http://soe.sagepub.com/Sociology of Education

http://soe.sagepub.com/content/86/2/174The online version of this article can be found at:

 DOI: 10.1177/0038040712472911

2013 86: 174 originally published online 10 February 2013Sociology of EducationBottia

Stephanie Moller, Roslyn Arlin Mickelson, Elizabeth Stearns, Neena Banerjee and Martha Ceciliaby Race, Ethnicity, and Socioeconomic Status

Collective Pedagogical Teacher Culture and Mathematics Achievement: Differences  

Published by:

http://www.sagepublications.com

On behalf of: 

  American Sociological Association

can be found at:Sociology of EducationAdditional services and information for    

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

 

http://soe.sagepub.com/subscriptionsSubscriptions:  

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

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

What is This? 

- Feb 10, 2013OnlineFirst Version of Record  

- Apr 4, 2013Version of Record >>

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

Collective PedagogicalTeacher Culture andMathematics Achievement:Differences by Race, Ethnicity,and Socioeconomic Status

Stephanie Moller1, Roslyn Arlin Mickelson1, Elizabeth Stearns1,Neena Banerjee1, and Martha Cecilia Bottia1

Abstract

Scholars have not adequately assessed how organizational cultures in schools differentially influence stu-dents’ mathematics achievement by race and socioeconomic status (SES). We focus on what we term col-lective pedagogical teacher culture, highlighting the role of professional communities and teachercollaboration in influencing mathematics achievement. Using cross-classified growth models, we analyzedata from the Early Childhood Longitudinal Study and illustrate that schools where teachers perceive thepresence of professional communities and teacher collaboration foster greater mathematics achievementthroughout elementary school. Furthermore, achievement gaps by race and socioeconomic status are less-ened in schools with professional communities and teacher collaboration.

Keywords

achievement, race, socioeconomic status, teacher collaboration, community, school culture

Most studies of educational organizations have

focused on structural features of schools, such as

classroom size, infrastructure, and resources.

Research on schools’ organizational cultures is

less widespread. Yet, the organizational cultures

of schools can have important implications for

teaching practices and student outcomes

(Gamoran, Secada, and Marrett 2000; Lee and

Smith 1996; Louis and Marks 1998; Vescio,

Ross, and Adams 2008). Schools’ organizational

cultures include the shared assumptions, rituals,

values, climate, and behaviors within organiza-

tions. Schools’ organizational cultures are critical

because they define how teachers interact with

one another and their students (Powers 2009).

Indeed, teachers are equipped with a broad range

of teaching tools, but the tools employed are often

constrained by the structural and cultural environ-

ment of schools (Powers 2009).

Education reformers and practitioners have at-

tempted to alter the cultures of schools through

localized reform efforts that foster the development

of professional communities among teachers

(Cuban 1990; Hess 1999; McLaughlin and

Talbert 2006; Tyack 1974; Tyack and Cuban

1995). Despite these efforts, scholars have not ade-

quately assessed the effect that schools’ organiza-

tional cultures have on achievement gaps by race,

ethnicity, and socioeconomic status (SES). This is

1Department of Sociology, University of North Carolina

at Charlotte, USA

Corresponding Author:

Stephanie Moller, Department of Sociology, University of

North Carolina at Charlotte, 9201 University City

Boulevard, Charlotte, NC 28032, USA

Email: smoller@uncc.edu

Sociology of Education86(2) 174–194

� American Sociological Association 2013DOI: 10.1177/0038040712472911

http://soe.sagepub.com

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

particularly problematic given large and well-

documented gaps in math achievement.

Despite localized reform initiatives that pro-

mote organizational cultures that value profes-

sional community, national reform efforts,

notably the No Child Left Behind Act (NCLB),

have undermined schools’ organizational cultures

by altering working conditions for teachers, princi-

pals, and students. High-stakes testing has created

an environment where the negative consequences

of student failure have placed immense pressure

on teachers, increasing stress and competition

among them (Crocco and Costigan 2006;

Lankford, Loeb, and Wyckoff 2002; Nichols and

Berliner 2007; Valli and Buese 2007). This compe-

tition undermines teachers’ abilities to build collab-

orative, student-centered lessons, as they must trust

one another to do so. In fact, the high-stakes testing

environment encouraged by NCLB has changed

schools’ organizational cultures by undermining

trust and reducing morale (Ravitch 2010). Given

that NCLB alters the organizational cultures of

schools, it is imperative that we clarify whether

schools’ organizational cultures are a critical piece

of the complicated solution to reducing early math-

ematics achievement gaps by race and SES. Our

research focuses on a sample of students that

entered kindergarten in 1998, prior to the imple-

mentation of NCLB. Yet our results will clarify

whether some of the negative consequences of

NCLB on the organizational cultures of schools,

found in other studies, have implications for stu-

dent achievement.

This study focuses on mathematics achieve-

ment because math is essential to students’ long-

term success as they become citizens and workers

in the twenty-first century (Moses and Cobb

2001). With the expansion of technology, workers

across occupational strata must employ mathemat-

ical concepts in their everyday lives (Burrill 2001).

We study achievement in the elementary years

because students who struggle at early ages con-

tinue to underachieve in the upper grades. And as

years pass, the achievement gaps that existed in

the early grades widen, and the need for interven-

tion grows (Lee, Grigg, and Dion 2007).

Our study is innovative in several respects.

Specifically, we assess components of what we

term collective pedagogical teacher culture—a

culture we find in schools where professional com-

munity and teacher collaboration are valued. Then,

we assess whether collective pedagogical teacher

culture influences students’ math trajectories in

elementary school. While prior studies have

offered important insights into mathematics

achievement, they have not presented a clear

view of how schools’ organizational cultures alter

students’ learning trajectories. Studies that have

focused on organizational cultures have limited

generalizability as their samples were small or pur-

posively selected (Gamoran et al. 2000).

Furthermore, studies have not assessed the extent

to which organizational cultures can reduce both

racial/ethnic and socioeconomic gaps in achieve-

ment; nor have they adequately established

whether components of collective pedagogical

teacher culture fit together conceptually.

We fill these gaps in the literature by assessing

the effects of organizational cultures on elementary

school students’ mathematics achievement trajec-

tories, focusing on racial and socioeconomic

achievement gaps. We analyze the mathematics

achievement of students sampled in the nationally

representative Early Childhood Longitudinal

Study (ECLS-K) through cross-classified growth

curve models, a methodologically sophisticated

technique (Cheadle 2008; Ladd and Dinella 2009;

Moller et al. 2006; Zvoch and Stevens 2006). The

theoretical and methodological advances of our

study result in a unique set of findings that clarifies

how students’ mathematics achievement trajecto-

ries and gaps in these trajectories are conditioned

by the perceived culture of schools.

SCHOOLS’ ORGANIZATIONALCULTURES

The culture of an organization defines the organiza-

tion. There are three broad components of organi-

zational culture: Artifacts reflect the visible and

identifiable structures, processes, and behaviors

in an organization; underlying assumptions reflect

unconscious and taken-for-granted beliefs and val-

ues in the organization; espoused beliefs and values

include values and norms that shape interactions

and expectations in the organization (Schein 2010).

The earliest studies of organizational culture

were qualitative, focusing on rich, deep description

of cultural evolution, organizational artifacts, and

underlying assumptions of organizations (Denison

1996; Schein 2010). More recently, this research tra-

dition has embraced quantitative and comparative

approaches. Researchers have increasingly focused

on shared values and norms across organizations

in that these values and norms are quantifiable

Moller et al. 175

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

(Denison 1996; O’Reilly, Chatman, and Caldwell

1991). Values and norms are essential to organiza-

tional identity and internal control systems for or-

ganizations (Black 2003; Pedersen and Dobbin

2006; Schein 2010). They undergird the technical

core of schools and prevent subcultures from under-

mining the mission of the organization (Schein

2010). Our research focuses on values and norms

inside elementary schools, understanding these to

be quantifiable manifestations of schools’ organiza-

tional cultures (Schein 2010).

Teachers are an integral part of generating and dif-

fusing organizational culture within schools. While

school administrators and leaders are generally

responsible for establishing cultural values within

schools, teachers must consent to and promote these

values (Kruse and Louis 2009; Schein 2010). In

fact, schools’ organizational cultures can only affect

student outcomes through teaching practices, and

teaching is the core function of schools (Gamoran

et al. 2000). This is particularly salient during the ele-

mentary years because elementary-age students are

more likely to conform to adult values than students

in higher grades, who tend to form their own distinc-

tive subcultures (Firestone and Louis 1999).

Therefore, we focus on teachers’ cultures as a compo-

nent of schools’ organizational cultures.

Much of the research that contributes to our

understanding of the organizational cultures of

schools has focused on professional learning com-

munities. The exact definition of these communi-

ties has varied across studies. Generally, they

have been referred to as communities where teach-

ers have a shared sense of purpose that is spear-

headed by a visionary principal; have a sense of

belonging, trust, and spirit; collaborate and reflect

on student learning; and are continually developing

professionally (Gamoran, Gunter, and Williams

2005; Kruse and Louis 2009; Louis and Kruse

1995; Louis, Marks, and Kruse 1996; Stoll et al.

2006).

We build on the professional learning commu-

nity literature by positing that schools with collec-

tive pedagogical teacher cultures are the most

effective at enhancing achievement and reducing

achievement gaps for students. We conceptualize

collective pedagogical teacher culture as a culture

where teachers perceive (1) the presence of strong

professional community and (2) a norm of collabo-

ration among teachers where students’ needs are

the focus.

While scholarship on professional communities

does not always conceptualize these communities

as cultures, there is a general understanding that

professional communities cultivate organizational

culture by creating shared languages, values, and

expectations among teachers and administrators

(McLaughlin and Talbert 2006).1 Generally, within

schools, the principal identifies the organizational

mission and communicates the mission to the fac-

ulty. The organizational culture is stronger if the

faculty agree on the mission. Organizational cul-

ture is also more community oriented if teachers

feel accepted by each other and if they have a sense

of pride or spirit. Prior scholarship has suggested

that teachers who are socialized into community-

oriented school culture (both professional and col-

legial) have a sense of belonging, attachment, and

pride (Anderson 1982; Wynne 1980).

An important attribute of professional commu-

nities within schools is an orientation toward learn-

ing. Scholars have suggested that professional

communities are most effective when teachers are

continually learning and searching for methods to

enhance their effectiveness (Little 1982;

McLaughlin and Talbert 2006; Patchen 2004;

Smey-Richman 1991; Yasumoto, Uekawa, and

Bidwell 2001). Thus, we contend that teachers

sense that they are part of professional communi-

ties when they perceive that there is an agreed

upon mission, school pride, an orientation toward

learning, and a sense of belonging.

This leads to the second main component of col-

lective pedagogical teacher culture: a norm of

teacher collaboration. Professional learning com-

munities lay the foundation for teachers to learn

from each other through collective understanding

as opposed to individualized, fragmented under-

standing and teaching (McLaughlin and Talbert

2006). Yet, a sense of community among teachers

does not necessarily translate into collaborative

teaching. Collaboration must be one of the norms

guiding the community. Collaboration reflects an

environment where teachers build their lessons

cooperatively, eliminating redundancy and

increasing compatibility across parts of the curric-

ulum and across grades. This allows teachers to

take collective responsibility for students, and it

permits teachers to interactively develop the best

strategies for teaching (Lee and Smith 1996;

Louis and Marks 1998; McLaughlin and Talbert

2006). Collaboration generates greater trust and

helps strengthen ties (Patchen 2004).

Yet, a collaborative, professional community

will not adequately serve the needs of students if

the community does not focus on students’ needs

176 Sociology of Education 86(2)

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

(Wood 2007). Here, teachers are more invested in

their students’ achievement, thereby increasing stu-

dents’ opportunities to learn. Students do better in

schools where teachers take collective responsibility

for students’ learning and collaboratively develop

interventions for individual students (Darling-

Hammond 2010; McLaughlin and Talbert 2006).

Therefore, collaboration among teachers where the

needs of students are prioritized is normative in a col-

lective pedagogical teacher culture.

Many studies incorporate collaboration in the

definition of professional learning communities,

yet we contend that teachers can perceive that

they are part of a community without collaborating

on lessons or centralizing individual students’

needs. We test this proposition to assess whether

professional community and teacher collaboration

are integrally linked together.2

SCHOOL CULTURE ANDMATHEMATICS ACHIEVEMENT

Components of collective pedagogical teacher cul-

ture are associated with greater teacher satisfaction,

greater accountability for student learning, a more

student-oriented environment, a more academi-

cally oriented student culture, and ultimately

higher achievement among students (Bryk and

Driscoll 1988; Eilers and Carnacho 2007; Langer

2000; Lee and Smith 1996; Louis et al. 1996;

Louis and Marks 1998; MacNeil, Prater, and

Busch 2009; Smey-Richman 1991). In essence,

professional and collaborative communities alter

teaching practices, leading to greater learning

opportunities for students (Louis and Marks 1998;

Newmann, Marks, and Gamoran 1996; Strahan

2003).

Prior research has clarified the importance of

schools’ organizational cultures for enhancing stu-

dents’ academic outcomes, including mathematics

achievement. However, these studies were based

on small selective samples, limiting their generaliz-

ability (Comer and Emmons 2006; Eilers and

Carnacho 2007; Strahan 2003; Supovitz 2002; see

also Gamoran et al. 2000; Vescio et al. 2008). Two

notable exceptions to this were Bryk and Driscol’s

(1988) study of high schools in the 1980s and

Lee, Smith, and Croninger’s series of studies (Lee

and Smith 1995, 1996; Lee, Smith, and Croninger

1997), but these studies were also focused on high

schools. Prior research has documented that learning

trajectories are largely established prior to high

school (Moller et al. 2011). Therefore, we focus on

the elementary years as a critical period in which stu-

dents are potentially propelled toward success.

Finally, we contribute to the literature by exam-

ining whether schools with collective pedagogical

teacher culture help reduce racial and socioeco-

nomic gaps in achievement. This is an important

line of inquiry given the national policy emphasis

on achievement gaps and the theoretical impor-

tance of establishing how organizational culture

can create more equitable learning environments.

Indeed, in prior research, Lee and Smith (1996)

found that professional communities can establish

more equitable learning environments as they can

help overcome the effect of SES on achievement

in high school; yet it remains unclear how profes-

sional communities might differentially affect

race-specific or race-by-SES–specific achievement

trajectories. Our limited knowledge of the impact

of professional communities on students’ achieve-

ment trajectories is particularly problematic since

school reform movements have emphasized the

importance of altering professional development

and organizational culture (Desimone 2002). To

better understand the effect of organizational cul-

ture on students’ achievement, we assess how com-

ponents of collective pedagogical teacher culture

interact with students’ race and SES to shape math-

ematics achievement trajectories.

We study mathematics trajectories for distinct

racial/ethnic and class groupings because scholars

have convincingly argued that in the modern econ-

omy race and class are intertwined. Indeed, there

are racial divisions within classes and class divi-

sions within races (Reardon and Galindo 2009;

Wilson 1978a, 1978b, 2011). Scholars have

debated the extent to which racial divisions within

the middle and upper classes have declined in the

modern economy. Nevertheless, there is broader

agreement that racial divisions persist among lower

socioeconomic individuals because they are

plagued by both racial and class-based isolation

(Freeman 2008; Oliver and Shapiro 2006; Wilson

1978a, 2009, 2011). These divisions generate and

reflect unique experiences and opportunities across

socioeconomic racial groups, experiences that

begin in early childhood.

Our research is in part a theoretical test, as prior

scholarship has established that components of col-

lective pedagogical teacher culture should augment

mathematics achievement. Our research is also

descriptive. We explain how this effect varies

across socioeconomic and racial/ethnic categories.

Moller et al. 177

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

The extant literature does not generate race- and

class-specific hypotheses. One could conjecture

that the racial and ethnic groups might have similar

responses to exposure to collective pedagogical

teacher culture in the middle and upper classes,

while responses may be divergent among the lower

class.

DATA AND METHODS

We analyze data from the Department of

Education’s Early Childhood Longitudinal Study.

This study began in 1998 with a nationally represen-

tative sample of 19,680 kindergarteners. Most stu-

dents were administered follow-up surveys when

they were in the first, third, fifth, and eighth grades.

In each wave students were tested and parents,

teachers, and school administrators were surveyed,

making this an ideal data set to examine students’

achievement trajectories in light of classroom and

school characteristics. Students are included in our

sample if they participated in the first four waves

of data collection; this yields 10,670 students. We

exclude the eighth-grade wave because this wave

does not include the same measures of organiza-

tional culture as the previous waves.

Sample attrition is normal in longitudinal

research. In this study, it reflects, in many cases,

students changing schools (only 50 percent of stu-

dents were followed into their new schools), stu-

dents moving out of the country, and parental

refusal to cooperate (Tourangeau et al. 2009).

Given attrition, nonresponse bias is small but mea-

surable in the ECLS-K data, but it is minimized

with the appropriate weights (Tourangeau et al.

2009); panel weights are applied in all analyses.

Given our research interests, we limit the sam-

ple to white, black, and Latino/a students. This nar-

rows our sample to 8,700 students (68 percent

white, 12 percent black, and 20 percent Latino/a).

We also limit our sample to students who attend

public schools because prior research has found

that organizational culture is a defining feature of

the divergence between Catholic and public

schools (Bryk, Lee, and Holland 1993). Dropping

private schools from the analysis ensures that our

results are not driven by this private school effect.

Finally, we limit our sample to students who attend

schools where both the teacher and the school

administrator complete questionnaires.

Additional missing data are imputed through mul-

tiple imputation as this approach is far superior to list-

wise deletion of missing data (Allison 2002; Schafer

1997).3 Data are imputed within waves to ensure that

the efficiency of imputation is not compromised by

attrition. To ensure high efficiency, the researchers

determined a priori to only impute variables that are

missing less than 20 percent of cases within waves.

Most of our variables have less than 10 percent miss-

ing data and are thus imputed. However, a potential

control variable, percentage of students on free and

reduced price lunch, has more than 20 percent miss-

ing data. This variable is not imputed. Instead, the

variable is tested in separate analysis (using a smaller

sample) to ensure that the results are robust. Those re-

sults are footnoted in the results section. The imputa-

tion is more than 93 percent efficient for all imputed

variables in all waves.

Our final sample includes 4,490 white, black,

and Latino/a students who attended public elemen-

tary schools between 1998 and 2003. Comparing

this final sample to the initial sample of black,

white, and Latino/a students, the students are com-

parable in race (13 percent black and 17 percent

Latino/a), socioeconomic status (30 percent of the

final sample are lower SES, compared to 34 percent

prior to sample selection, and 32 percent are higher

SES in the final sample, compared to 34 percent in

the original sample), and math scores (the average

kindergarten and fifth-grade scores were 37.5 and

124.7 in the initial sample, and they are 36.4 and

124.2, respectively, in the final sample). As these

descriptive statistics indicate, the final sample is

not substantially different from the initial sample

in terms of race, SES, and achievement.

Collective Pedagogical Teacher Culture

Before turning to the analysis of mathematics

achievement, we will describe dimensions of col-

lective pedagogical teacher culture. We conceptu-

alize collective pedagogical teacher culture as

a workplace environment where teachers perceive

(1) a strong community orientation and (2) teacher

collaboration. A strong professional community is

measured with five variables: (1) Teachers have

school spirit, (2) leadership has communicated

a school mission, (3) teachers agree on a school

mission, (4) teachers feel accepted and respected

as a colleague, and (5) teachers are constantly

engaged in learning. The norm of teacher collabo-

ration is calculated with three variables that mea-

sure the extent to which individual teachers

perceive that within the school, their colleagues:

(1) collaborate on lesson planning, (2) collaborate

on curriculum development, and (3) meet to discuss

178 Sociology of Education 86(2)

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

children. Each of these variables is gathered from

the teacher questionnaire, and they are described

in Appendix A (online at soe.sagepub.com).

We measure collective pedagogical teacher cul-

ture through exploratory factor analysis (EFA). We

use exploratory factor analysis because the extant

literature does not clearly articulate the degree to

which professional communities are collaborative.

There is an assumption in the literature that they

flow together, but we test this assumption because

community does not necessarily generate collabo-

ration. We test this assumption through maximum

likelihood exploratory factor analysis with promax

rotation.4 Each of these variables is ordinal

(Appendix A lists values), and therefore we ran

the EFA on a polychoric correlation matrix. Our

analysis produced two factors. The first factor rep-

resents strong professional learning communities

(as the first five measures presented in Appendix

A have a moderate to strong loading on this factor),

and the second factor represents collaborative,

child-oriented planning among teachers (the final

three measures have a moderate to strong loading

on this factor). We call our factors Professional

Community and Teacher Collaboration.

The lack of correspondence between the collab-

oration variables and the remaining community

variables suggests that schools with a professional

learning community are not necessarily collabora-

tive. This is an important finding given that prior

research suggests that schools must have all of

the components of collective pedagogical teacher

culture to be effective (Wood 2007).5

Mathematics Achievement

Our dependent variable is achievement scores in

mathematics between kindergarten and the fifth

grade. We use item response theory (IRT) scale

scores because these scores permit evaluation of

achievement trajectories over time even though

the tests changed to reflect age-appropriate meas-

ures. The IRT math scores assess the probability

of a correct response by estimating the number of

correct answers expected if the students had

answered all questions for the math test in all waves

(Tourangeau et al. 2009). Using these scores ena-

bles us to examine growth over time.

Additional Independent Variables

Student race and SES. We measure race and

SES using kindergarten data. Race/ethnicity is

coded as white, black, and Latino/a. SES is coded

as low (in the bottom third), middle (in the middle

third), and high (in the top third). We chose this

coding scheme to allow us to combine race/ethnic-

ity and SES into categories with sufficiently large

samples in each time period. SES is a composite

of five variables: father’s education and occupa-

tion, mother’s education and occupation, and

household income.6

Control variables. In all models, we control for

variables correlated with math scores and our pri-

mary independent variables (see Appendix B, on-

line at soe.sagepub.com, for descriptive statistics).

Our time-invariant controls include gender,

English as a second language in kindergarten, and

student’s approach to learning in kindergarten.

English language learners clearly face more chal-

lenges in the classroom than do other students

(Gersten and Baker 2000). A student’s approach

to learning has been found to be one of the stron-

gest predictors of mathematics achievement and

must be controlled in the models (Bodovski and

Farkas 2007; Singh, Granville, and Dika 2002). It

is measured as a scale downloaded from the origi-

nal data set and measures behaviors that affect

learning, including child’s attentiveness, task per-

sistence, eagerness to learn, learning independence,

flexibility, and organization. This measure is pulled

from the kindergarten survey because measures of

collective pedagogical teacher culture should have

an impact on students’ learning. Thus, we control

for kindergarten orientation toward learning to con-

trol for students’ orientation when they enter

school. Each of these time-invariant variables is

interacted with time in the analysis to account for

achievement trajectories of students of different

genders, English language status, and learning

approaches in kindergarten.

We also include time-variant controls, including

same race (coded 1 if student and teacher are of the

same race or ethnicity), school size (logged), percent-

age of students in the school who are white, teacher’s

highest education (coded 1 for master’s degree, edu-

cation specialist, or doctorate), teacher enjoys teach-

ing (1 = yes), region (South is excluded), and rural/

suburban (urban is excluded). We control for same-

race teacher because research shows that same-race

teachers are better role models and are a source of

academic and socioemotional support—especially

for minority students—through their interactions,

instructional mechanisms, and time allocation. As

a result, these students may flourish in schools, as

seen in their course of study or overall achievement

Moller et al. 179

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

(Dee 2004; Ehrenberg, Goldhaber, and Brewer 1995).

We control for school size because it influences equi-

table distributions of learning among students

through its effect on class size (Lee and Smith

1997). The variable measuring percentage of students

in a school who are white helps to capture the effects

of diversity in the student body, an important compo-

nent of student learning outcomes (Rumberger and

Palardy 2005). We control for teacher education since

it reflects teacher quality, which has been found to

affect student achievement (Hanushek and Rivkin

2010; Nye, Konstantopoulos, and Hedges 2004).

Finally, we control for teacher satisfaction as it affects

teachers’ organizational commitment and student

learning (Ostroff 1992; Park 2005). Time-varying

control variables are centered around their grand

means.

Given that students’ achievement trajectories

should reflect their cumulative experiences

throughout their academic trajectory, teacher- and

school-level variables are lagged for each student.

While the literature suggests that a lag is necessary

because students’ experiences cumulate over time

and because school and teacher effects are long-

lasting (Krueger and Whitmore 2001), the degree

of the lag necessary continues to spark lively debate

(Kane and Staiger 2008; Konstantopoulos 2007,

2008). Therefore, we empirically identify the lag

by fitting a series of exponential decay curves to

the data:

decay 5 100 � e�:tt;

where t is the rate of decay and t reflects time

elapsed. Decay curves are widely used across the

physical and social sciences to explain fade-out

of a variety of phenomena. In fact, Kane and

Staiger’s (2008) analysis of teacher effects on stu-

dent achievement suggests a constant rate of

fade-out at 50 percent (or 50 percent at t = 1 and

25 percent in t = 2). This rate is reflective of a decay

curve where the rate of decay is .69. In separate

analysis (not shown), we test decay rates ranging

from t = .001 (no decay) to t = 1. We compare fit

statistics across models and find that model fit be-

gins to decline once we reach a decay of t = .5.

Based on the application of the decay (see

Appendix C online at soe.sagepub.com, for

a detailed discussion of the method for applying

the decay), our cumulative lag variables are calcu-

lated as follows. There is no lag given that kinder-

garten is the first year—100 percent of the lag

variables in kindergarten are based on kindergar-

ten. In the first grade, the lag variables are

calculated as 61 percent kindergarten and 39 per-

cent first grade. The third-grade values are calcu-

lated as 29 percent third grade, 39 percent first

grade, and 32 percent kindergarten. For the fifth

grade, the cumulative lag effects are calculated as

26 percent fifth grade, 33 percent third grade, 23

percent first grade, and 18 percent kindergarten.

This measurement allows students’ school experi-

ences to accumulate over time. In this way, meas-

ures of collective pedagogical teacher culture

reflect students’ experiences with teachers, as

opposed to the effects of individual teachers. By

creating a lag of our key variables, we can better

establish a causal link between measures of organi-

zational culture and mathematics achievement.

Analytic Technique

We use cross-classified growth modeling to exam-

ine mathematics achievement over four time peri-

ods. Cross-classified growth models permit

analysis of achievement scores over more than

two time periods when the number of time periods

is limited, trajectories are nonlinear, and students

change schools (Goldstein 1999; Raudenbush and

Bryk 2002). We are able to predict both initial

scores in kindergarten and growth in scores

between kindergarten, first, third, and fifth grades.

This permits us to examine how professional com-

munity, teacher collaboration, and child-centered

planning affect achievement in school, controlling

for students’ initial scores:

ytðijÞ5 bo1X3

q 5 0

pqðijÞxqtðijÞ1Xp

p 5 1

lpwpi

1Xp

p 5 1

bpzpj1X3

q 5 0

pqðijÞxqtðijÞðXp

p 5 1

lpwpi

1Xp

p 5 1

bpzpjÞ1etðijÞ1u1i1m2j:

The outcome variable is math scores at time t for

student i in school j, yt(ij), where i and j are placed in

parentheses to reflect cross-classification. Math

scores are a function of time, xqt(ij), student varia-

bles, lpwpi, and school variables, bpzpj. Time

(coded 0, 1, 2, and 3 for K, first, third, and fifth

grades, respectively) is also interacted with student

and school variables. The direct effects of the stu-

dent and school variables, then, are the effects at

time 0, when students are in kindergarten. The

180 Sociology of Education 86(2)

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

interactive effects reflect the impact of student and

school variables at each time period. Time is entered

as multiple dichotomous variables rather than as

a scaled variable to permit nonlinear growth in

achievement over time. Growth in achievement by

race and SES is measured through interactions

between race/ethnicity, SES, and time. The effects

of professional communities on achievement for

each racial/ethnic and SES group is measured

through interactions between race/ethnicity, SES,

time, and professional community. The effects of

teacher collaboration on achievement for each

racial/ethnic and SES group are also measured

through interactions between race/ethnicity, SES,

time, and teacher collaboration. The equation in-

cludes a between-student error term, etðijÞ, and ran-

dom components for students and schools, u1i and

u2j (Littell et al. 1996; Raudenbush and Bryk 2002).7

RESULTS

Table 1 presents results of cross-classified growth

models with all controls discussed previously,

with the exception of professional community or

teacher collaboration. This table illustrates that in

comparison to white middle-SES students, lower-

SES students begin kindergarten with lower math

scores regardless of race or ethnicity. Black

middle-SES students begin school with signifi-

cantly lower scores than white middle-SES stu-

dents, and white high-SES students begin school

with higher scores. In first grade, black students,

regardless of SES, have lower growth in scores

than white middle-SES students. This disadvantage

accumulates over time for low- and middle-SES

black students as their growth significantly lags

behind white middle-SES students in third and fifth

grade. Latino/a low-SES students also have signif-

icantly lower growth than white middle-SES stu-

dents over time.

Table 2 presents the results for the effects of pro-

fessional community and teacher collaboration by

race/ethnicity and SES, controlling for all variables

described previously.8 Given the complicated nature

of interactions in the analyses, predicted least square

means (from Table 2) are also plotted in Figures 1

through 6. The first three figures present predicted

growth in mathematics achievement between kin-

dergarten and the fifth grade by race/ethnicity and

strength of the professional community for low-

SES (Figure 1), middle-SES (Figure 2), and high-

SES (Figure 3) students. The final three figures pres-

ent predicted growth in mathematics achievement

by the extent of teacher collaboration for low-

(Figure 4), middle- (Figure 5), and high-SES

(Figure 6) students. In each figure, gray lines reflect

Table 1. Slopes and Standard Errors from Cross-classified Growth Model Predicting MathematicsAchievement in Kindergarten, First, Third, and Fifth Grades

Initial score in kindergarten 14.48 (1.14)*** Growth in third grade 39.75 (1.15)***Black high SES 21.72 (1.65) Black high SES 22.54 (1.97)Black low SES 25.35 (0.90)*** Black low SES 215.01 (0.89)***Black middle SES 23.48 (1.06)** Black middle SES 210.82 (1.06)***Latino high SES 21.06 (1.47) Latino high SES 2.6 (1.54)^Latino low SES 25.8 (0.99)*** Latino low SES 27.73 (1.05)***Latino middle SES 21.98 (1.05)^ Latino middle SES 24.11 (1.08)***White high SES 3.98 (0.62)*** White high SES 2.99 (0.60)***White low SES 21.31 (0.61)* White low SES 26.19 (0.64)***Growth in first grade 16.8 (1.07)*** Growth in fifth grade 62.98 (1.35)***Black high SES 26.17 (1.58)*** Black high SES 23.89 (2.14)^Black low SES 25.17 (0.74)*** Black low SES 217.77 (1.02)***Black middle SES 24.45 (0.91)*** Black middle SES 213.79 (1.28)***Latino high SES 2.41 (1.33)^ Latino high SES 1.41 (1.58)Latino low SES 21.51 (0.87)^ Latino low SES 24.6 (1.14)***Latino middle SES 20.83 (0.93) Latino middle SES 21.07 (1.14)White high SES 1.4 (0.56)* White high SES 2.55 (0.64)***White low SES 22.45 (0.57)*** White low SES 25.25 (0.67)***

Note: Standard errors in parentheses. SES = socioeconomic status.^p \ .10. *p \ .05. **p \ .01. ***p \ .001.

Moller et al. 181

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

the predicted growth in achievement at the tenth per-

centiles of the Professional Community (Figures 1-

3) or Teacher Collaboration (Figures 4-5) distribu-

tions (from Table 2). The darker lines represent pre-

dicted growth in achievement at the ninetieth

percentiles of the Professional Community or

Teacher Collaboration distribution (again, based

on slopes from Table 2). Growth in achievement is

calculated by subtracting predicted scores (i.e., least

square means) in kindergarten from predicted scores

in the later years.

The results in Table 2 suggest that black low-

SES students benefit in the first, third, and fifth

grades from studying in schools where teachers per-

ceive the existence of a stronger professional com-

munity. These results are plotted in Figure 1. It is

clear from this graph that the mathematics trajecto-

ries of black low-SES students who study in schools

where teachers report weak professional communi-

ties (the gray dotted line) begin to diverge from their

peers in the third grade. These students have sub-

stantially (and significantly) lower growth in pre-

dicted math scores by the fifth grade than any

other group (at 67.29), including black low-SES stu-

dents who study with teachers who perceive there to

be strong professional communities in their schools.

Yet while studying in schools with strong profes-

sional communities clearly boosts the achievement

of low-SES black students, their growth by the fifth

grade trails that of their Latino/a and white peers.

Despite this fact, the graph clearly shows the impor-

tance of considering the strength of professional

communities when discussing racial gaps in

achievement among low-SES students.

Returning to Table 1, black low-SES students

trail their low-SES peers by approximately 13 points

in achievement growth by the fifth grade when col-

laboration and community are not controlled in the

model.9 However, turning to the darker line in

Figure 1 (from Table 2), students who spend their

elementary years in schools with teachers who con-

sistently sense the presence of a strong professional

community only lag behind their peers in growth by

9 points. Indeed, the gap jumps to 18 points for black

students who spend their elementary years with

teachers who sense the existence of a weak commu-

nity. Once again, this result highlights the impor-

tance of professional community for reducing

racial and socioeconomic gaps in achievement, as

low-SES black students are the most disadvantaged

demographic group in American schools.

Table 2 also illustrates that professional com-

munity is not a significant determinant of math

scores in kindergarten, although middle-SES

Latino/a students who begin their education in

schools where teachers perceive a stronger commu-

nity have lower mathematics scores. This is likely

not a result of the community, but merely an indica-

tion that initial disparities exist given that exposure

to professional communities is limited at this early

time, and there is no evidence of initial effects for

other race-SES groups. In the first and third grades,

the effect of Professional Community is negative

(in light of the interaction terms, this is the effect

for white middle-SES students), but Figure 2 illus-

trates that this effect is not substantial. White

middle-SES students who study in schools where

teachers perceive the existence of a strong profes-

sional community (indicated by the dark solid

line) are slightly disadvantaged compared to their

counterparts who attend schools where teachers

perceive there to be a weak professional commu-

nity (indicated by the gray solid line). However,

the point estimates on the graph are not signifi-

cantly different from each other, and this slight dis-

advantage subsides by the fifth grade. Indeed, by

the fifth grade, racial disparities in achievement

growth among middle-SES students are compara-

ble regardless of whether students spend their ele-

mentary years with teachers who report strong or

weak professional communities: Black middle-

SES students are disadvantaged regardless.

In contrast, the results in Table 2 suggest that

teachers’ sense of professional community does not

influence high-SES students’ math trajectories. Yet

when we plot growth in scores (see Figure 3), a strik-

ing story emerges: High-SES black students who

study in schools with teachers who perceive the exis-

tence of a weak community face serious obstacles as

their predicted math achievement growth between the

third and fifth grades is not significantly different

from zero. As a result, by the fifth grade, the growth

in math achievement of high-SES black students who

spend their elementary years with teachers who sense

weak communities falls significantly below white

students who study with teachers who sense strong

professional communities. Another potentially trou-

blesome picture materializes from the results pre-

sented in Figure 3: Black high-SES students who

study in schools with stronger professional commu-

nity appear to trail their peers in the first and third

grades. But these point estimates are not significant

(neither are the slopes in Table 2).

To summarize, black low-SES students experi-

ence the greatest benefit from studying with teach-

ers who sense the existence of strong professional

182 Sociology of Education 86(2)

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

Table 2. Slopes and Standard Errors from Cross-classified Growth Model Predicting MathematicsAchievement in Kindergarten, First, Third, and Fifth Grades by Race, Ethnicity, Socioeconomic Status (SES),Collaborative, Child-centered Planning and Professional Learning Community

Initial score in kindergarten 11.03 (1.74)*** Growth in third grade 40.07 (1.76)***Black high SES 0.38 (2.48) Black high SES 22.91 (2.42)Black low SES 24.48 (1.31)*** Black low SES 215.21 (0.93)***Black middle SES 22.49 (1.38)^ Black middle SES 211.68 (1.20)***Latino high SES 20.70 (1.83) Latino high SES 1.80 (1.75)Latino low SES 26.35 (1.28)*** Latino low SES 27.32 (1.16)***Latino middle SES 21.57 (1.37) Latino middle SES 23.42 (1.22)***White high SES 3.13 (0.73)*** White high SES 2.75 (0.64)***White low SES 20.86 (0.73) White low SES 26.21 (0.66)***Professional community 0.42 (0.56) Professional community 21.65 (0.81)*Professional Community 3

Black High SES2.09 (3.62) Professional Community 3

Black High SES23.79 (3.81)

Professional Community 3

Black Low SES20.90 (1.02) Professional Community 3

Black Low SES6.31 (1.49)***

Professional Community 3

Black Middle SES23.42 (1.82)^ Professional Community 3

Black Middle SES3.86 (2.29)

Professional Community 3

Latino High SES21.84 (2.16) Professional Community 3

Latino High SES4.76 (3.47)

Professional Community 3

Latino Low SES20.40 (0.95) Professional Community 3

Latino Low SES20.43 (1.50)

Professional Community 3

Latino Middle SES23.42 (1.55)* Professional Community 3

Latino Middle SES20.24 (2.42)

Professional Community 3

White High SES20.16 (0.74) Professional Community 3

White High SES4.70 (1.16)***

Professional Community 3

White Low SES0.96 (0.75) Professional Community 3

White Low SES3.94 (1.26)**

Collaborative planning 20.13 (0.48) Collaborative planning 3.48 (0.75)***Collaborative Planning 3

Black High SES5.72 (2.63)* Collaborative Planning 3

Black High SES22.95 (4.65)

Collaborative Planning 3

Black Low SES21.24 (1.02) Collaborative Planning 3

Black Low SES21.40 (1.59)

Collaborative Planning 3

Black Middle SES20.35 (1.36) Collaborative Planning 3

Black Middle SES0.65 (1.93)

Collaborative Planning 3

Latino High SES21.37 (1.79) Collaborative Planning 3

Latino High SES20.90 (3.79)

Collaborative Planning 3

Latino Low SES0.71 (0.90) Collaborative Planning 3

Latino Low SES20.42 (1.38)

Collaborative Planning 3

Latino Middle SES0.33 (1.31) Collaborative Planning 3

Latino Middle SES24.81 (2.18)*

Collaborative Planning 3

White High SES0.99 (0.71) Collaborative Planning 3

White High SES23.44 (1.15)**

Collaborative Planning 3

White Low SES20.17 (0.70) Collaborative Planning 3

White Low SES22.69 (1.18)*

Growth in first grade 16.39 (1.52)*** Growth in fifth grade 62.24 (1.76)***Black high SES 26.13 (2.02)** Black high SES 26.34 (2.33)**Black low SES 25.04 (0.88)*** Black low SES 217.64 (1.05)***Black middle SES 25.61 (1.12)*** Black middle SES 213.92 (1.47)***Latino high SES 1.96 (1.60) Latino high SES 0.22 (1.79)Latino low SES 21.33 (1.04) Latino low SES 24.28 (1.19)***Latino middle SES 20.11 (1.12) Latino middle SES 20.32 (1.25)

(continued)

Moller et al. 183

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

communities, not because this component of col-

lective pedagogical teacher culture dissolves their

disadvantage in mathematics achievement but

because it protects them from the deleterious ef-

fects of studying in schools where teachers do not

sense this community. This finding is in line with

prior research. For example, teacher interviews

from three elementary schools that were successful

in raising achievement for low-SES black students

suggested that teachers in those schools developed

collaborative, professional communities that

helped to enhance student achievement (Strahan

2003).

Turning to Teacher Collaboration, our findings

indicate that the results are not robustly significant

over time for any of the lower SES categories,

but they are significant for some middle- and

higher-SES students. Black lower-SES students

face serious challenges by the fifth grade regardless

of whether they spend their elementary years with

collaborative teachers (see Table 2 and Figure 4).

For middle-SES white students, the effects of

Table 2. (continued)

White high SES 1.64 (0.63)** White high SES 2.87 (0.68)***White low SES 22.04 (0.63)** White low SES 25.30 (0.70)***Professional community 21.75 (0.65)** Professional community 0.23 (1.00)Professional Community 3

Black High SES23.48 (2.61) Professional Community 3

Black High SES1.79 (5.03)

Professional Community 3

Black Low SES3.53 (1.32)** Professional Community 3

Black Low SES4.67 (1.76)**

Professional Community 3

Black Middle SES3.02 (1.86) Professional Community 3

Black Middle SES20.27 (2.98)

Professional Community 3

Latino High SES3.80 (3.07) Professional Community 3

Latino High SES4.72 (3.62)

Professional Community 3

Latino Low SES2.12 (1.26)^ Professional Community 3

Latino Low SES21.36 (1.85)

Professional Community 3

Latino Middle SES0.40 (1.92) Professional Community 3

Latino Middle SES24.00 (2.99)

Professional Community 3

White High SES1.40 (0.92) Professional Community 3

White High SES1.20 (1.42)

Professional Community 3

White Low SES1.58 (0.97) Professional Community 3

White Low SES1.52 (1.46)

Collaborative planning 0.84 (0.66) Collaborative planning 2.59 (0.88)**Collaborative Planning 3

Black High SES21.00 (3.10) Collaborative Planning 3

Black High SES5.80 (5.53)

Collaborative Planning 3

Black Low SES20.15 (1.44) Collaborative Planning 3

Black Low SES20.38 (2.30)

Collaborative Planning 3

Black Middle SES20.04 (1.62) Collaborative Planning 3

Black Middle SES1.73 (2.99)

Collaborative Planning 3

Latino High SES0.24 (2.42) Collaborative Planning 3

Latino High SES3.90 (3.68)

Collaborative Planning 3

Latino Low SES0.10 (1.13) Collaborative Planning 3

Latino Low SES21.05 (1.49)

Collaborative Planning 3

Latino Middle SES24.31 (1.76)* Collaborative Planning 3

Latino Middle SES1.84 (2.47)

Collaborative Planning 3

White High SES21.38 (0.99) Collaborative Planning 3

White High SES23.49 (1.31)**

Collaborative Planning 3

White Low SES20.92 (0.94) Collaborative Planning 3

White Low SES22.02 (1.48)

Note: Standard errors in parentheses. Controls for gender, student’s orientation toward learning in kindergarten, Englishas a second language in kindergarten, teacher of same race, teacher’s education, teacher satisfaction, percentage white inschool, region, urban, and school size.^p \ .10. *p \ .05. **p \ .01. ***p \ .001.

184 Sociology of Education 86(2)

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

teacher collaboration in the third and fifth grade are

significantly positive, indicating measurable

growth for middle-SES white students. These re-

sults are illustrated in Figure 5. White middle-

SES students who study in schools where their

teachers engage in collaborative planning (the

darker solid line) significantly outperform their

counterparts who study in schools where teachers

report little time on collaborative planning (the

gray solid line). Although white middle-SES stu-

dents experience 4 points greater growth in

achievement by the fifth grade when they study in

classrooms where teachers report collaboration,

the racial gap in achievement growth among mid-

dle-SES students is not significantly higher

because all students receive a boost from attending

schools where teachers collaborate. The gap

between white and black middle-SES students is

12.6 points among those students who spent their

elementary years in highly collaborative schools,

while it is 15.4 points among those white and black

middle-SES students who studied in noncollabora-

tive schools. Teacher collaboration helps white

middle-SES students without increasing racial

gaps in achievement.

Interestingly, Latino/a middle-SES students

who study with teachers who collaborate exten-

sively are disadvantaged in the first and third

grades, but this disadvantage dissipates by the fifth

grade as their scores approach white middle-SES

students. Finally, Figure 4 illustrates (in congru-

ence with Figure 2) that black middle-SES students

dramatically underperform other racial/ethnic

groups by the fifth grade regardless of their experi-

ence with teachers who collaborate.

Before turning to the results plotted in Figure 6,

we return to the results presented in Table 1, where

community and collaboration are not controlled.

This table illustrates that racial and ethnic gaps in

achievement among high-SES students are relatively

small, and black students only trail white and Latino/

a student by 5 points in growth by the fifth grade. Yet,

Figure 6 (which plots the results from Table 2) illus-

trates that examining racial/ethnic effects without

consideration of teacher collaboration masks gaps

in achievement across schools. Racial disparities in

Figure 1. Predicted math achievement growth forlow socioeconomic status (SES) students by teach-ers’ perception of professional community and byrace/ethnicity

Figure 2. Predicted math achievement growth formiddle socioeconomic status (SES) students byteachers’ perception of professional communityand by race/ethnicity

Moller et al. 185

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

math achievement trajectories are not evident among

high-SES students who spend their elementary years

with teachers who collaborate (as differences in

growth are not significant). Yet, racial disparities

are present among high-SES students who study in

elementary schools where teachers do not collabo-

rate, as white students outperform black and Latino/

a students and black students trail their peers by

approximately 15 points (compared to only 5 points

when we do not control for collaboration and commu-

nity, from Table 1).

Once again, summarizing the effects for teacher

collaboration, Latino/a and black high-SES students

and white middle-SES students experience the

greatest benefit from teacher collaboration.

Notably, the results illustrate that teacher collabora-

tion can reduce racial gaps in achievement while

also helping to boost the achievement of some white

students. This result is particularly interesting given

that prior studies have assumed that collaboration

accompanies community. However, the results

from this study indicate that the lowest SES students

do not necessarily benefit from collaboration.

DISCUSSION ANDCONCLUSIONS

There is a scarcity of nationally representative

research that examines how the organizational cul-

tures of elementary schools augment mathematics

achievement and reduce gaps in achievement by

race and SES (Vescio et al. 2008). Most studies that

have linked components of collective pedagogical

teacher culture—professional community and

teacher collaboration—with student achievement

were based on interviews and observational research

in a small number of schools (Berry, Johnson, and

Montgomery 2005; Hollins et al. 2004; Louis and

Marks 1998; Phillips 2003; Strahan 2003; Supovitz

2002). All of these studies found that overall achieve-

ment was higher in schools with strong professional

communities. Our research builds on these studies

by illustrating, with nationally representative data,

how achievement gains vary across students of differ-

ent racial/ethnic and socioeconomic status.

A limited number of quantitative studies also

addressed this topic, but those studies focused on

Figure 3. Predicted math achievement growth forhigh socioeconomic status (SES) students by teach-ers’ perception of professional community and byrace/ethnicity

Figure 4. Predicted math achievement growth forlow socioeconomic status (SES) students by teach-ers’ perception of collaboration and by race/ethnicity

186 Sociology of Education 86(2)

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

middle and high schools, and they did not assess

how components of collective pedagogical teacher

culture reduced achievement gaps by SES and race

(Bryk and Driscoll 1988; Lee and Smith 1993,

1995, 1996). Our research clarifies how schools’

organizational cultures affect achievement gaps

by race, ethnicity, and SES in elementary school.

In doing so, our study builds most directly on the

work of Lee and Smith (1996), who found that

socioeconomic gaps in achievement were smaller

in high schools where teachers took collective

responsibility for student learning.

We conceptualize an organizational culture

that should—and has been found to—enhance

students’ achievement trajectories. This organi-

zational culture, collective pedagogical teacher

culture, has two main components: (1) a strong

community orientation and (2) teacher collabora-

tion. Our findings indicate that community orien-

tation and teacher collaboration do not fit neatly

into a single component, and they must be mea-

sured separately. Yet, both of these elements of

collective pedagogical teacher culture are

important for reducing racial and socioeconomic

gaps in achievement.

Importantly, the majority of students are not

studying in schools with both of these factors. For

fifth-grade students in our sample, less than one-

fifth spend all of their elementary years in schools

where their teachers frequently collaborate with

each other, and even fewer students were fortunate

enough to spend all of their surveyed elementary

years with teachers who sensed that their schools

had all of the components of a professional commu-

nity (a mission, a sense of belonging and respect,

continual learning, and spirit for the school).

When thinking about this in terms of race, approx-

imately 30 percent of black and Latino/a students

and 35 percent of white students studied in schools

with relatively strong professional communities

(i.e., in the top 20 percent of the distribution) in kin-

dergarten. By fifth grade, 13 percent of black stu-

dents and 17 percent of Latino/a and white

students spent all of their elementary years in these

schools with relatively strong professional

communities.

Figure 5. Predicted math achievement growth formiddle socioeconomic status (SES) students byteachers’ perception of collaboration and by race/ethnicity

Figure 6. Predicted math achievement growth forhigh socioeconomic status (SES) students by teach-ers’ perception of collaboration and by race/ethnicity

Moller et al. 187

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

In light of our findings, the infrequencies of these

important experiences are problematic. We find that

math performance can be enhanced and racial and

socioeconomic gaps in achievement can be reduced

simply by changing the organizational culture of

schools. Specifically, a professional community

helps to reduce the extensive disadvantage faced

by black students. Indeed, lower- and higher-SES

black students are less disadvantaged by the fifth

grade when they have studied with teachers who

sense a professional community. Teacher collabora-

tion is also beneficial for higher-SES black students,

net of the community effect, because it reduces

achievement gaps. Interestingly, middle-SES black

students’ achievement trajectories are not signifi-

cantly altered by the measures of collective peda-

gogical teacher culture.

Consistent with findings of Reardon and

Galindo (2009), results from this study show that

within socioeconomic categories Latino/a students

perform at levels that are very similar to those of

white students. The predicted math trajectories

for white and Latino/a low-, middle-, and high-

SES students, irrespective of teachers’ perceptions

of professional community, are almost identical,

while the predicted math trajectories for black stu-

dents appear to be considerably lower across all

socioeconomic categories. This is an important

area for further investigation.

However, Latino/a students are differentially

affected by teacher collaboration. High-SES

Latino/a students benefit from spending their ele-

mentary years in schools with teachers who sense

a norm of collaboration while low-SES Latino/

a students are temporarily harmed. While their

deceleration in achievement rebounds by the fifth

grade, it is important to better understand the

source of this deceleration in an effort to help all

Latino/a students achieve their fullest potential.

Perhaps in the late 1990s many teachers had not

been trained to understand the unique cultural her-

itage of Latino/a students or their predispositions

(Valenzuela 1999). According to Espinosa

(1995), ‘‘Educators need to develop a greater

understanding of the features of the Latino/a culture

that influence parents’ childrearing and socializa-

tion practices, communication styles, and orienta-

tion toward formal education.’’ As such, teachers

are not always adequately aware of the pedagogical

approach best suited to Latino/a students (De

Gaetano 2007; Valenzuela 1999; Zea, Quezada,

and Belgrave 1994). This is an area for future

investigation.

A key question that researchers must better

understand is why a collective pedagogical teacher

culture boosts student achievement and reduces

gaps in achievement. As Gamoran and others

have argued, the primary mechanisms for this to

occur must be teaching practices. In a separate

analysis, not shown, we find that teachers who

report collaborative child-centered planning and

teachers who sense the presence of professional

communities in their schools are more likely to

assist individual children. Indeed, in kindergarten,

first, and third grades, around two-thirds of teach-

ers who fall in the top 25 percent of the teacher col-

laboration or professional community measures

offer students extra assistance at least three times

a week, compared to approximately 50 percent of

teachers who report very little collaboration or pro-

fessional community. Therefore, components of

collective pedagogical teacher culture influence

students’ mathematics achievement through teach-

ing practices due to the fact that teachers who per-

ceive stronger collective pedagogical teacher

cultures spend more time assisting individual chil-

dren. In separate research, we are further exploring

the extent to which collective pedagogical teacher

culture influences peer culture, curriculum, and

students’ orientation toward learning.

The policy implications of our findings are

important. Schools can improve math performance

and reduce achievement gaps by altering the orga-

nizational culture of schools. While teachers and

leaders can work toward developing child-centered

professional learning communities, there are

numerous challenges that they will need to over-

come. First, district and school leadership must be

supportive of this approach and put structures in

place that foster collaborative planning, such as

materials, time, and space (Gilrane, Roberts, and

Russell 2008; Levine and Marcus 2007; MacIver

and Epstein 1991; Wood 2007). Second, the devel-

opment of professional communities does not nec-

essarily create a collaborative environment among

teachers, either because teachers have not been

adequately professionalized as collaborative and

cooperative team players or because they do not

appreciate forced ‘‘community’’ (Pomson 2005).

Thus, it is important to include teachers in the pro-

cess of defining goals and professional development

needs (Gilrane et al. 2008). Third, the results for

Latino/a students suggest that collaboration will

only be effective for the greatest number of students

if planning is preceded by diversity training and cul-

tural understanding. This would serve to enhance

188 Sociology of Education 86(2)

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

trust between families and schools (De Gaetano

2007; Nicolau and Ramos 1990; Ream 2003).

Our findings suggest that politicians and bu-

reaucrats must consider the implications of educa-

tional reform on the organizational culture of

schools, asking whether reforms enhance schools’

cultures, thereby making them more accessible to

a diverse group of students. Educational reform

should be based on the premise that educators

have multiple types of resources to use when teach-

ing their students; schools’ organizational cultures

is one such resource (Grubb 2008). Indeed, Charles

Payne (2008) has argued that educational reforms

have failed to produce consistently positive results,

in part because reformers have overlooked the

extent of ‘‘collective depression’’ in schools. This

includes demoralized school cultures. Indeed, the

No Child Left Behind Act makes it difficult for

schools to develop collective pedagogical teacher

cultures because it increases competition among

teachers while simultaneously reducing trust and

morale (Crocco and Costigan 2006; Lankford et

al. 2002; Ravitch 2010). This is incompatible

with the development of collective pedagogical

teacher cultures. Reformers must address this issue

to more effectively reduce achievement gaps.

The final implication of this study concerns the

value of organizational sociology for research in

the sociology of education. As noted earlier, studies

of educational organizations rarely focus on

schools’ organizational cultures. We illustrate

that the organizational cultures of schools have

important implications for student outcomes. In

the spirit of the organizational research tradition,

future quantitative research on values and norms

should be accompanied by qualitative research

that highlights detailed processes and behaviors

and other organizational artifacts. This would

more fully elucidate values and norms and how

these norms are developed, maintained, and dif-

fused within elementary schools. Indeed, while

our research shows unequivocally the importance

of collective pedagogical teacher culture to reduc-

ing racial, ethnic, and socioeconomic mathematics

achievement gaps, one must turn to the qualitative

literature to understand the processes underlying

these trends. For example, Berry et al. (2005) clar-

ified in a rural elementary school how teachers col-

laborated to promote student achievement; this

included taking systematic notes on their teaching

to share with colleagues. Phillips (2003) also illus-

trated the process of collaboration through inter-

view research in a middle school, finding that

teachers collaborated through study circles, friends

groups, and classroom observations. A multime-

thod approach would help us further understand

the processes through which schools with collec-

tive pedagogical teacher cultures reduce racial, eth-

nic, and socioeconomic achievement gaps. Ideally,

future quantitative research will also include more

nuanced measures of teaching practices and stu-

dent-oriented teaching techniques and values to

determine exactly how collective pedagogical

teacher culture generates greater achievement. In

the meantime, this study provides important sup-

port for the theoretical claims that the organiza-

tional cultures of schools are essential elements

of successful reform efforts that augment mathe-

matics achievement for students across diverse

racial/ethnic and social class backgrounds.

FUNDING

The authors disclosed receipt of the following financial

support for the research, authorship, and/or publication

of this article: The research reported here was supported

by the Institute of Education Sciences, U.S. Department

of Education, through Grant R305A100822 to the

University of North Carolina at Charlotte. The opinions

expressed are those of the authors and do not represent

views of the Institute or the U.S. Department of

Education.

NOTES

1. A visionary leader is important for professional com-

munities because he or she permits a centralized set of

goals. This leader, typically the principal, can change

or bolster the organizational culture of his or her

school by working with teachers to instill values, trust,

and expectations (Bizar and Barr 2001; Bryk and

Schneider 2002; McLaughlin and Talbert 2006). A

lack of leadership, in contrast, inhibits the develop-

ment of shared norms.

2. It is important to note that our description of collective

pedagogical teacher culture assumes that culture has

developed dynamically (following the important theo-

retical work of Gamoran, Secada, and Marrett 2000).

In schools where teachers are continually learning,

a sense of collaboration may develop, fueling a desire

for continued learning.

3. Scaled variables are imputed with the Markov Chain

Monte Carlo method because we have an arbitrary

missing data pattern (Schafer 1997). Categorical var-

iables are imputed with a logistic regression method.

4. Oblique rotations are generally favorable to orthogonal ro-

tations (Conway and Huffcutt 2003; Fabrigar et al. 1999).

5. We also tested the two-factor solution through confir-

matory factor analysis (CFA) for each grade. We fit

Moller et al. 189

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

models with goodness of fit and Comparative Fit

Index indices above .98 and root mean square error

of approximation below .06 in each time period. The

exploratory factor analysis (EFA) and CFA factor

scores are correlated above .97 in each time period

for the first factor, and they are correlated above .9

for the second factor. This suggests that the factor

scores created from the EFA represent constructs

that fit the data well.

6. Data on socioeconomic status (SES) are asked of parents

in each wave, permitting us to model change in SES over

time. Less than 10 percent of the sample has a substantial,

lasting change in socioeconomic status over time. Thus,

we measure SES in kindergarten to assess how SES at

school entry affects achievement growth.

7. We do not include an interaction between u1i and u2j

because the models are memory intensive and will

not run.

8. In separate analysis (not shown) we incorporate a con-

trol for percentage of students in the school on reduced

price lunch. Our results are robust to that control.

9. The slopes for black low-SES, Latino/a low-SES, and

white low-SES are –17.77, –4.6, and –5.25,

respectively.

REFERENCES

Allison, Paul D. 2002. Missing Data. Thousand Oaks,

CA: Sage.

Anderson, Carolyn S. 1982. ‘‘The Search for School

Climate: A Review of the Research.’’ Review of

Educational Research 52:368-420.

Berry, Barnett, Dylan Johnson, and Diana Montgomery.

2005. ‘‘The Power of Teacher Leadership.’’

Educational Leadership 62:56.

Bizar, Marilyn and Rebecca Barr. 2001. School

Leadership in Times of Urban Reform. Mahwah, NJ:

Lawrence Erlbaum Associates.

Black, Richard J. 2003. Organisational Culture: Creating

the Influence Needed for Strategic Success. London:

dissertation.com.

Bodovski, Katerina and George Farkas. 2007.

‘‘Mathematics Growth in Early Elementary School:

The Roles of Beginning Knowledge, Student

Engagement, and Instruction.’’ Elementary School

Journal 108:115-30.

Bryk, Anthony S. and Mary Erina Driscoll. 1988. The

High School as Community: Contextual Influences

and Consequences for Students and Teachers.

Madison, WI: National Center on Effective

Secondary Schools: Wisconsin Center for Education

Research University of Wisconsin-Madison.

Bryk, Anthony S., Valerie E. Lee, and Peter Blakeley

Holland. 1993. Catholic Schools and the Common

Good. Cambridge, MA: Harvard University Press.

Bryk, Anthony S. and Barbara Schneider. 2002. Trust in

Schools: A Core Resource for Improvement. New

York: Russell Sage Foundation.

Burrill, Gail. 2001. ‘‘Mathematics Education: The Future

and the Past Create a Context for Today’s Issues.’’ Pp.

25-41 in The Great Curriculum Debate, edited by T.

Loveless. Washington, DC: Brookings Institution Press.

Cheadle, Jacob E. 2008. ‘‘Educational Investment, Family

Context, and Children’s Math and Reading Growth

from Kindergarten through the Third Grade.’’

Sociology of Education 81:1-31.

Comer, James P. and Christine Emmons. 2006. ‘‘The

Research Program of the Yale Child Study Center

School Development Program.’’ The Journal of

Negro Education 75:353-72.

Conway, J. M. and A. I. Huffcutt. 2003. ‘‘A Review and

Evaluation of Exploratory Factor Analysis Practices

in Organizational Research.’’ Organizational

Research Methods 6:147-68.

Crocco, Margaret Smith and Arthur T. Costigan. 2006.

‘‘High-stakes Teaching: What’s at Stake for

Teachers (and Students) in the Age of

Accountability.’’ The New Educator 2:1-13.

Cuban, Larry. 1990. ‘‘Reforming Again, Again, and

Again.’’ Educational Researcher 19:3-13.

Darling-Hammond, Linda. 2010. The Flat World and

Education: How America’s Commitment to Equity

Will Determine Our Future. New York: Teachers

College Press.

De Gaetano, Yvonne. 2007. ‘‘The Role of Culture in

Engaging Latino Parents’ Involvement in School.’’

Urban Education 42:145-62.

Dee, T. S. 2004. ‘‘Teachers, Race, and Student

Achievement in a Randomized Experiment.’’

Review of Economics and Statistics 86:195-210.

Denison, D. 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 21:619-54.

Desimone, Laura. 2002. ‘‘How Can Comprehensive

School Reform Models Be Successfully

Implemented?’’ Review of Educational Research 72:

433-79.

Ehrenberg, R. G., D. D. Goldhaber, and D. J. Brewer.

1995. ‘‘Do Teachers Race, Gender, and Ethnicity

Matter—Evidence from the National Educational

Longitudinal Study of 1988.’’ Industrial & Labor

Relations Review 48:547-61.

Eilers, Angela M. and Armando Carnacho. 2007. ‘‘School

Culture Change in the Making: Leadership Factors

That Matter.’’ Urban Education 42:616-37.

Espinosa, Linda M. 1995. ‘‘Hispanic Parent Involvement

in Early Childhood Programs’’ (ED382412). Urbana,

IL: ERIC Clearinghouse on Elementary and Early

Childhood Education.

Fabrigar, L. R., D. T. Wegener, R. C. MacCallum, and E.

J. Strahan. 1999. ‘‘Evaluating the Use of Exploratory

Factor Analysis in Psychological Research.’’

Psychological Methods 4:272-99.

Firestone, W. A. and Karen Seashore Louis. 1999.

‘‘Schools as Cultures.’’ Pp. 297-319 in Handbook on

190 Sociology of Education 86(2)

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

Research on Educational Administration, edited by J.

Murphy and K. S. Louis. San Francisco: Jossey-Bass

Publishers.

Freeman, Lance. 2008. ‘‘Is Class Becoming a More

Important Determinant of Neighborhood Attainment

for African-Americans?’’ Urban Affairs Review 44:

3-26.

Gamoran, Adam, R. Gunter, and T. Williams. 2005.

‘‘Professional Community by Design: Building

Social Capital through Teacher Professional

Development.’’ Pp. 129-46 in Social Organization

of Schooling, edited by L. V. Hedges and B.

Schneider. New York: Russell Sage Foundation.

Gamoran, Adam, Walter G. Secada, and Cora B. Marrett.

2000. ‘‘The Organizational Context of Teaching and

Learning.’’ Pp. 37-63 in Handbook of the Sociology

of Education, edited by M. T. Hallinan. New York:

Kluwer Academic/Plenum Publishers.

Gersten, Russell and Scott Baker. 2000. ‘‘What We

Know about Effective Instructional Practices for

English-language Learners.’’ Exceptional Children

66:454-70.

Gilrane, Colleen P, Melba L Roberts, and Lisa Anne

Russell. 2008. ‘‘Building a Community in Which

Everyone Teaches, Learns, and Reads: A Case

Study.’’ The Journal of Educational Research 101:

333-49.

Goldstein, Harvey. 1999. ‘‘Multilevel Statistical Models.’’

Retrieved December 21, 2012 (http://www.soziologie

.uni-halle.de/langer/multilevel/books/goldstein.pdf).

Grubb, W. Norton. 2008. ‘‘Multiple Resources, Multiple

Outcomes: Testing the ‘Improved’ School Finance

with NELS88.’’ American Educational Research

Journal 45:104-44.

Hanushek, E. A. and S. G. Rivkin. 2010.

‘‘Generalizations about Using Value-added

Measures of Teacher Quality.’’ American Economic

Review 100:267-71.

Hess, Frederick M. 1999. Spinning Wheels: The Politics

of Urban School Reform. Washington, DC:

Brookings Institution Press.

Hollins, Etta, Linda McIntyre, Charles DeBose, Kimberly

Hollins, and Arthurlene Towner. 2004. ‘‘Promoting

a Self-sustaining Learning Community: Investigating

an Internal Model for Teacher Development.’’

International Journal of Qualitative Studies in

Education 17:247-64.

Kane, Thomas J. and Douglas O. Staiger. 2008.

‘‘Estimating Teacher Impacts on Student

Achievement: An Experimental Evaluation.’’ Working

Paper No. 14607, National Bureau of Economic

Research, Chicago, IL.

Konstantopoulos, Spyros. 2007. How Long Do Teacher

Effects Persist? Bonn: IZA.

Konstantopoulos, Spyros. 2008. ‘‘Do Small Classes

Reduce the Achievement Gap between Low and

High Achievers? Evidence from Project Star.’’

Elementary School Journal 108:275-91.

Krueger, A. B. and D. M. Whitmore. 2001. ‘‘The Effect of

Attending a Small Class in the Early Grades on

College Test Taking and Middle School Test

Results: Evidence from Project Star.’’ Economic

Journal 111:1-28.

Kruse, Sharon D. and Karen Seashore Louis. 2009.

Building Strong School Cultures: A Guide

to Leading Change. Thousand Oaks, CA: Corwin

Press.

Ladd, Gary W. and Lisa M. Dinella. 2009. ‘‘Continuity

and Change in Early School Engagement: Predictive

of Children’s Achievement Trajectories from First

to Eighth Grade?’’ Journal of Educational

Psychology 101:190-206.

Langer, Judith A. 2000. ‘‘Excellence in English in Middle

and High School: How Teachers’ Professional Lives

Support Student Achievement.’’ American

Educational Research Journal 37:397-439.

Lankford, Hamilton, Susanna Loeb, and James Wyckoff.

2002. ‘‘Teacher Sorting and the Plight of Urban

Schools: A Descriptive Analysis.’’ Educational

Evaluation and Policy Analysis 24:37-62.

Lee, J., W. Grigg, and G. Dion. 2007. The Nation’s Report

Card: Mathematics 2007—National Assessment of

Educational Progress at Grades 4 and 8. NCES

2007-494. Washington, DC: National Center for

Education Statistics.

Lee, Valerie E. and Julia B. Smith. 1993. ‘‘Effects of

School Restructuring on the Achievement and

Engagement of Middle-grade Students.’’ Sociology

of Education 66:164-87.

Lee, Valerie E. and Julia B. Smith. 1995. ‘‘Effects of

High-school Restructuring and Size on Early Gains

in Achievement and Engagement.’’ Sociology of

Education 68:241-70.

Lee, Valerie E. and Julia B. Smith. 1996. ‘‘Collective

Responsibility for Learning and Its Effects on Gains

in Achievement for Early Secondary School

Students.’’ American Journal of Education 104:

103-47.

Lee, Valerie E. and Julia B. Smith. 1997. ‘‘High School

Size: Which Works Best and for Whom?’’

Educational Evaluation and Policy Analysis 19:

205-27.

Lee, Valerie E., Julia B. Smith, and Robert G. Croninger.

1997. ‘‘How High School Organization Influences the

Equitable Distribution of Learning in Mathematics

and Science.’’ Sociology of Education 70:128-50.

Levine, Thomas H. and Alan S. Marcus. 2007. ‘‘Closing

the Achievement Gap through Teacher Collaboration.’’

Journal of Advanced Academics 19:116-38.

Littell, R. C., G. A. Milliden, W. W. Stroup, and R. D.

Wolfinger. 1996. SAS System for Mixed Models.

Cary, NC: SAS Institute.

Little, Judith W. 1982. ‘‘Norms of Collegiality and

Experimentation: Workplace Conditions of School

Success.’’ American Educational Research Journal

19:325-40.

Moller et al. 191

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

Louis, Karen Seashore and Sharon D. Kruse. 1995.

Professionalism and Community: Perspectives on

Reforming Urban Schools. Thousand Oaks, CA: Sage.

Louis, Karen Seashore and Helen M. Marks. 1998. ‘‘Does

Professional Community Affect the Classroom?

Teachers’ Work and Student Experiences in

Restructuring Schools.’’ American Journal of

Education 106:532-75.

Louis, Karen Seashore, Helen M. Marks, and Sharon

Kruse. 1996. ‘‘Teachers’ Professional Community in

Restructuring Schools.’’ American Educational

Research Journal 33:757-98.

MacIver, Douglas J. and Joyce L. Epstein. 1991.

‘‘Responsive Practices in the Middle Grades:

Teacher Teams, Advisory Groups, Remedial

Instruction, and School Transition Programs.’’

American Journal of Education 99:587-622.

MacNeil, Angus, Doris Prater, and Steve Busch. 2009.

‘‘The Effects of School Culture and Climate on

Student Achievement.’’ International Journal of

Leadership in Education 12:73-84.

McLaughlin, Milbrey Wallin and Joan E. Talbert. 2006.

Building School-based Teacher Learning Communities:

Professional Strategies to Improve Student

Achievement. New York: Teachers College Press.

Moller, Stephanie, Elizabeth Stearns, Judith R. Blau, and

Kenneth C. Land. 2006. ‘‘Smooth and Rough Roads

to Academic Achievement: Retention and Race/

Class Disparities in High School.’’ Social Science

Research 35:157-80.

Moller, Stephanie, Elizabeth Stearns, Stephanie

Potochnick, and Stephanie Southworth. 2011.

‘‘Student Achievement and College Selectivity:

How Changes in Achievement during High School

Affect the Selectivity of College Attended.’’ Youth

& Society 43:639-63.

Moses, Robert Parris and Charles E. Cobb. 2001. Radical

Equations: Math Literacy and Civil Rights. Boston:

Beacon Press.

Newmann, Fred M., Helen M. Marks, and Adam

Gamoran. 1996. ‘‘Authentic Pedagogy and Student

Performance.’’ American Journal of Education 104:

280-312.

Nichols, Sharon L. and David C. Berliner. 2007.

Collateral Damage: How High-stakes Testing

Corrupts America’s Schools. Cambridge, MA:

Harvard Education Press.

Nicolau, Siobhan and Carmen Lydia Ramos. 1990.

‘‘Together Is Better: Building Strong Relationships

between Schools and Hispanic Parents.’’ Hispanic

Policy Development Project, New York.

Nye, Barbara, Spyros Konstantopoulos, and Larry V.

Hedges. 2004. ‘‘How Large Are Teacher Effects?’’

Educational Evaluation and Policy Analysis 26:

237-57.

Oliver, Melvin L. and Thomas M. Shapiro. 2006. Black

Wealth, White Wealth: A New Perspective on Racial

Inequality. New York, NY: Routledge.

O’Reilly, C. A., J. Chatman, and D. F. Caldwell. 1991.

‘‘People and Organizational Culture—A Profile

Comparison Approach to Assessing Person-

organization Fit.’’ Academy of Management Journal

34:487-516.

Ostroff, C. 1992. ‘‘The Relationship between Satisfaction,

Attitudes, and Performance—An Organizational

Level Analysis.’’ Journal of Applied Psychology 77:

963-74.

Park, Insim. 2005. ‘‘Teacher Commitment and Its Effects

on Student Achievement in American High Schools.’’

Educational Research and Evaluation 11:461-85.

Patchen, Martin. 2004. Making Our Schools More

Effective: What Matters and What Works.

Springfield, IL: C.C. Thomas Publishers.

Payne, Charles M. 2008. So Much Reform, So Little

Change: The Persistence of Failure in Urban

Schools. Cambridge, MA: Harvard Education Press.

Pedersen, Jesper Strandgaard and Frank Dobbin. 2006.

‘‘In Search of Identity and Legitimation: Bridging

Organizational Culture and Neoinstitutionalism.’’

American Behavioral Scientist 49:897-907.

Phillips, Joy. 2003. ‘‘Powerful Learning: Creating

Learning Communities in Urban School Reform.’’

Journal of Curriculum and Supervision 18:240-58.

Pomson, Alex D. 2005. ‘‘One Classroom at a Time?

Teacher Isolation and Community Viewed through

the Prism of the Particular.’’ Teachers College

Record 107:783-802.

Powers, Jeanne M. 2009. Charter Schools: From Reform

Imagery to Reform Reality. New York: Palgrave

Macmillan.

Raudenbush, S. and A. S. Bryk. 2002. Hierarchical

Linear Models. Vol. 2. Thousand Oaks, CA: Sage.

Ravitch, Diane. 2010. The Death and Life of the Great

American School System: How Testing and Choice

Are Undermining Education. New York: Basic

Books.

Ream, Robert K. 2003. ‘‘Counterfeit Social Capital and

Mexican-American Underachievement.’’ Educational

Evaluation and Policy Analysis 25:237-62.

Reardon, Sean F. and Claudia Galindo. 2009. ‘‘The

Hispanic-white Achievement Gap in Math and

Reading in the Elementary Grades.’’ American

Educational Research Journal 46:853-91.

Rumberger, Russell W. and Gregory J. Palardy. 2005.

‘‘Does Segregation Still Matter? The Impact of

Student Composition on Academic Achievement in

High School.’’ Teachers College Record 107:1999-2045.

Schafer, Joseph L. 1997. Analysis of Incomplete

Multivariate Data. London: Chapman & Hall.

Schein, Edgar H. 2010. Organizational Culture and

Leadership. San Francisco: Jossey-Bass.

Singh, Kusum, Monique Granville, and Sandra Dika.

2002. ‘‘Mathematics and Science Achievement:

Effects of Motivation, Interest, and Academic

Engagement.’’ Journal of Educational Research 95:

323-32.

192 Sociology of Education 86(2)

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

Smey-Richman, Barbara. 1991. School Climate and

Restructuring for Low-achieving Students.

Philadelphia: Research for Better Schools.

Stoll, Louise, Ray Bolam, Agnes McMahon, Mike

Wallace, and Sally Thomas. 2006. ‘‘Professional

Learning Communities: A Review of the Literature.’’

Journal of Educational Change 7:1573-812.

Strahan, D. 2003. ‘‘Promoting a Collaborative

Professional Culture in Three Elementary Schools

That Have Beaten the Odds.’’ The Elementary

School Journal 104:127-46.

Supovitz, Jonathan A. 2002. ‘‘Developing Communities

of Instructional Practice.’’ Teachers College Record

104:1591-626.

Tourangeau, Karen, Christine Nord, Thanh Le, Alberto

Sorongon, Michelle Najaran, and Elvira Germino

Hausken. 2009. Early Childhood Longitudinal Study,

Kindergarten Class of 1998-99 (ECLS-K): Combined

Users Manual for the ECLS-K Eight-grade and K-8

Full Sample Data Files and Electonic Codebooks.

Washington, DC: U.S. Department of Education,

National Center for Education Statistics.

Tyack, David B. 1974. The One Best System: A History of

American Urban Education. Cambridge, MA:

Harvard University Press.

Tyack, David B. and Larry Cuban. 1995. Tinkering

toward Utopia: A Century of Public School Reform.

Cambridge, MA: Harvard University Press.

Valenzuela, Angela. 1999. Subtractive Schooling: U.S.-

Mexican Youth and the Politics of Caring. SUNY

Series, the Social Context of Education. Ithaca, NY:

State University of New York Press.

Valli, Linda and Daria Buese. 2007. ‘‘The Changing

Roles of Teachers in an Era of High-stakes

Accountability.’’ American Educational Research

Journal 44:519-58.

Vescio, Vicki, Dorene Ross, and Alyson Adams. 2008.

‘‘A Review of Research on the Impact of

Professional Learning Communities on Teaching

Practice and Student Learning.’’ Teaching and

Teacher Education 24:80-91.

Wilson, William J. 1978a. The Declining Significance of

Race: Blacks and Changing American Institutions.

Chicago: University of Chicago Press.

Wilson, William J. 1978b. ‘‘Declining Significance of

Race—Revisited but Not Revised.’’ Society 15:11.

Wilson, William J. 2009. More Than Just Race: Being

Black and Poor in the Inner City. New York:

Norton & Company.

Wilson, William J. 2011. ‘‘The Declining Significance of

Race: Revisited & Revised.’’ Daedalus 140:55-69.

Wood, Diane. 2007. ‘‘Teachers’ Learning Communities:

Catalyst for Change or a New Infrastructure for the

Status Quo?’’ Teachers College Record 109:699-739.

Wynne, Edward A. 1980. Looking at Schools: Good, Bad,

and Indifferent. Lexington, MA: Lexington Books.

Yasumoto, Jeffrey Y., Kazuaki Uekawa, and Charles E.

Bidwell. 2001. ‘‘The Collegial Focus and High

School Students’ Achievement.’’ Sociology of

Education 74:181-209.

Zea, Maria C., T. Quezada, and F. Belgrave. 1994.

‘‘Cultural Values and Adjustment to Disability among

Latinos.’’ Journal of Social Behavior and Personality

9:185-200.

Zvoch, Keith and Joseph J. Stevens. 2006. ‘‘Longitudinal

Effects of School Context and Practice on Middle

School Mathematics Achievement.’’ Journal of

Educational Research 99:347-56.

BIOS

Stephanie Moller is an associate professor of sociology

and public policy at the University of North Carolina at

Charlotte. She conducts research on mathematics

achievement in primary and secondary schools, examin-

ing racial, ethnic, and socioeconomic gaps in achieve-

ment. Recent publications have appeared in Sociology

of Education, Youth & Society, Social Science

Research, and Urban Education. She has received fund-

ing from the National Science Foundation, the Spencer

Foundation, AERA, and the U.S. Department of

Education.

Roslyn Arlin Mickelson is professor of sociology, public

policy, women and gender studies, and information tech-

nology at the University of North Carolina at Charlotte. A

former high school social studies teacher, her research in-

terests include pathways to STEM, minority educational

issues, desegregation, social science and the law, gender

and education, school reform, and educational policy.

Mickelson has investigated school reform in the

Charlotte-Mecklenburg Schools since 1989, chronicling

the consequences of the district’s transformation from

a desegregated to a resegregated school system.

Elizabeth Stearns is an associate professor of sociology

at the University of North Carolina at Charlotte. Her

research interests include racial and gender differences

in student outcomes, interracial friendships, and inequal-

ity in the contemporary United States. Recent publica-

tions have appeared in Teachers College Record, Social

Science Research, and Sociology of Education. She has

received funding from the National Science Foundation,

the Spencer Foundation, AERA, and the U.S.

Department of Education.

Neena Banerjee is a PhD candidate in public policy at the

University of North Carolina at Charlotte. Her research

focuses on racial and ethnic diversity among teachers,

teachers’ job satisfaction, and their implications on racial,

ethnic, and socioeconomic gaps in student achievement.

Martha Cecilia Bottia is assistant research professor of

sociology at the University of North Carolina at

Moller et al. 193

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from

Charlotte. Since 2005, Bottia has been surveying and syn-

thesizing the educational, social, and behavioral science

literature on the effects of school racial and socioeco-

nomic demographic composition on various educational

outcomes. Her other research interests include illicit

drugs and terrorist organizations and the education of

immigrant students. Currently, Bottia is working on

a series of articles related to the unequal impact of the im-

plemented curriculum on a racial and socioeconomic

diverse students and on the role of structural characteris-

tics of K-12 schools on the decision of students to major

and graduate from a STEM major.

194 Sociology of Education 86(2)

at UNIV NORTH CAROLINA-CHARLOTTE on February 18, 2014soe.sagepub.comDownloaded from