

78)

(0.4

62)

(0.3

36)

(0.3

45)

(0.3

59)

(0.3

37)

(0.3

41)

(0.5

43)

Ach

ieve

men

t−

0.02

3*0.

008

0.00

7−

0.02

3*−

0.02

1*−

0.00

7−

0.00

10.

013

0.00

30.

001

0.00

40.

011

0.00

40.

005

−0.

003

(0

.011

)(0

.010

)(0

.010

)(0

.009

)(0

.010

)(0

.009

)(0

.008

)(0

.009

)(0

.011

)(0

.008

)(0

.008

)(0

.008

)(0

.008

)(0

.008

)(0

.013

)Pr

edom

inat

ely

His

pani

c−

0.59

4−

0.86

50.

168

0.12

2−

0.41

70.

863

−0.

715

−0.

302

0.15

0−

0.10

3−

0.05

20.

021

−0.

133

0.64

41.

307

(0

.599

)(0

.481

)(0

.473

)(0

.469

)(0

.519

)(0

.465

)(0

.413

)(0

.452

)(0

.535

)(0

.410

)(0

.407

)(0

.419

)(0

.383

)(0

.406

)(0

.746

)Pr

edom

inat

ely

min

ority

−0.

477

−0.

377

−0.

173

0.06

90.

254

0.72

2−

0.53

20.

381

0.34

8−

0.33

3−

0.45

20.

291

0.06

4−

0.00

70.

933

(0

.501

)(0

.399

)(0

.396

)(0

.394

)(0

.370

)(0

.398

)(0

.325

)(0

.383

)(0

.463

)(0

.322

)(0

.324

)(0

.346

)(0

.326

)(0

.329

)(0

.640

)R

acia

lly m

ixed

/in

tegr

ated

−0.

045

−0.

525

−0.

288

0.36

10.

791

1.36

7*0.

120

−0.

269

0.55

60.

677

0.09

90.

301

−0.

942*

−0.

287

0.75

2

(0

.682

)(0

.569

)(0

.560

)(0

.547

)(0

.567

)(0

.550

)(0

.475

)(0

.541

)(0

.639

)(0

.485

)(0

.484

)(0

.503

)(0

.475

)(0

.484

)(0

.794

)Pr

inci

pal a

ge in

ye

ars

<

40−

0.75

8−

0.32

7−

0.16

7−

0.13

40.

151

0.48

8−

0.01

40.

233

−0.

326

0.38

70.

505

0.23

1−

0.17

8−

0.01

50.

307

(0

.462

)(0

.400

)(0

.395

)(0

.394

)(0

.376

)(0

.391

)(0

.332

)(0

.385

)(0

.437

)(0

.330

)(0

.337

)(0

.350

)(0

.314

)(0

.345

)(0

.614

)

40-5

00.

530

0.68

60.

452

−0.

261

0.25

40.

183

−0.

283

0.43

50.

720

−0.

012

0.02

50.

245

−0.

669*

−0.

229

0.06

8

(0.4

91)

(0.3

66)

(0.3

51)

(0.3

31)

(0.3

33)

(0.3

26)

(0.2

94)

(0.3

35)

(0.4

14)

(0.2

91)

(0.2

91)

(0.2

98)

(0.2

95)

(0.3

03)

(0.4

81)

App

endi

x B

(con

tinue

d)



85

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

Scho

ol a

nd

Prin

cipa

l C

hara

cter

istic

sq1

q2q3

q4q5

q6q7

q8q9

q10

q11

q12

q13

q14

q15

Mal

e pr

inci

pal

−0.

244

−0.

595*

−0.

441

−0.

239

0.32

5−

0.32

90.

171

−0.

145

−0.

797*

*0.

067

−0.

109

0.19

60.

126

0.17

70.

165

(0

.338

)(0

.274

)(0

.270

)(0

.263

)(0

.280

)(0

.255

)(0

.235

)(0

.264

)(0

.309

)(0

.235

)(0

.239

)(0

.244

)(0

.233

)(0

.244

)(0

.387

)Bl

ack

prin

cipa

l−

0.04

4−

0.21

30.

105

0.35

10.

362

0.99

1**

0.51

20.

008

−0.

001

0.06

8−

0.31

5−

0.17

1−

0.34

5−

0.21

10.

622

(0

.472

)(0

.386

)(0

.381

)(0

.360

)(0

.386

)(0

.373

)(0

.325

)(0

.371

)(0

.451

)(0

.324

)(0

.326

)(0

.343

)(0

.322

)(0

.328

)(0

.567

)H

ispa

nic

prin

cipa

l

0.17

80.

238

−0.

376

0.12

30.

215

−0.

595

0.93

4*0.

535

−0.

009

−0.

407

−0.

289

−0.

299

0.47

3−

0.42

3−

0.33

9(0

.534

)(0

.444

)(0

.435

)(0

.409

)(0

.485

)(0

.399

)(0

.374

)(0

.431

)(0

.500

)(0

.385

)(0

.395

)(0

.383

)(0

.365

)(0

.389

)(0

.609

)

Prin

cipa

l fai

led

one

or m

ore

cert

ifica

tion

exam

s

−0.

207

0.60

50.

049

0.47

00.

190

0.86

1−

0.14

00.

081

1.01

5−

0.32

30.

038

−0.

051

0.36

00.

058

0.21

0(0

.285

)(0

.432

)(0

.302

)(0

.399

)(0

.254

)(0

.451

)(0

.228

)(0

.252

)(0

.561

)(0

.218

)(0

.218

)(0

.222

)(0

.218

)(0

.204

)(0

.490

)

Barr

on’s

ratin

g, pr

inci

pal

unde

rgra

duat

e in

stitu

tion

−0.

543*

*−

0.03

3−

0.15

4−

0.35

1*−

0.03

7−

0.03

10.

361*

*0.

001

−0.

076

0.09

00.

194

−0.

083

−0.

026

0.09

1−

0.23

0(0

.192

)(0

.151

)(0

.147

)(0

.144

)(0

.147

)(0

.139

)(0

.130

)(0

.145

)(0

.169

)(0

.129

)(0

.128

)(0

.129

)(0

.121

)(0

.128

)(0

.212

)

Ave

rage

sco

re

on 1

4 ot

her

prin

cipa

l ite

ms

1.64

6**

2.22

8**

1.47

4**

1.78

4**

1.93

2**

1.87

9**

2.76

2**

2.44

1**

2.62

9**

2.68

6**

2.81

6**

1.47

3**

2.28

5**

2.67

5**

1.72

4**

(0.4

06)

(0.3

58)

(0.3

23)

(0.3

34)

(0.3

49)

(0.3

27)

(0.3

02)

(0.3

39)

(0.4

09)

(0.3

01)

(0.3

13)

(0.2

82)

(0.2

92)

(0.3

24)

(0.4

40)

Obs

erva

tions

368

368

368

368

368

368

368

368

368

368

368

368

368

368

368

Mea

n sc

ore

on t

his

char

acte

rist

ic

4.8

4.59

4.6

4.51

1.56

4.45

3.22

4.45

4.69

3.23

2.59

4.12

2.74

2.04

4.85

Not

e. R

obus

t st

anda

rd e

rror

s ar

e gi

ven

in p

aren

thes

es. P

redo

min

atel

y Bl

ack

scho

ols,

prin

cipa

ls o

lder

tha

n 50

yea

rs, W

hite

pri

ncip

als,

and

fem

ale

are

the

omitt

ed c

ateg

o-ri

es. S

ee T

able

3 o

r App

endi

x D

for

list

of d

epen

dent

var

iabl

es (

surv

ey it

ems)

.*p

< .0

5.**

p <

.01.

App

endi

x B

(co

ntin

ued)




Appendix C2006 Chicago Public Schools Principal Interview Protocol

Principal Interview Protocol

School/Unit___________ Participants name___________

Introduction. As the consent form mentioned, we’re trying to learn more about how principals go about hiring teachers. We want to understand what characteristics you look for in teachers, and how you identify those character-istics . . . Feel free to stop me at any time if you have any questions. Also, if I don’t ask you about something that you believe is important regarding this issue, please let me know. Do you have any questions before we get started?

Teacher Qualities and Qualifications. Take a minute to think about the two best teachers you’ve known/worked with. What characteristics did they have that made them good teachers?

More generally, in your opinion, what are the characteristics of a quali-fied teacher?

What qualities are most important in making someone a good teacher?

How does this vary by subject and/or grade level?

What specific qualities do you look for in a new teacher?

Which of the characteristics you just described are most impor-tant to you when you’re hiring a new teacher?

Does this vary by the subject or grade level or other things? If so, how?

How do you view experience (i.e. who do you tend to hire, newer or more experienced teachers)?

What about candidates who just graduated and have never taught? Is cost (your budget) a factor in this?

How about people with alternative certification?How can you tell if a job candidate has the qualities you’re look-

ing for?Are some of these qualities more difficult to spot than others?

How so?



Engel 87

Appendix D

Survey Items From Survey Administered Online to Chicago Public Schools Principals During Spring and Summer 2006

The questions below ask you to describe which qualities are important to you in hiring a new teacher. We realize that you might consider different qualities when hiring different teachers. In answering the questions below, please consider hiring a core subject area teacher (e.g., math, science, his-tory, English, or foreign language) for a “typical” class in your school.

1. The candidate’s classroom management skills 2. The candidate’s ability to relate well with parents 3. The extent to which the candidate’s philosophy of learning is a

good fit with the school’s 4. The candidate’s ability to raise student test scores 5. The candidate’s gender 6. The candidate’s ability to create a fun and stimulating classroom

environment for 7. The candidate’s prior teaching experience 8. The candidate’s ability to relate well with colleagues (i.e., other

teachers and administrators in the school) 9. The candidate’s ability to provide a positive role model10. The quality of the college or graduate program the candidate attended11. Whether the candidate has a MA or other advanced degree in edu-

cation12. The candidate’s ability to increase student achievement in ways that

will not necessarily show up on standardized test scores13. Whether the candidate has a traditional versus an alternative certifi-

cation14. How far the candidate lives from the school15. The candidate’s enthusiasm for teaching

AcknowledgmentsI am grateful to Brian Jacob for sharing his survey data and for feedback on earlier drafts of this article. Thanks also to Greg Duncan, Larry Hedges, Jim Spillane, and seminar participants at Northwestern University and Vanderbilt University for com-ments. Thanks to F. Chris Curran for research assistance and to the Chicago Public Schools principals and Department of Human Resources for allowing me to collect data and for sharing administrative records. All errors are my own.




Declaration of Conflicting InterestsThe author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

FundingThe author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: WT Grant provided funding that partially supported this research, as did an AERA/IES dissertation grant.

Notes

1. All principal names and identifying information have been changed.2. Although intriguing, these results should be interpreted with caution. The survey

had a very low response rate (27%), and sample schools from New York City and its immediate suburbs were excluded from the analyses.

3. Data on CPS schools, principals, students, and teachers are for the 2005-2006 and 2006-2007 school years and were taken from the CPS website: http://www.cps.k12.il.us/AtAGlance.html

4. The response rate is 85% among the 432 schools that attended at least one job fair, as charter school principals and schools whose principals registered for fairs on-site did not complete the survey.

5. In Table 4, I show ordered logistic models for the subgroup of survey items that are of the most interest. Ordered logistic results for all survey items are provided in Appendix B. In addition, I ran logistic models using ranking infor-mation as dichotomous outcomes and also ran ordinary least squares models predicting responses (results not shown). The pattern of results was similar across all models (results not shown).

References

Aaronson, D., Barrow, L., & Sander, W. (2007). Teachers and student achieve-ment in the Chicago public high schools. Journal of Labor Economics, 25, 95-135.

Baker, B. D., & Cooper, B. S. (2005). Do principals with stronger academic back-grounds hire better teachers? Policy implications for improving high-poverty schools. Educational Administration Quarterly, 41, 449-479.

Ballou, D. (1996). Do public schools hire the best applicants? Quarterly Journal of Economics, 111, 97-133.

Ballou, D., & Podgursky, M (1997). Teacher pay and teacher quality. Kalamazoo, MI: W. E. Upjohn Institute for Employment Research.

Balter, D., & Duncombe, W. (2005a, November). Staffing classrooms: How do New York school districts find their teachers? Paper presented at the Southern Economic Asso-ciation Annual Conference, Washington, DC.



Engel 89

Balter, D., & Duncombe, W. (2005b, January). Teacher hiring practices in New York State school districts. Report prepared for the Education Finance Research Con-sortium, Albany, NY.

Balter, D., & Duncombe, W. (2006). Staffing classrooms: Do teacher hiring practices affect teacher qualifications? Paper presented at the meeting of the American Education Finance Association, Denver, CO.

Brewer, D. J. (1993). Principals and student outcomes: Evidence from U. S. high schools. Economics of Education Review, 12, 281-292.

Clotfelter, C. T., Ladd, H. F., & Vigdor, J. L. (2006). Teacher-student matching and the assessment of teacher effectiveness. Journal of Human Resources, XLI, 778-820.

Clotfelter, C. T., Ladd, H. F., & Vigdor, J. L. (2007a). Teacher credentials and student achievement: Longitudinal analysis with student fixed effects. Economics of Edu-cation Review, 26, 673-682.

Clotflelter, C. T., Ladd, H. F., & Vigdor, J. L. (2007b). How and why do teacher cre-dentials matter for student achievement (National Bureau of Economic Research Working Paper 12828). Retrieved from http://www.nber.org/papers/w12828.pdf

Croninger, R., Rice, J., Rathbun, A., & Nishio, M. (2007). Teacher qualifications and early learning: Effects of certification, degree, and experience on first-grade stu-dent achievement. Economics of Education Review, 26, 312-324.

Dunton, J. (2001). Selection criteria used by high school principals in Virginia when hiring first year career and technical teachers (Unpublished doctoral disserta-tion). Virginia Polytechnic Institute and State University, Blacksburg.

Ehrenberg, R. G., & Brewer, D. J. (1994). Do school and teacher characteristics matter? Evidence from high school and beyond. Economics of Education Review, 13, 1-17.

Ehrenberg, R. G., & Brewer, D. J. (1995). Did teachers’ verbal ability and race matter in the 1960s? Coleman revisited. Economics of Education Review, 14, 1-21.

Emley, K., & Ebmeier, H. (1997). The effect of employment interview format on principals’evaluations of teachers. Journal of Personnel Evaluation in Education, 11, 39-56.

Engel, M., & Jacob, B. A. (2011). New evidence on teacher labor supply (National Bureau of Economic Research Working Paper 16802). Retrieved from http://www.nber.org/papers/w16802.pdf

Goldhaber, D. D. (2002). The mystery of good teaching: Surveying the evidence on student achievement and teachers’ characteristics. Education Next, 2(1), 50-55.

Goldhaber, D. D., & Brewer, D. J. (2000). Does teacher certification matter? High school teacher certification status and student achievement. Educational Evalua-tion and Policy Analysis, 22, 129-146.

Grissom, J. A., & Loeb, S. (2011). Triangulating principal effectiveness: How perspec-tives of parents, teachers, and assistant principals identify the central importance of managerial skills. American Educational Research Journal, 48, 1091-1123.




Hanushek, E. A. (1986). The economics of schooling: Production and efficiency in public schools. Journal of Economic Literature, 24, 1141-1177.

Hanushek, E. A. (1997). Assessing the effects of school resources on student perfor-mance: An update. Educational Evaluation and Policy Analysis, 19, 141-64.

Hanushek, E. A., Kain, J. F., O’Brien, D. M., & Rivkin, S. G. (2005). The market for teacher quality (National Bureau of Economic Research Working Paper 11154). Retrieved from http://www.nber.org/papers/w11154.pdf

Harris, D. N., Rutledge, S. A., Ingle, W. K., & Thompson, C. C. (2010). Mix and match: What principals really look for when hiring teachers. Education Finance and Policy, 5, 228-246.

Harris, D. N., & Sass, T. R. (2008). Teacher training, teacher quality and student achievement. (National Center for Analysis of Longitudinal Data in Education Research Working Paper No. 3).

Harris, D. N., & Sass, T. R. (2009). What makes for a good teacher and who can tell? (CALDER Working Paper No. 30).

Hill, H., Rowan, B., & Ball, D. (2005). Effects of teacher’s mathematical knowledge for teaching on student achievement. American Educational Research Journal, 42, 371-406.

Horng, E., Klasik, D., & Loeb, S. (2010) Principal time-use and school effectiveness. American Journal of Education, 116, 491-524.

Jacob, B. A. (2007). The challenges of staffing urban schools with effective teachers. Future of Children, 17, 129-154.

Jacob, B. A. (2010). Do principals fire the worst teachers? (National Bureau of Eco-nomic Research Working Paper 15715). Retrieved from http://www.nber.org/papers/w15715.pdf

Jacob, B. A., & Lefgren, L. (2008). Can principals identify effective teachers? Evi-dence on subjective performance evaluation in education. Journal of Labor Eco-nomics, 26, 101-136.

Kane, T., Rockoff, J., & Staiger, D. (2008). What does certification tell us about teacher effectiveness? Evidence from New York City. Economics of Education Review, 27, 615-631.

Liu, E., & Johnson, S. M. (2006). New teachers’ experiences of hiring: Late, rushed, and information poor. Educational Administration Quarterly, 43, 324-360.

Nye, B., Konstantopoulos, S., & Hedges, L. V. (2004). How large are teacher effects? Educational Evaluation and Policy Analysis, 26, 237-257.

Papa, F., & Baxter, I. (2008). Hiring teachers in New York’s public schools: Can the principal make a difference? Leadership and Policy in Schools, 7, 87-117.

Pflaum, S. W., & Abramson, T. (1990). Teacher assignment, hiring, and preparation: Minority teachers in New York City. The Urban Review, 22, 17-31.



Engel 91

Rockoff, J. E. (2004). The impact of individual teachers on student achievement: Evi-dence from panel data. American Economic Review, 94, 247-252.

Strauss, R. P. (2003, March). The preparation and selection of public school admin-istrators in Pennsylvania: Supply and demand and the effects on student achieve-ment. Paper presented at the annual meeting of the American Education Finance Association, Orlando, FL.

Strauss, R. P., Bowes, L. R., Marks, M. S., & Plesko, M. R. (2000). Improving teacher preparation and selection: Lessons from the Pennsylvania experience. Economics of Education Review, 19, 387-415.

Young, I. P., & Allison, B. (1982). Effects of candidate age and teaching experience on school superintendents and principals in selecting teachers. Planning and Chang-ing, 13, 245-256.

Bio

Mimi Engel is an assistant professor of public policy and education at Vanderbilt University’s Peabody College in the Department of Leadership, Policy, and Organizations. Her research interests include teacher labor markets, urban education, school reform, program evaluation, and early skill formation.

problematic preferences? a mixed method examination of principals' preferences for teacher...

Documents