Collegial Climate and Novice Teachers’ Intent to Remain Teaching

Author(s): Ben Pogodzinski, Peter Youngs, and Kenneth A. Frank

Source: American Journal of Education, Vol. 120, No. 1 (November 2013), pp. 27-54
Published by: The University of Chicago Press
Stable URL: http://www.jstor.org/stable/10.1086/673123 .
Accessed: 11/11/2013 15:07

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to
American Journal of Education.

http://www.jstor.org

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
Collegial Climate and Novice Teachers’ Intent
to Remain Teaching

BEN POGODZINSKI
Wayne State University
PETER YOUNGS
Michigan State University
KENNETH A. FRANK
Michigan State University

Using survey data from novice teachers across 99 schools, we estimated multilevel
regressions to identify the association between novices’ intent to remain teaching
within their schools and their perceptions of the collegial climate. The results
suggest that novice teachers who perceive a more positive collegial climate
marked by higher degrees of professional fit and collective responsibility among
colleagues are more likely to report intent to remain teaching within their schools.
The findings from this study have implications for understanding novice teachers’
intended career decisions as a function of their perceptions of the collegial climate
within their schools.

Novice teachers (generally defined as teachers in the first 3 years of the pro-
fession) enter the profession with a particular idea of what school is all about
based on their own experiences as students, their university-based teacher
preparation, and their preservice teaching (Lortie 1975; Staton and Hunt
1992). In other words, they have an idea of what effective teaching is, how
students behave, and how adults behave within a school. Yet in each school
the routines and practices of adults may or may not provide a supportive
environment for novice teachers. The reality of a given school context may
not live up to the idea of school that novice teachers hold and may fail to
provide them with the support and resources they feel necessary to succeed
in their work (Flores and Day 2006).
Organizational climate arises from the everyday policies and practices that
in part define an organization and its members, as well as how these policies

Electronically published August 21, 2013

American Journal of Education 120 (November 2013)
䉷 2013 by The University of Chicago. All rights reserved.
0195-6744/2013/12001-0002$10.00

NOVEMBER 2013 27

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Collegial Climate

and practices are perceived by members of the organization (Hoy and Miskel
2008; Litwin and Stringer 1968; Tagiuri 1968). Therefore, the organizational
climate influences the attitudes and behaviors of organizational members (Gil-
mer 1966; Hoy and Miskel 2008). In this study, we were particularly interested
in identifying the association between novice teachers’ perceptions of the col-
legial climate (i.e., the routine beliefs and actions associated with teacher-
teacher relationships) and their intent to remain teaching within their schools.
Specifically, the extent to which novice teachers are aligned with the goals
and beliefs of their colleagues, the levels of trust among teachers, and the
degree of collective responsibility among teachers likely influence the likelihood
a teacher would want to remain teaching within a particular school (Bryk and
Schneider 2002; Kristof-Brown et al. 2005; Penuel et al. 2010).
Therefore, the purpose of this study was (a) to measure novice teachers’
perceptions of the collegial climate, including their perceptions of professional
fit with colleagues (i.e., the degree to which their professional practices and
goals align with those of their colleagues), their perceptions of the degree of
relational trust among teachers within the school, and the degree of collective
responsibility among teachers within the school; and (b) then to identify the
association between these perceptions and their intent to remain teaching
within their schools. Novices’ intent to remain teaching should be a concern
for researchers and practitioners because it is related to the amount of effort
teachers put into their current work, and it is often highly correlated with
actual career decisions (Chan et al. 2008; Kushman 1992; Ladd 2011; Mitchell
et al. 2001).
Developing a better understanding of the predictors of novice teacher com-
mitment and career decisions is important because these factors have been
found to be directly and indirectly related to school effectiveness (Ronfeldt et

BEN POGODZINSKI is an assistant professor of educational leadership and

policy studies at Wayne State University. His research interests focus on how
state and district policies, school organizational context, and labor relations
influence teachers’ instructional practices, effectiveness, and labor market de-
cisions. PETER YOUNGS is an associate professor of teacher education and
educational policy at Michigan State University. His research interests focus
on state and district policy related to teacher preparation, induction, and
professional development in the United States and their effects on teachers’
instructional practices, commitment to teaching, and retention in the teaching
profession. KENNETH A. FRANK is a professor of measurement and quanti-
tative methods at Michigan State University. His substantive interests include
the study of schools as organizations, how teachers influence one another to
affect classroom practices and school decision making, social networks, and
the social context of learning.

28 American Journal of Education

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Pogodzinski, Youngs, and Frank

al. 2011). Particularly, research has found that the least effective teachers (in
terms of value-added measures) are more likely to leave a school, especially
in schools that serve high percentages of low-achieving and black students
(Boyd et al. 2011; Hanushek and Rivkin 2010). Yet it is not known whether
or not, on average, these teachers are replaced by more effective teachers
(Ronfeldt et al. 2011). Additionally, teacher turnover enacts substantial finan-
cial costs on schools and districts (e.g., costs associated with recruiting, hiring,
and training new teachers), and it is likely that high rates of teacher turnover
affect schools’ ability to maintain or build effectiveness as organizations (Al-
liance for Excellent Education 2004; Ronfeldt et al. 2011; Smith and Ingersoll
2004).

Collegial Climate and Intent to Remain Teaching

Individuals base their career decisions on personal factors as well as profes-

sional factors. Generally, in addition to factors exogenous to the organization
(e.g., spousal relocation, illness, etc.), “job attitudes combined with job alter-
natives predict intent to leave, which is the direct antecedent to turnover”
(Mitchell et al. 2001, 1102). How novice teachers understand and evaluate
the organizational climate in part influences these job attitudes (Gilmer 1966;
Litwin and Stringer 1968).
Researchers have identified several characteristics of a teaching position
that affect job attitudes and have been found to be associated with teacher
turnover, including teacher, student, and school attributes, as well as admin-
istrator support and economic factors (e.g., salary and benefits; Boyd et al.
2006; Goldhaber et al. 2011; Guarino et al. 2006; Hanushek et al. 2004).
Additionally, researchers have shown that teachers’ understanding of the col-
legial climate within their school may ultimately influence their intent to
remain teaching within their school. Specifically, it is through interactions with
colleagues that (a) norms and expectations for teacher behavior are com-
municated to novice teachers (Feiman-Nemser 2010; Feiman-Nemser et al.
1999) and (b) networks are established through which novice teachers may
access resources and support (Coburn and Russell 2008; Coleman 1988; Pen-
uel et al. 2009, 2010).

Novice Teacher Mentoring

Novice teachers evaluate the collegial climate in part on the basis of their
interactions with their formally assigned mentors whose primary purpose is
to socialize novices into the teaching profession and their local context and

NOVEMBER 2013 29

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Collegial Climate

to provide them with additional learning opportunities. In 2010, more than

30 states required new teachers to participate in some form of induction
( Johnson et al. 2010), and teacher mentoring programs have become the
primary element of formal novice teacher induction (Ingersoll and Strong
2011; Strong 2009). Of particular importance is the extent to which formal
mentors reflect the collegial climate (e.g., the extent to which teachers col-
laborate and offer support) and connect novice teachers to the broader com-
munity of teachers. As Feiman-Nemser et al. (1999) state, “if mentors do not
represent shared standards of practice or promote a sense of collective re-
sponsibility for student learning, novices may not come to see themselves as
part of a broader collectivity working toward improved teaching and learning
for all students” (29).
Research, though, has found uneven effects of mentoring on novices’ com-
mitment and retention. For example, in their study using nationally repre-
sentative data from the 1999–2000 Schools and Staffing Survey (SASS), Smith
and Ingersoll (2004) found that having a mentor in one’s field reduced the
likelihood of leaving teaching at the end of the first year by 30%. Similarly,
in a study of 1,180 novice elementary teachers in Chicago, Kapadia et al.
(2007) reported that those novices receiving strong mentoring “were much
more likely (than other novices) to report a good experience, intend to continue
teaching, and plan to remain in the same school” (28).
Yet, a recent multiyear randomized control trial study (n p 1,009 teachers)
of two comprehensive induction programs is perhaps the most comprehensive
study of new teacher induction thus far (Glazerman et al. 2010). Compre-
hensive induction programs in the study were characterized by full-time men-
tors who received extensive training, weekly meetings between mentors and
mentees, professional development opportunities, and formative assessments
of novice teachers. The researchers found that comprehensive induction over
2 years had no statistically significant effect on teacher retention (Glazerman
et al. 2010).

Interactions with Other Colleagues

In contrast to socialization to the collegial climate as experienced through

formal mentoring, socialization also occurs through more everyday interac-
tions with colleagues throughout a school (Feldman 1989; Jones 1986; Meyer
et al. 1981). It is often thought that socialization is done to novice teachers,
yet novice teachers enact agency when self-selecting into particular relation-
ships that affect the degree to which they are open to the socialization efforts
of the organization (Bryk and Schneider 2002; Lounsbury and Ventresca 2003;
Powell and Colyvas 2007). Therefore, the relationships that novice teachers

30 American Journal of Education

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Pogodzinski, Youngs, and Frank

choose to enter into within a school (both professionally and personally) shape
the way in which they understand and evaluate the collegial climate (Bryk
and Schneider 2002; Coburn and Russell 2008; Penuel et al. 2010).
Kapadia et al. (2007) found that novice elementary teachers who had regular
opportunities to collaborate with peers in the same field were much more
likely than other novices to “report a good experience and intend to remain
in the same school” (30). In a second study, using the 1999–2000 SASS, Smith
and Ingersoll (2004) found that for first-year teachers, collaborating with col-
leagues on instruction reduced the risk that they would leave teaching (by
43%) or migrate to other schools (by 25%). Kardos et al. (2001) found that
new teachers sought signals from their colleagues about how to interact with
students and colleagues, which instructional approaches were acceptable, ap-
propriate behavior in meetings, and appropriate use of time. Their findings
suggest that school context is an important factor in determining the social-
ization that novice teachers will receive.

A Framework of Collegial Climate

The way in which organizational members evaluate organizational climate

influences their behaviors and attitudes (Denison 1996; Gilmer 1966; Hoy
and Miskel 2008; Litwin and Stringer 1968; Poole 1985; Tagiuri 1968). Spe-
cifically, the individual and collective perceptions of schoolwide relations shape
the attitudes and behaviors of organization members (Hoy and Miskel 2008).
The focus of this study is on one critical component of the organizational
climate, the collegial climate, which relates to the routine beliefs and practices
of teachers within a school.
The collegial climate is ingrained within the entire school as an organization.
In other words, perceptions of teacher-teacher relations are based on the
collective relationship among teachers within groups and across a school (Macy
1990). This conceptualization builds on many previous studies of novice teach-
ers’ relationships with their colleagues and their career decisions (e.g., Johnson
2004) because it focuses not only on the extent to which an individual novice
teacher feels supported by his or her colleagues but also more broadly on the
extent to which novice teachers feel that teachers throughout the school sup-
port each other. Therefore, all teachers are influenced by the collegial climate
across the school regardless of a specific personal relationship an individual
teacher has with another teacher or group of teachers.
There are three aspects of collegial climate that are the focus of this study:
(a) degree of professional fit among teachers, (b) level of relational trust among
teachers, and (c) degree of collective responsibility among teachers. These
conceptualizations of collegial climate are certainly not the only aspects that

NOVEMBER 2013 31

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Collegial Climate

could be studied, but they did emerge from our review of the literature on
teachers’ social-organizational context as important themes. These aspects of
collegial climate are often closely linked within an organization but remain
distinct from each other. These aspects of collegial climate are also related to
elements of teacher efficacy (at both the individual and group levels), but
efficacy is directly related to beliefs about competency while the aspects of
collegial climate that are the focus of this study are more aligned with beliefs
about school community and personal and professional accountability among
colleagues (Whalen 2012).

Professional Fit

Drawing on research in organizational psychology (Chatman 1989; Kristof

1996; Kristof-Brown et al. 2005), we propose that the degree of a novice
teacher’s fit with her colleagues is an integral component of the collegial climate
that is likely related to her intent to remain teaching within her school. That
is, novice teachers who do not share in organizational preferences for attitudes
and practices that permeate throughout a school are likely to seek employment
in another setting (Chan et al. 2008). It is likely that novice teachers not only
have emotional responses to these work socialization experiences (e.g., they
like or dislike coworkers) but also evaluate the degree to which they are aligned
with the goals, values, and expectations of their work (Kristof-Brown et al.
2005).
The notion of fit has been conceptualized as having both affective and
normative elements. Affectively, those who have positive feelings toward their
coworkers may perceive higher levels of fit (Allen and Meyer 1990; Mitchell
et al. 2001). Normatively, fit represents a cognitive belief that their needs and
talents fit with those of other members of the organization (Allen and Meyer
1990; Mitchell et al. 2001). In several organizational contexts, individuals’
perceived fit with their organization has been associated with their career
decisions; particularly, those who perceive low levels of fit (i.e., lack of con-
gruence with the norms and expectations of behavior) are more likely to seek
employment in a different context (Chatman 1989; Kristof 1996; Kristof-
Brown et al. 2005).

Relational Trust

Relational trust derives from the set of role relations characterizing the social
organization of schooling (Bryk and Schneider 2002), including teacher-
teacher relations, teacher-principal relations, and family–school staff relations.

32 American Journal of Education

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Pogodzinski, Youngs, and Frank

Tschannen-Moran and Hoy (2000) describe trust in schools as “one party’s

willingness to be vulnerable to another party based on the confidence that
the latter party is (a) benevolent, (b) reliable, (c) competent, (d) honest, and
(e) open” (556). When there are high levels of relational trust among teachers
within a school, the context is expected to be more conducive to collaboration
among teachers and effective decision making (Penuel et al. 2010). Further,
in an atmosphere in which it is perceived to be accepted to discuss feelings,
challenges, and shortcomings, novice teachers are more likely to seek help
from their colleagues.

Collective Responsibility

Collective responsibility has been defined as the extent to which teachers in

a school are committed to the goals of the school, particularly regarding student
learning (Bryk and Schneider 2002; Penuel et al. 2009). Researchers have
shown that ties between members of an organization are important in mo-
tivating collective action, such as supporting schoolwide reforms and providing
support and resources to one another (Granovetter 1978; Macy 1990). Par-
ticularly, individuals within an organization assess the level of involvement of
others around them and adjust their level of participation accordingly (Macy
1990). In other words, if a novice teacher perceives others around her as
investing their time and resources toward achieving a schoolwide goal, she is
more likely to invest her own time and resources. Conversely, if a teacher
perceives an insular environment, she is less likely to put effort toward achiev-
ing goals outside of her own personal goals. Therefore, teachers who identify
with the collective are more likely to direct their support and resources to
others independent of their specific social relations (Frank et al. 2004; Penuel
et al. 2009).
In sum, the framework used to guide this study suggests that the collegial
climate in a school (reflected in novices’ perceptions of the degree of profes-
sional fit, relational trust, and collective responsibility among teachers) in part
shapes the way in which novice teachers evaluate their work environment.
Therefore, it is expected that novice teachers’ perceptions of their work en-
vironment will influence their intent to remain teaching within their schools.

Methods for Data Collection and Analysis

This study draws on survey data to identify the association between novices’
perceptions of the collegial climate at their schools and their intent to remain
teaching within their schools. We sought to recruit medium to large districts

NOVEMBER 2013 33

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Collegial Climate

TABLE 1

2007–8 Enrollment and Demographic Information for Participating Districts

Free/Reduced
Student Enrollment Lunch Nonwhite
Category District (Approximate; %) (Approximate; %)
20,000–23,000 District A (MI) 60–65 75–80
District B (IN) 60–65 55–60
District C (MI) 10–15 10–15
15,000–19,999 District D (MI) 50–55 10–15
District E (IN) 40–45 55–60
10,000–14,999 District F (IN) 60–65 45–50
District G (IN) 55–60 55–60
District H (IN) 45–50 80–85
District I (MI) 25–30 15–20
7,000–9,999 District J (MI) 40–45 45–50
District K (MI) 35–40 45–50

in Michigan and Indiana (conveniently located near the researchers) that (a)
served varying student populations with regard to race/ethnicity and socio-
economic status and (b) had at least 10 full-time core content teachers in their
first 3 years of the profession. Data were collected within 99 schools across
11 districts (see table 1 for descriptive information about the districts).

Novice Teacher Sample

We invited all novice teachers (N p 380) to complete two surveys (one in the
fall of 2007 [October/November] and one in the spring of 2008 [April/May])
who met the following criteria: (a) were in the first 3 years of the teaching
profession and (b) taught in core content areas (i.e., general elementary or
math, science, English language arts, or social studies at the middle school
level). Overall, 184 novice teachers completed both the fall and spring surveys
(63% fall response rate and a spring retention rate of those teachers of 76%;
overall response rate of 48%). See table 2 for information on teacher char-
acteristics. It should be noted that this is a sample of convenience and is not
necessarily representative of the broader teacher population. At the same time,
this moderate sample size is sufficient to deepen our knowledge regarding the
association between collegial climate and novices’ intent to remain teaching.
Although the final novice teacher response rate was approximately 48%,
which could be cause for concern, there was no indication that nonrespondents
were significantly different from responders in observable characteristics on
the basis of descriptive information received from the districts (i.e., grade level

34 American Journal of Education

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Pogodzinski, Youngs, and Frank

TABLE 2

Novice Teacher Characteristics

District n Female White Elementary 1st Year 2nd Year 3rd Year
District A (MI) 10 9 9 8 3 7 0
District B (IN) 25 19 24 20 7 11 7
District C (MI) 13 10 12 8 0 8 5
District D (MI) 11 10 10 7 5 2 4
District E (IN) 11 7 8 8 2 5 4
District F (IN) 42 30 40 30 15 15 12
District G (IN) 23 18 19 15 6 10 7
District H (IN) 20 19 19 18 5 11 4
District I (MI) 8 8 8 2 0 4 4
District J (MI) 14 12 11 9 4 4 6
District K (MI) 7 6 6 7 2 5 0
All districts 184 83% 90% 72% 27% 44% 29%

taught and years of experience). We acknowledge, though, that nonobservable

factors may make nonresponders fundamentally different from responders; we
were not able to account for such factors. There were no statistically significant
differences in the baseline measure of intent to remain teaching or the measures
of collegial climate between those who completed only the fall survey and
those who completed both the fall and spring surveys.

Colleague Sample

Following the conceptual framework, it was important to identify the asso-

ciation between novice teachers’ perceptions of the collegial climate net of
any group effects (Enders and Tofighi 2007; Raudenbush and Bryk 2002). In
other words, novice teachers are located within specific groups of teachers
within a school that project their own beliefs and attitudes regarding the
organizational climate within the school. The extent to which an individual
novice teacher’s perceptions are associated with her intent to remain teaching
within that school may be conditional on the extent to which those perceptions
are aligned or diverge from those of other teachers within the school.
To account for contextual effects, we calculated group-level means of the
measures of collegial climate for each school included in the analysis. Spe-
cifically, in the fall 2007 survey, novice teachers were asked to name their
formal mentor and up to eight colleagues with whom they regularly interact.
Surveys were then sent to these colleagues to measure their perceptions of the
collegial climate within their schools. We were not able to distribute surveys
to all teachers within the schools because of resource constraints. Therefore,

NOVEMBER 2013 35

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Collegial Climate

the teacher colleague sample is not necessarily representative of the overall

population of teachers within these schools, but this sample represents key
colleagues in the novices’ work lives, which are important in measuring po-
tential contextual effects related to novices’ perceptions of the collegial climate.
In addition to the 184 novice teachers who completed both the fall and
spring surveys, a total of 351 of the novices’ more senior colleagues completed
a survey in the winter of 2008 (response rate of 60%). These colleagues were
also predominantly female and white (83% and 92%, respectively) and taught
at the elementary school level (approximately 60%). They averaged 16 years
of experience, with a minimum of 4 years and a maximum of 40 years (SD
9.87).

Analytic Approach

To identify both teacher-level and group-level associations between teachers’

perceptions of the collegial climate within their schools and novice teachers’
intent to remain teaching, a series of multilevel regression models were esti-
mated using restricted maximum likelihood (using the proc mixed command
in SAS 9.2) to model the nested structure of the data (teachers nested within
schools). We first estimated an unconditional model to identify the parameter
variance in the outcome variable, specifically identifying the extent to which
variance was concentrated among teachers and between schools. Following
the framework for random effects of multilevel models, error terms were
assumed to be distributed at the two levels: rij ∼ N(0, j 2 ), u 0j ∼ N(0, tp): level
1 model:
Yijt p b0j ⫹ rij ;
level 2 model:
b0j p g00 ⫹ u 0j ;
combined model:
Yijt p g00 ⫹ u 0j ⫹ rij , (1)
where Yijt is the measure of novice teacher i’s intent to remain teaching in
school j at time t (spring 2008), b0j is the school mean for measure of intent
to remain teaching, g00 is the grand mean, u 0j is the level 2 variance (school-
specific error term), and rij is the level 1 variance (teacher error term).
Second, we estimated a conditional two-level model to identify the asso-
ciation between novices’ perception of the collegial climate and their intent
to remain teaching within their schools. We also controlled for variables that
may be associated with selection into particular school contexts (Steiner et al.
2010), specifically related to both teacher and school attributes. Although

36 American Journal of Education

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Pogodzinski, Youngs, and Frank

teacher candidates often have limited information regarding schools that they
apply to for jobs (Liu and Johnson 2006), previous research has shown that
elements of the local context such as school and student characteristics, eco-
nomic factors, and other work conditions influence selection of workplace
(Boyd et al. 2005; Guin 2004; Hanushek et al. 2004). Therefore, we estimated
the following model: level 1 model:
Yijt p b0j ⫹ b1j X t⫺1 ⫹ b2j C ⫹ rij ;
level 2 model:
b0j p g00 ⫹ g01S ⫹ u 0j ,

b1j p g10 ,
b2j p g20 ;
combined model:
Yijt p g00 ⫹ g10 X t⫺1 ⫹ g20C ⫹ g01S ⫹ u 0j ⫹ rij . (2)
The model predicts the extent to which a novice teacher i in school j
indicated on the spring 2008 survey (t) intent to remain teaching at his or her
school as a function of his or her perceptions of the collegial climate within
the school taken from the fall 2007 survey (X t⫺1), time-invariant teacher at-
tributes (C), and school attributes (S). To measure the collegial climate, the
teachers were asked to report on the degree of their professional fit with their
colleagues, relational trust among teachers, and collective responsibility among
teachers within their schools. Regarding teacher attributes, we controlled for
novices’ gender, school level taught (elementary or middle school), first-year
status, whether or not they had a formal mentor, and whether or not the
teacher was new to the school. The continuous predictor variables entered in
at level 1 were group mean centered. At the school level, we controlled for
the percentage of students in the school who were eligible for free or reduced
lunch and the percentage of students who were white. The control variables
entered at the second level were grand mean centered.
In the next stage of analysis, we sought to examine possible contextual
effects related to perceptions of the collegial climate. In order to do this, we
estimated another model that included group-level means of the measures of
collegial climate at the school level. Data collected from novice and veteran
teachers were used to calculate the group-level school means. In this model,
we also controlled for the total number of responses per school (T ) and the
response rate per school (R) to help account for school differences in calculating
the group means. Level 1 model:
Yijt p b0j ⫹ b1j X t⫺1 ⫹ b2j C ⫹ rij ;
level 2 model:

NOVEMBER 2013 37

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Collegial Climate

b0j p b00 ⫹ g01S ⫹ g02X t⫺1 ⫹ g03T ⫹ g04R ⫹ u 0j ,

b1j p g10 ,
b2j p g20 ;
combined model:
Yijt p g00 ⫹ g10 X t⫺1 ⫹ g20C ⫹ g01S ⫹ g02X t⫺1 ⫹ g03T ⫹ g04R
(3)
⫹ u 0j ⫹ rij .
There were concerns regarding interpreting the directionality of the asso-
ciations between the independent variables and the dependent variable, as
well as a need to further address the fact that teachers often self-select into
schools for various personal and professional reasons. So although our focal
independent variables were measured in fall 2007 and the outcome variable
was measured in spring 2008, we estimated a fourth model in which we
controlled for a prior measure of intent to remain teaching measured in fall
2007 (Yt⫺1). Level 1 model:
Yijt p b0j ⫹ b1j Yt⫺1 ⫹ b2j X t⫺1 ⫹ b3j C ⫹ rij ;
level 2 model:
b0j p g00 ⫹ g01S ⫹ g02X t⫺1 ⫹ g03T ⫹ g04R ⫹ u 0j ,

b1j p g10 ,
b2j p g20 ;
combined model:
Yijt p g00 ⫹ g10 Yt⫺1 ⫹ g20 X t⫺1 ⫹ g30C ⫹ g01S ⫹ g02X t⫺1 ⫹ g03T
(4)
⫹ g04R ⫹ u 0j ⫹ rij .
By controlling for the prior measure of intent, we controlled for several
unobservable differences among individual teachers related to their career
decisions that existed prior to the spring 2008 data collection. In other words,
this prior measure of intent can be considered a cause of future intent as a
function of its persistent nature. For example, a teacher may enter the teaching
profession with the express intent of remaining in the profession for only a
short time (Smethem 2007). The persistent nature of this prior measure justifies
using it as a control rather than calculating a change score to use as the
dependent variable (Allison 1990).
Finally, we estimated a best-fitting model that produced the lowest deviance
score (⫺2#log likelihood). Specifically, we estimated a model with all inde-
pendent variables included and then removed variables one by one to examine
changes in the deviance score.
There were missing values that had to be accounted for in the analysis.

38 American Journal of Education

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Pogodzinski, Youngs, and Frank

Specifically, cases were dropped from the analysis if there was a missing value
for any of the variables used in the analysis. The number of observations used
in the analyses (n p 144 in 87 schools) was therefore reduced because of
missing data in the focal predictor variables and control variables. Cases
included in the analysis were, on average, more likely (a) to report higher
levels of collective responsibility among colleagues in their school ( p ! .05),
(b) to have been teaching in the same school as in the previous year ( p !
.05), and (c) to be located in schools where the group mean for reported
collective responsibility was higher ( p ! .05). For the remainder of the inde-
pendent variables and dependent variables, there were no statistically signif-
icant differences ( p ! .05) between this subsample and the larger sample.

Construction of Variables

A measure of novices’ intent to remain teaching within their schools was the
outcome variable of interest. Although intent to remain teaching does not
necessarily equate to actual career decisions, it has been shown to be highly
correlated with career decisions and is likely indicative of novices’ attitudes
and behaviors regarding their current work (Chan et al. 2008; Kushman 1992;
Ladd 2011; Mitchell et al. 2001). Therefore, in the spring 2008 survey the
novice teachers were asked to indicate their level of agreement (1 p strongly
disagree to 4 p strongly agree) with the following statements: (a) I would
prefer to continue working in this school the following year and (b) I could
see myself working in this school in 5 years. In part because the two measures
were highly correlated (r p .81), the measure of intent to remain teaching in
the school used in the analyses was the average of the two measures. Also,
novice teachers may be willing to “stick it out” for a year or two even if they
are generally unhappy in their work environment, so taking the average is a
more robust measure of their early career intentions to remain teaching within
their schools. These same questions were asked of novices in the fall 2007
survey, and this measure was used as a control variable (see table 3 for de-
scriptive information for each variable included in the analysis).
There were three specific aspects of the collegial climate that were the focus
of this study: perceptions of professional fit, relational trust, and collective
responsibility among colleagues. To measure perceptions of professional fit,
novices were asked to indicate the degree to which they agreed (1 p strongly
disagree to 4 p strongly agree) with six statements, from which a composite
variable was created by taking the mean response across the six measures.
There was a correlation a of .88 among the six measures, indicating a high
degree of internal validity. To measure teachers’ perceptions of relational trust
among teachers, they were asked to indicate the degree to which they agreed

NOVEMBER 2013 39

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
TABLE 3

Descriptive Information about Variables

Variable Description Mean SD
Intent (r p .81) Spring 2008: intent to remain teaching 3.264 .902
within school
Intent next yeara Spring 2008: I would prefer to continue 3.378 .894
teaching in this school next year
Intent 5 yearsa Spring 2008: I could see myself teaching 3.167 .989
in this school in 5 years
Prior intent (r p .84) Fall 2007: intent to remain teaching 3.337 .871
within school
Intent next yeara Fall 2007: I would prefer to continue 3.468 .810
teaching in this school next year
Intent 5 yearsa Fall 2007: I could see myself teaching in 3.270 .959
this school in 5 years
Fita (a p .88) 3.256 .464
Teaching fit Approach to teaching fits with others 3.359 .563
Interest fit Professional interests fit with others 3.151 .613
Identify Identify with others 3.280 .610
Goal fit Professional goals fit with others 3.068 .620
I matter I matter to other teachers 3.236 .570
Others matter Other teachers matter to me 3.441 .539
Trusta (a p .89) 3.167 .580
Discuss It’s ok to discuss feelings with others 3.379 .662
Teacher trust Teachers trust each other 2.993 .740
Respect leaders Teachers respect others who are leaders 3.129 .668
Respect experts Teachers respect others who are experts 3.261 .596
Collectiveb (a p .91) 3.692 .659
Discipline Help maintain discipline 3.618 .885
Help teachers Take responsibility for helping others 3.500 .836
Improve quality Take responsibility for improving quality 3.507 .939
Help students Feel responsible for helping students 3.713 .802
Expectations Set high expectations for academic work 3.846 .772
Ensure learning Feel responsible for ensuring students 3.979 .782
learn
Mentor Teacher has an assigned mentor (1 p .660 .475
yes, 0 p no)
Same school Teaches in the same school as the previ- .368 .484
ous year (1 p same school, 0 p new
to the school)
First year First-year teacher (1 p first year, 0 p .236 .426
second/third year)
Middle school Teaches middle school (1 p middle .264 .442
school, 0 p elementary school)
Female Female (1 p female, 0 p male) .833 .374

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Pogodzinski, Youngs, and Frank

TABLE 3 (Continued )

Variable Description Mean SD

School level:
Fit mean School-level group mean for measure of 3.314 .262
administrator-teacher relations as
poor
Trust mean School-level group mean for measure of 3.220 .419
adequate teaching resources
Collective mean School-level group mean for measure of 3.792 .490
manageable administrative duties
Total response Total number of responders by school 8.097 5.095
Response rate Response rate by school .611 .175
Percentage white Percentage of students in school who .486 .273
are white
Percentage lunch Percentage of students eligible for free/ .541 .215
reduced lunch
NOTE.—Means are based on responses of individuals included in the final analysis
(n p 144).
a
Based on the following scale: 1 p strongly disagree, 2 p disagree, 3 p agree, 4
p strongly agree.
b
Based on the following scale: 1 p none, 2 p less than half, 3 p about half, 4 p
most, 5 p all.

(1 p strongly disagree to 4 p strongly agree) with four statements, from which

a composite variable was created by taking the mean response across the four
measures (correlation a p .89). To gauge teachers’ perceptions of the degree
of collective responsibility shared by the teachers in the school, they were
asked to indicate the proportion of teachers (1 p none, 2 p less than half,
3 p about half, 4 p most, and 5 p all) who participated in six different
schoolwide activities, from which a composite measure was created by taking
the mean response across the six measures (correlation a p .91). The items
used to measure perceptions of professional fit, relational trust, and collective
responsibility were all derived from the work of Bryk and Schneider (2002).
Correlations among the collegial climate variables are reported in table 4.
The survey instrument is available on request.
In the fall 2007 survey, the novice teachers were also asked to indicate
whether or not they had a formally assigned mentor (1 p mentor assigned,
0 p mentor not assigned), whether or not they were teaching in the same
school as the previous year (1 p taught in the same school, 0 p taught at
a different school/did not teach), the school level they taught (1 p middle
school, 0 p elementary school), whether or not they were in their first year
of teaching (1 p first year, 0 p second/third year), and their gender (1 p
female, 0 p male). Measures of school attributes (i.e., percentage of students

NOVEMBER 2013 41

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Collegial Climate

TABLE 4

Correlations among Measures of Collegial Climate

Variable (1) (2) (3)
1. Fit 1.000 .542 .377
2. Trust .542 1.000 .611
3. Collective .377 .611 1.000

who were white and percentage who were eligible for free/reduced-price
lunch) were taken from the National Center for Education Statistics’ Common
Core of Data (2007–8).
In the second model, we included the school-level group means for each
of the three collegial climate measures in the model to examine group asso-
ciations, and we also controlled for the total number of responses per school
as well as the response rate per school. The mean number of responses per
school was 4.84 (SD 0.90) with a range of 1–18, and the mean response rate
was 0.54 (SD 0.10) with a range of 0.46–0.79.

Findings

Distribution of Variance

We began the analysis by calculating the variance parameters for the uncon-
ditional model in order to identify the distribution of variance among teachers
and between schools. Column 1 in table 5 shows the variance components
for the unconditional model. While approximately 70% of the variance is
among teachers, there is statistically significant variation between schools based
on the statistical significance of the variance component (tp p 0.240). This
indicates that multilevel modeling is appropriate as there are likely organi-
zational factors that are important in influencing novices’ intent to remain
teaching within a given school.

Teacher-Level Associations between Collegial Climate and Intent to Remain

Teaching

Column 2 in table 5 reports the results from estimating model (2), focusing
on the teacher-level association between novices’ perceptions of the collegial
climate and their intent to remain teaching. The coefficient for the measure
of professional fit was positive and statistically significant ( p ! .05), indicating

42 American Journal of Education

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
TABLE 5

Estimated Effects of Perceived Fit on Intent to Remain Teaching in a Particular School

Variable (1) (2) (3) (4) (5)
Intercept 3.289*** 3.412*** .615 .731 .809
(.084) (.238) (1.029) (1.004) (.938)
Prior intent .708*** .779***
(.091) (.081)
Fit .447*** .575*** .321*** .384***
(.208) (.206) (.159) (.125)
Trust .141 .128 .129
(.208) (.203) (.155)
Collective .412*** .360*** .114
(.173) (.169) (.135)
Mentor ⫺.183 ⫺.229 ⫺.182
(.148) (.145) (.115)
Same school .187 .224 .212* .215*
(.154) (.154) (.123) (.110)
First year .359* .403** .194
(.173) (.171) (.136)
Middle school ⫺.194 ⫺.041 ⫺.071
(.192) (.198) (.187)
Female ⫺.055 ⫺.109 ⫺.033
(.184) (.181) (.141)
Fit mean .273 .333 .364
(.365) (.354) (.351)
Trust mean .219 .214 .375*
(.288) (.276) (.222)
Collective mean .313 .219
(.234) (.227)
Total response ⫺.010 ⫺.006
(.017) (.018)
Total response rate .486 .553 .513
(.482) (.463) (.449)
Percentage lunch ⫺.252 .289 .175 .076
(.461) (.456) (.450) (.449)
Percentage white .249 .476 .404 .363
(.367) (.375) (.371) (.364)
j2 .560*** .437*** .431*** .204*** .191***
(.092) (.082) (.079) (.041) (.036)
tp .240*** .235*** .177** .335*** .364***
(.094) (.096) (.086) (.084) (.083)
Deviance score 371.8 349.2 342.3 305.3 292.4
n 144 144 144 144 144

* Statistically significant at p ≤ .10.

** Statistically significant at p ≤ .05.
*** Statistically significant at p ≤ .01.

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Collegial Climate

that, on average, novices who report having higher levels of professional fit
with their colleagues also report greater intent to remain teaching within their
school, ceteris paribus. While the coefficient for the measure of perceptions
of relational trust among teachers is positive, it lacks statistical significance,
and therefore, the association with the outcome variable is not distinguishable
from zero. The coefficient for the measure of perceived collective responsibility
among teachers is positive and statistically significant ( p ! .05), indicating that
those novices who perceive higher levels of collective responsibility among
teachers reported greater intent to remain teaching within their schools. Ap-
proximately 22% of the variation in intent to remain teaching among teachers
was accounted for in this model. Additionally, the deviance score was reduced
to 349.2, which represents a statistically significant improvement in model fit
based on the chi-square distribution (x 2 p 22.6; critical x 2 p 18.31 for p !
.05 and df p 10).

School-Level Group Associations between Collegial Climate and Novices’ Intent to

Remain Teaching

Column 3 in table 5 reports the results from estimating model (3), which
included group-level means of the three collegial climate measures. At the
teacher level, the coefficient for the measure of novices’ perceived professional
fit with colleagues remained positive and statistically significant ( p ! .01), as
did the coefficient for the measure of perceived collective responsibility among
teachers ( p ! .05). At the group level, none of the coefficients for the mean
measures of collegial climate reached statistical significance. This suggests that
novices who are located within groups of teachers who, on average, perceive
a more positive collegial climate (marked by high levels of professional fit,
relational trust, and collective responsibility) are not likely to report greater
intent to remain teaching within their schools compared to novices who are
located within groups of teachers who perceive a less positive collegial climate.
The inclusion of these group-level measures, though, did account for ap-
proximately 24% of the variation in teachers’ intent to remain teaching be-
tween schools in novices’ intent. The deviance score was also reduced to 342.3
(x 2 p 6.9); therefore, this model represents a statistically significant improve-
ment in model fit over model (2) (critical x 2 p 5.99 for p ! .05 and df p 2).

Controlling for Prior Intent

Column 4 in table 5 reports on the results from estimating model (4), which
was identical to model (3) with the exception that the prior measure of intent

44 American Journal of Education

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Pogodzinski, Youngs, and Frank

was included as a control variable. As expected, the coefficient for the prior
measure of intent to remain teaching (fall 2007) was statistically significant
( p ! .01). After we controlled for the prior measure of intent, of the focal
variables, only the measure of perceived fit was statistically significant ( p !
.05). None of the group-level coefficients for the measures of collegial climate
reached statistical significance by traditional standards.
As expected, the inclusion of the prior measure of intent to the model
accounted for a significant amount of variation in parameter variance among
teachers. Specifically, the independent variables included in model (4) ac-
counted for approximately 63% of the variation among teachers. The deviance
score was reduced to 305.3 (x 2 p 37); therefore, not surprisingly, this model
represents a statistically significant improvement in model fit over model (3)
(critical x 2 p 3.84 for p ! .05 and df p 1).

Best-Fitting Model

While the inclusion of the prior measure of intent greatly improved model
fit, given the large number of non–statistically significant variables in model
(4), we sought to estimate the best-fitting model on the basis of comparisons
of the deviance statistic. At the teacher level, this final model included as
independent variables the prior measure of intent to remain teaching, per-
ceptions of professional fit, and the measure of whether the novice was teaching
in the same school as in the previous year. At the group level, the model
included the mean measure of perceived fit and the mean measure of perceived
relational trust among teachers, a control for the response rate per school,
and the measures of the percentage of students who were eligible for free or
reduced-priced lunch and the percentage of students who were white.
The coefficient for the teacher-level measure of professional fit remained
positive and statistically significant ( p ! .01 ), again indicating that, on average,
those novices who reported higher levels of professional fit with their colleagues
had higher reported levels of intent to remain teaching, ceteris paribus. Besides
the measure of fit and the measure of prior intent to remain teaching, none
of the other measures were statistically significant ( p ! .05 ). It should be noted,
though, that the coefficient for the group-level measure of relational trust was
positive and statistically significant at a lower threshold ( p ! .10), suggesting
that when novice teachers are located within groups in which there are higher
perceptions of relational trust among teachers, they are more likely to report
intention to remain teaching within their given school. This final model,
though, further reduced the deviance statistic to 292.4 (x 2 p 12.9), and al-
though it did not reach statistical significance for improvement over model

NOVEMBER 2013 45

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Collegial Climate

(4) (critical x 2 p 15.51 for p ! .05 and df p 8), it did approach statistical
significance at a lower threshold ( p p .1153).

Discussion

By examining novices’ perceptions of their professional fit with their colleagues,

levels of relational trust among colleagues, and levels of collective responsibility
among colleagues, this study builds on previous research by examining not
only teacher-level perceptions but also contextual elements of the collegial
climate. The findings suggest that the ways in which novice teachers perceive
the collegial climate are important in influencing their intended career de-
cisions. The findings are of particular interest in that the measure of novices’
perceptions of their professional fit with their colleagues emerged as statistically
significant ( p ! .01) in the best-fitting model, which included the prior measure
of intent to remain teaching.

Collegial Climate and Teacher Interactions

Novice teachers’ perceptions of the collegial climate in their schools are largely
developed through interactions with both their formal mentors and other
colleagues (Kapadia et al. 2007; Smith and Ingersoll 2004). In other words,
it is through these interactions that novice teachers understand and evaluate
the beliefs and practices of their teacher colleagues. This has implications for
how school leaders (both administrators and teacher leaders) can influence
the collegial climate among teachers through both formal mechanisms (e.g.,
formal mentoring) and promotion of less formal collaboration among teachers.
Formal mentors can play an important role in transmitting expectations
for behavior, reflecting the collegial climate of the school through commu-
nicating their beliefs and practices and connecting novices to networks of
support throughout the school (Feiman-Nemser 2010; Feiman-Nemser et al.
1999). Researchers have shown that the portrayal of a positive climate is more
likely when novice teachers and their formally assigned mentors are well
aligned by grade level or subject matter area because they are more likely to
engage in frequent interactions and are often better able to provide support
and resources that shape novices’ attitudes and behaviors (Achinstein et al.
2004; Grossman and Thompson 2004; Smith and Ingersoll 2004; Youngs
2007). Additionally, mentoring is more likely to be successful in influencing
novice teacher commitment to the organization if the mentor’s beliefs and

46 American Journal of Education

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Pogodzinski, Youngs, and Frank

practices reflect a sense of collective responsibility inherent in the school’s

collegial climate (Feiman-Nemser et al. 1999).
Interactions with other teacher colleagues also influence the socialization
of novice teachers, thus influencing their perceptions of the organizational
climate (Bidwell 2001; Frank et al. 2004). Novice teachers enact agency when
self-selecting into particular relationships that affects the degree to which they
are open to the socialization efforts of the organization (Bryk and Schneider
2002; Lounsbury and Ventresca 2003; Powell and Colyvas 2007). They are
likely to enter into relationships with individuals with whom they share both
professional and personal interests, and these relationships communicate their
own unique set of norms and expectations (Coleman 1988). Additionally,
different types of informal teacher relations will mediate the effects of district
and school policies (Bryk and Schneider 2002; Coburn and Russell 2008;
Granovetter 1978). In other words, many elements of the collegial climate
may be resistant to exogenous forces simply because climate is inherent in the
everyday beliefs and practices of teachers throughout the school that often
operate disconnected from formal policy.
Administrators concerned about the quality of collegial climate, though,
can take advantage of the social structures within their schools to ensure that
novice teachers have greater access to resources and support (Friedkin and
Slater 1994; Lounsbury and Ventresca 2003; Powell and Colyvas 2007). One
way to do this is to manipulate the structural elements of the organization,
such as class schedules, room assignments, or team assignments (Crow and
Pounder 2000). These structural elements can be organized to increase the
probability that novice teachers will interact with several teachers across a
school. Additionally, administrators can more effectively communicate the
goals and values of the school through these networks, particularly when they
facilitate ties among teachers that span naturally forming subgroups, thus
encouraging fit among teachers and a sense of collective responsibility (Coburn
and Russell 2008; Granovetter 1978; Macy 1990).
Although administrators largely shape the organizational climate (Denison
1996; Lewin 1951), teacher leaders can also influence this climate. These
leaders often emerge informally as they are seen as competent by their col-
leagues, have the personality to relate to a diverse group of teachers, and
identify with the collective (Penuel et al. 2009; Spillane 2006). Teacher leaders
are charged with creating a collegial climate that promotes greater professional
alignment and a sense of collective responsibility among teachers throughout
a school. Therefore, encouraging collaboration among all teachers within a
school likely influences not only individual teachers’ retention decisions but
also teachers’ engagement with their schools’ mission (Granovetter 1978; Kris-
tof 1996; Kristof-Brown et al. 2005; Macy 1990).

NOVEMBER 2013 47

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Collegial Climate

Limitations and Implications for Future Research

There are limitations associated with this study that can be addressed by future
research to advance what we know about influences on teachers’ career de-
cisions. As the nature of our data did not allow us to make any causal claims
and focused on novice teachers’ intent to remain teaching, we recommend
that researchers seek to specifically identify causal relationships between mea-
sures of collegial climate and measures of teacher retention.

Teacher Retention

Although researchers have shown that intent to remain is highly correlated

with future career decisions (e.g., Chan et al. 2008; Kushman 1992; Ladd
2011; Mitchell et al. 2001), examining associations between collegial climate
and actual teacher retention may have broader practical and policy impli-
cations. To identify those relationships would likely require data collection
over multiple school years. Additionally, it is important to consider that in-
dividuals enter the teaching profession with different short- and long-term
career goals (Smethem 2007). In other words, a significant number of indi-
viduals enter the profession with the expressed interest of remaining a teacher
for only a very limited number of years. This suggests that it would be difficult
to accurately identify associations between specific organizational factors and
actual retention for this group of teachers. It is also important to point out
that some teacher attrition is desired, and researchers have shown that, on
average, less effective teachers are more likely to exit schools and the profession
(Boyd et al. 2011; Goldhaber et al. 2011; Hanushek and Rivkin 2010), though
it has not been shown whether or not the teachers who replace those who
leave are more or less effective on average (Ronfeldt et al. 2011).

Measuring Collegial Climate

The focal independent variables of interest in this study included measures of

professional fit, relational trust, and collective responsibility. Although these
represent important elements of the collegial context, they certainly are not
the only measures. Researchers should consider expanding the conceptuali-
zation of collegial climate to include other measures, such as measures of self
and collective efficacy (Bandura 1977). Researchers can also investigate the
content and frequency of teacher interactions throughout a school through
schoolwide social network analysis (Friedkin and Slater 1994; Penuel et al.

48 American Journal of Education

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Pogodzinski, Youngs, and Frank

2009; Pogodzinski 2012). Additionally, although the study we present here

was based on survey data, intensive qualitative research that investigates how
novice teachers shape their understanding of their school context can provide
some deep insight into the association between teachers’ perceptions of their
schools’ collegial climate and their career decisions (e.g., Flores and Day 2006;
Schaefer and Clandinin 2011).
More broadly, this study was limited in focusing on one aspect of organi-
zational climate. Several factors endogenous and exogenous to a school in-
fluence an individual teacher’s decision whether or not to remain teaching
within a particular school. For example, Hanushek et al. (2004) have shown
that student demographics and achievement status are a strong predictor of
teachers’ career decisions. Additionally, researchers have shown that teachers’
perceptions of support from their administration are often a stronger predictor
of teacher career decisions than student characteristics (Boyd et al. 2011;
Horng 2009; Ladd 2011). Finally, several of these organizational factors related
to teachers’ career decisions are also associated with the overall economic
condition of the school. While controlling for all these factors was beyond the
scope of this study, by controlling for a prior measure of intent in the final
two models, we accounted for many differences among teachers that are
manifest at the time the prior measure is obtained (Allison 1990). Additionally,
we did control for measures of students’ economic status and race at the school
level.

Identifying Causal Relationships

There is concern that teachers self-select into particular school environments

on the basis of personal and professional preferences and novices enter the
profession with different career trajectories in mind; this can make it difficult
to identify the direction of the association between collegial climate and nov-
ices’ intent to remain teaching within their schools. Teachers in the study may
have selected themselves into particular schools on the basis of expectations
as to what the working conditions would be like, even if they had limited
information (Liu and Johnson 2006). As a result, their perceptions of the
collegial climate may have formed before they actually began working in a
given school. Additionally, teachers in their second or third year of the pro-
fession may have transferred from one school to another as a function of their
preferences for work environment. We did attempt, though, to control for
selection into schools by controlling for the prior measure of intent (Allison
1990) and key teacher and school attributes known to be associated with job
preferences.

NOVEMBER 2013 49

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Collegial Climate

Conclusion

Despite these limitations, this study is significant for both researchers and
practitioners. First, it applied the notion of collegial climate to the field of
educational research, more generally, and to novice teachers, in particular.
While organizational climate is a common focus of empirical investigation in
research in organizational psychology and management, it has not been thor-
oughly used to understand how novice teachers experience the social systems
of their schools. Previous research on novice teacher socialization and retention
decisions has typically focused on mentors’ characteristics, grade level or con-
tent area matches between new teachers and mentors, and the content and
frequency of new teacher-mentor interactions (e.g., Glazerman et al. 2010;
Smith and Ingersoll 2004). In contrast to these studies, this study examines
how novice teachers experience the social context of their schools and, more
specifically, how they perceive the beliefs and practices of their teacher col-
leagues throughout their schools.
The findings from this study highlight the fact that perceptions of the overall
collegial climate within the school influence novice teachers’ intent to remain
teaching within their schools, which likely affects their future career decisions.

References

Achinstein, Betty, Rodney T. Ogawa, and Anna Speiglman. 2004. “Are We Creating
Separate and Unequal Tracks of Teachers? The Impact of State Policy, Local Con-
ditions, and Teacher Characteristics on New Teacher Socialization.” American Edu-
cational Research Journal 41 (3): 557–603.
Allen, Natalie J., and John P. Meyer. 1990. “Organizational Socialization Tactics: A
Longitudinal Analysis of Links to Newcomers’ Commitment and Role Orientation.”
Academy of Management Journal 33 (4): 847–58.
Alliance for Excellent Education. 2004. Tapping the Potential: Retaining and Developing High-
Quality New Teachers. New York: Carnegie Corporation.
Allison, Paul D. 1990. “Change Scores as Dependent Variables in Regression Analysis.”
Sociological Methodology 20 (1): 93–114.
Bandura, Albert. 1977. “Self-Efficacy: Toward a Unifying Theory of Behavioral
Change.” Psychological Review 84 (2): 191–215.
Bidwell, Charles. 2001. “Analyzing Schools as Organizations: Long-Term Permanence
and Short-Term Change.” Sociology of Education 74 (2): 100–114.
Boyd, Donald, Pamela Grossman, Marsha Ing, Hamilton Lankford, Susanna Loeb,
and James Wyckoff. 2011. “The Influence of School Administrators on Teacher
Retention Decisions.” American Educational Research Journal 48 (2): 303–33.
Boyd, Donald, Pamela Grossman, Hamilton Lankford, Susanna Loeb, and James
Wyckoff. 2006. “How Changes in Entry Requirements Alter the Teacher Workforce
and Affect Student Achievement.” Education Finance and Policy 1 (2): 176–216.
Boyd, Donald, Hamilton Lankford, Susanna Loeb, Matthew Ronfeldt, and James

50 American Journal of Education

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Pogodzinski, Youngs, and Frank

Wyckoff. 2011. “The Role of Teacher Quality in Retention and Hiring: Using
Applications to Transfer to Uncover Preferences of Teachers and Schools.” Journal
of Policy Analysis and Management 30 (1): 88–110.
Boyd, Donald, Hamilton Lankford, Susanna Loeb, and James Wyckoff. 2005. “The
Draw of Home: How Teachers’ Preferences for Proximity Disadvantage Urban
Schools.” Journal of Policy Analysis and Management 24 (Winter): 113–32.
Bryk, Anthony, and Barbara Schneider. 2002. Trust in Schools: A Core Resource for Im-
provement. New York: Russell Sage Foundation.
Chan, Wai-Yen, Shun Lau, Youyan Nie, Sandy Lim, and David Hogan. 2008. “Or-
ganizational and Personal Predictors of Teacher Commitment: The Mediating Role
of Teacher Efficacy and Identification with School.” American Educational Research
Journal 45 (3): 597–630.
Chatman, Jennifer A. 1989. “Improving Interactional Organizational Research: A
Model of Person-Organization Fit.” Academy of Management Review 14 (3): 333–49.
Coburn, Cynthia E., and Jennifer L. Russell. 2008. “District Policy and Teachers’
Social Networks.” Educational Evaluation and Policy Analysis 30 (3): 203–35.
Coleman, James S. 1988. “Social Capital in the Creation of Human Capital.” American
Journal of Sociology 94 (suppl.): S95–S120.
Crow, Gary M., and Diana G. Pounder. 2000. “Interdisciplinary Teacher Teams:
Context, Design, and Process.” Educational Administration Quarterly 36 (2): 216–54.
Denison, Daniel R. 1996. “What Is the Difference between Organizational Culture
and Organizational Climate? A Native’s Point of View on a Decade of Paradigm
Wars.” Academy of Management Review 21 (3): 619–54.
Enders, Craig, and Davood Tofighi. 2007. “Centering Predictor Variables in Cross-
Sectional Multi-level Models: A New Look at an Old Issue.” Psychological Issues 12
(2): 121–38.
Feiman-Nemser, Sharon. 2010. “Multiple Means of New Teacher Induction.” In Past,
Present, and Future Research on Teacher Induction: An Anthology for Researchers, Policy Makers,
and Practitioners, ed. Jian Wang, Sandra J. Odell, and Renee T. Clift. Lanham, MD:
Rowman & Littlefield and the Association of Teacher Educators.
Feiman-Nemser, Sharon, Sharon Schwille, Cindy Carver, and Brian Yusko. 1999. A
Conceptual Review of Literature on New Teacher Induction. Washington, DC: National Part-
nership for Excellence and Accountability in Teaching.
Feldman, Daniel C. 1989. “Socialization, Resocialization and Training: Reframing the
Research Agenda.” In Training and Development in Organizations, ed. Irwin L. Goldstein.
San Francisco: Jossey-Bass.
Flores, Maria Assunção, and Christopher Day. 2006. “Contexts Which Shape and
Reshape New Teachers’ Identities: A Multi-perspective Study.” Teaching and Teacher
Education 22 (2): 219–32.
Frank, Kenneth A., Yong Zhao, and Kathryn Borman. 2004. “Social Capital and the
Diffusion of Innovations within Organizations: The Case of Computer Technology
in Schools.” Sociology of Education 77 (2): 148–71.
Friedkin, Noah E., and Michael R. Slater. 1994. “School Leadership and Performance:
A Social Network Approach.” Sociology of Education 67 (2): 139–57.
Gilmer, B. Von Haller. 1966. Industrial Psychology. New York: McGraw-Hill.
Glazerman, Steven, Eric Isenberg, Sarah Dolfin, Martha Bleeker, Amy Johnson, Mary
Grider, Matthew Jacobus, and Melanie Ali. 2010. Impacts of Comprehensive Teacher
Induction: Final Results from a Randomized Controlled Study. Washington, DC: US De-
partment of Education, Institute of Education Sciences.
Goldhaber, Dan, Betheny Gross, and Daniel Player. 2011. “Teacher Career Paths,

NOVEMBER 2013 51

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Collegial Climate

Teacher Quality, and Persistence in the Classroom: Are Public Schools Keeping
Their Best?” Journal of Policy Analysis and Management 30 (1): 57–87.
Granovetter, Mark. 1978. “Threshold Models of Collective Behavior.” American Journal
of Sociology 83 (6): 1420–43.
Grossman, Pamela L., and Clarissa Thompson. 2004. “District Policy and Beginning
Teachers: A Lens on Teacher Learning.” Educational Evaluation and Policy Analysis 26
(4): 281–301.
Guarino, Cassandra, Lucrecia Santibanez, and Glenn Daley. 2006. “Teacher Recruit-
ment and Retention: A Review of the Recent Empirical Literature.” Review of Ed-
ucational Research 76 (2): 173–208.
Guin, Kacey. 2004. “Chronic Teacher Turnover in Urban Elementary Schools.” Ed-
ucation Policy Analysis Archives 12 (42), http://epaa.asu.edu/epaa/v12n42/.
Hanushek, Eric A., John Kain, and Steven G. Rivkin. 2004. “The Revolving Door:
A Path-Breaking Study of Teachers in Texas Reveals That Working Conditions
Matter More than Salary.” Education Next 4 (Winter): 77–82.
Hanushek, Eric A., and Steven G. Rivkin. 2010. “Constrained Job Matching: Does
Teacher Job Search Harm Disadvantaged Urban Schools?” Unpublished manu-
script, National Bureau of Economic Research, Cambridge, MA.
Horng, Eileen L. 2009. “Teacher Tradeoffs: Disentangling Teachers’ Preferences for
Working Conditions and Student Demographics.” American Educational Research Journal
46 (3): 690–717.
Hoy, Wayne, and Cecil Miskel. 2008. Educational Administration: Theory, Research, and
Practice. New York: McGraw-Hill.
Ingersoll, Richard, and Michael Strong. 2011. “The Impact of Induction and Men-
toring Programs for Beginning Teachers: A Critical Review of the Research.” Review
of Educational Research 81 (2): 201–33.
Johnson, Lisa S., Liam Goldrick, and Molly Lasagna. 2010. New Teacher Excellence: The
Impact of State Policy on Induction Program Implementation. Santa Cruz, CA: New Teacher
Center.
Johnson, Susan M. 2004. Finders and Keepers: Helping New Teachers Survive and Thrive in
Our Schools. San Francisco: Jossey-Bass.
Jones, Gareth R. 1986. “Socialization Tactics, Self-Efficacy, and Newcomers’ Adjust-
ments to Organizations.” Academy of Management Journal 29 (2): 262–79.
Kapadia, Kavita, Vanessa Coca, and John Q. Easton. 2007. Keeping New Teachers: A
First Look at the Influences of Induction in the Chicago Public Schools. Chicago: Consortium
on Chicago School Research, University of Chicago.
Kardos, Susan M., Susan M. Johnson, Heather G. Peske, David Kauffman, and Ed-
ward Liu. 2001. “Counting on Colleagues: New Teachers Encounter the Professional
Cultures of Their Schools.” Educational Administration Quarterly 37 (2): 250–90.
Kristof, Amy. 1996. “Person-Organization Fit: An Integrative Review of Its Concep-
tualizations, Measurement, and Implications.” Personnel Psychology 49 (1): 1–49.
Kristof-Brown, Amy, Ryan Zimmerman, and Erin Johnson. 2005. “Consequences of
Individuals’ Fit at Work: A Meta-Analysis of Person-Job, Person-Organization, Per-
son-Group, and Person-Supervisor Fit.” Personnel Psychology 58 (2): 281–342.
Kushman, James W. 1992. “The Organizational Dynamics of Teacher Workplace
Commitment: A Study of Urban Elementary and Middle Schools.” Educational Ad-
ministration Quarterly 28 (1): 5–42.
Ladd, Helen F. 2011. “Teachers’ Perceptions of Their Working Conditions: How Pre-
dictive of Planned and Actual Teacher Movement?” Educational Evaluation and Policy
Analysis 33 (2): 235–61.
Lewin, Kurt. 1951. Field Theory in Social Science. New York: Harper & Row.

52 American Journal of Education

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Pogodzinski, Youngs, and Frank

Litwin, George, and Robert Stringer. 1968. Motivation and Organizational Climate. Boston:
Harvard University, Graduate School of Business Administration.
Liu, Edward, and Susan M. Johnson. 2006. “New Teachers’ Experiences of Hiring:
Late, Rushed, and Information Poor.” Educational Administration Quarterly 42 (3): 324–
60.
Lortie, Dan. 1975. Schoolteacher: A Sociological Study. Chicago: University of Chicago
Press.
Lounsbury, Michael, and Marc Ventresca. 2003. “The New Structuralism in Orga-
nizational Theory.” Organization 10 (3): 457–80.
Macy, Michael W. 1990. “Chains of Cooperation: Threshold Effects in Collective
Action.” American Sociological Review 56 (6): 730–47.
Meyer, John W., W. Richard Scott, and Terrence Deal. 1981. “Institutional and Tech-
nical Sources of Organizational Structure: Explaining the Structure of Educational
Organizations.” In Organization and the Human Services, ed. Herman D. Stein. Phila-
delphia: Temple University Press.
Mitchell, Terence R., Brooks C. Holtom, Thomas W. Lee, Chris Sablynski, and Miriam
Erez. 2001. “Why People Stay: Using Job Embeddedness to Predict Voluntary
Turnover.” Academy of Management Journal 44 (6): 1102–22.
Penuel, William R., Margaret Riel, Aasha Joshi, Leslie Pearlman, Chong M. Kim,
and Kenneth A. Frank. 2010. “The Alignment of the Informal and Formal Or-
ganizational Supports for Reform: Implications for Improving Teaching in Schools.”
Educational Administration Quarterly 46 (1): 57–95.
Penuel, William R., Margaret Riel, Anne E. Krause, and Kenneth A. Frank. 2009.
“Analyzing Teachers’ Professional Interactions in a School as Social Capital: A
Social Network Approach.” Teachers College Record 111 (1): 124–63.
Pogodzinski, Ben. 2012. “Considering the Social Context of Schools: A Framework
for Investigating New Teacher Induction.” Mentoring and Tutoring: Partnership in Learning
20 (3): 325–42.
Poole, Marshall S. 1985. “Communication and Organizational Climates: Review, Cri-
tique, and a New Perspective.” In Organizational Communications: Traditional Themes and
New Directions, ed. Robert D. McPhee and Phillip K. Thompkis. Beverly Hills, CA:
Sage.
Powell, Walter W., and Jeannette A. Colyvas. 2007. “Microfoundations of Institutional
Theory.” In Handbook of Organizational Institutionalism, ed. Royston Greenwood, Chris-
tine Oliver, Kerstin Sahlin, and Roy Suddaby. New York: Sage.
Raudenbush, Stephen W., and Anthony Bryk. 2002. Hierarchical Linear Models: Applications
and Data Analysis Methods. Newbury Park, CA: Sage.
Ronfeldt, Matthew, Hamilton Lankford, Susanna Loeb, and James Wyckoff. 2011.
“How Teacher Turnover Harms Student Achievement.” Unpublished manuscript,
National Bureau of Economic Research, Cambridge, MA.
Schaefer, Lee, and D. Jean Clandinin. 2011. “Stories of Being Sustained: A Narrative
Inquiry into the Experiences of Two Beginning Teachers.” Learning Landscapes 4 (2):
275–95.
Smethem, Lindsey. 2007. “Retention and Intention in Teaching Careers: Will the New
Generation Stay?” Teachers and Teaching: Theory and Practice 13 (5): 465–80.
Smith, Thomas M., and Richard M. Ingersoll. 2004. “Reducing Teacher Turnover:
What Are the Components of Effective Induction?” American Educational Research
Journal 41 (3): 681–714.
Spillane, James. 2006. Distributed Leadership. San Francisco: Jossey-Bass.
Staton, Ann Q., and Sandra L. Hunt. 1992. “Teacher Socialization: Review and
Conceptualization.” Communication Education 41 (2): 109–37.

NOVEMBER 2013 53

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions
 Collegial Climate

Steiner, Peter M., Thomas D. Cook, William R. Shadish, and M. H. Clark. 2010.
“The Importance of Covariate Selection in Controlling for Selection Bias in Ob-
servational Studies.” Psychological Methods 15 (3): 250–67.
Strong, Michael. 2009. Effective Teacher Induction and Mentoring: Assessing the Evidence. New
York: Teachers College Press.
Tagiuri, Renato. 1968. “The Concept of Organizational Climate.” In Organizational
Climate: Explorations of a Concept, ed. Renato Tagiuri and George Litwin. Boston:
Harvard University, Graduate School of Business Administration.
Tschannen-Moran, Megan, and Wayne K. Hoy. 2000. “A Multidisciplinary Approach
to the Nature, Meaning, and Measurement of Trust.” Review of Educational Research
70 (4): 547–93.
Whalen, Frances. 2012. Collective Responsibility: Redefining What Falls through the Cracks for
School Reform. Rotterdam: Sense Publishers.
Youngs, Peter. 2007. “District Induction Policy and New Teachers’ Experiences: An
Examination of Local Policy Implementation in Connecticut.” Teachers College Record
109 (4): 797–837.

54 American Journal of Education

This content downloaded from 141.217.105.143 on Mon, 11 Nov 2013 15:07:12 PM

All use subject to JSTOR Terms and Conditions