Essay about racism in high school setting
JINKA
InequalityinBlackandWhiteHighSchoolStudentsPerceptions.pdf
E M P I R I C A L R E S E A R C H
Inequality in Black and White High School Students’ Perceptions of School Support: An Examination of Race in Context
Jessika H. Bottiani1 • Catherine P. Bradshaw1 • Tamar Mendelson2
Received: 3 November 2015 / Accepted: 19 December 2015 / Published online: 8 January 2016
� Springer Science+Business Media New York 2016
Abstract Supportive relationships with adults at school
are critical to student engagement in adolescence. Addi-
tional research is needed to understand how students’ racial
backgrounds interact with the school context to shape their
perceptions of school support. This study employed mul-
tilevel, latent variable methods with a sample of Black and
White students (N = 19,726, 35.8 % Black, 49.9 % male,
mean age = 15.9) in 58 high schools to explore variation
in perceived caring, equity, and high expectations by stu-
dent race, school diversity, and socioeconomic context.
The results indicated that Black students perceived less
caring and equity relative to White students overall, and
that equity and high expectations were lower in diverse
schools for both Black and White students. Nonetheless,
racial disparities were attenuated in more diverse schools.
The findings point to the need for intervention to improve
perceptions of school support for Black youth and for all
students in lower income and more diverse schools.
Keywords School climate � Engagement � Racial disparities � School diversity � Social support
Introduction
Racial disparities between Black students and their White
peers in academic and disciplinary outcomes are among the
most pressing concerns facing U.S. schools (Aud et al.
2012; Gregory et al. 2010) and are drawing increasing
policy and research attention (Advancement Project 2010;
Fabelo et al. 2011). Supportive relationships at school are
linked consistently to more positive academic and behav-
ioral outcomes (Eccles and Roeser 2011) and thus present a
promising avenue of research to better understand racial
gaps in achievement and school discipline exposure.
Although some studies suggest that Black students perceive
less support than White students (Hughes and Kwok 2007),
additional research is needed to examine disparate per-
ceptions of school support with attention to variation by
school context (e.g., the diversity of the school) and to
other aspects of students’ backgrounds (e.g., socioeco-
nomic status). This study explored how students’ race
interacted with the diversity and socioeconomic context of
the school to shape perceptions of school support in three
dimensions (caring, equity, and high expectations).
School Support
School support is an emerging construct theorized to fulfill
youth needs for belonging, competence, and autonomy
(Hanson and Kim 2007; Skinner and Pitzer 2012) and
conceptualized as an aspect of school climate shaped by
students’ relationships with adults at school. Research
indicates that supportive relationships with adults at school
predict students’ academic engagement and social-emo-
tional well-being (Roeser et al. 2000), particularly for
behaviorally at-risk Black youth (Decker et al. 2007).
Several conceptual models, such as the dynamic model of
& Jessika H. Bottiani [email protected]
Catherine P. Bradshaw
Tamar Mendelson
1 Curry School of Education, University of Virginia, Ruffner
Hall, 212, 407 Emmet Street South, Charlottesville,
VA 22903, USA
2 Johns Hopkins Bloomberg School of Public Health, 642 N.
Broadway, Hampton House 853, Baltimore, MD 21205, USA
123
J Youth Adolescence (2016) 45:1176–1191
DOI 10.1007/s10964-015-0411-0
motivational development of engagement (Skinner and
Pitzer 2012) and the youth development and resiliency
models (Benard 2004; Hanson and Kim 2007) provide a
theoretical framework of the dimensions of school support
that motivate student engagement and promote positive
student outcomes. Two dimensions of school support—
caring and high expectations—are consistently regarded as
essential to youth social-emotional and school outcomes
(Baker et al. 2008; Benard 2004; Furrer and Skinner 2003;
Gregory and Weinstein 2008; Hanson and Kim 2007;
Hughes et al. 2008). Caring refers to warmth and regard for
students as individuals, and is theorized to promote stu-
dents’ sense of belonging at school. The dimension of high
expectations refers to support for students’ ability to meet a
high standard of academic effort, and is theorized to pro-
mote academic competence. Theory suggests that sup-
portive school contexts encourage engagement motivation
by fulfilling students’ psychological needs (Skinner and
Pitzer 2012). Unfortunately, efforts to identify threats to
student engagement have relied too heavily on student risk
factors and tended to overlook the important role of school
context (Finn and Zimmer 2012) and person-context fit
(Byrd and Chavous 2011). Furthermore, although these
dimensions are well supported by research, theoretical
models tend to neglect the role of race and racial inequity
in schools, influences that have potential to explain dis-
parate school outcomes among Black and White youth.
More inclusive models addressing the intersections of race
and social context are necessary to further our under-
standing of the development of young people of color
(Garcı́a Coll et al. 1996).
Equity as a Dimension of School Support
Students’ perceptions of differential treatment, exclusion,
and discrimination by teachers and other adults in school
appear to play a role in poor outcomes among youth of
color in school. For example, perceived discrimination has
been linked with mental health problems, including anti-
social behavior (Bogart et al. 2013), depression, and low
self-esteem (Zeiders et al. 2012). Among Black students,
perceived discrimination is inversely associated with
valuing of education and academic persistence (e.g., Dot-
terer et al. 2009; Smalls et al. 2007; Wong et al. 2003).
Conversely, research on school climate suggests that per-
ceived school equity, which includes dimensions of fair-
ness and inclusiveness [Organization for Economic
Cooperation and Development (OECD) 2008], may posi-
tively influence students’ sense of connectedness and aca-
demic motivation in school (Debnam et al. 2014).
Adolescents’ perceptions of fairness in the school envi-
ronment can enhance both students’ sense of competence
(Elliot and Dweck 2005) and connectedness (Lowman
1984). Existing conceptualizations of school support such
as the dynamic model of motivational development
(Skinner and Pitzer 2012) and the youth development and
resiliency model (Hanson and Kim 2007) have not inclu-
ded dimensions reflecting students’ perceptions of equity.
Thus, there has been limited research exploring fairness
and inclusion as a component of school support.
Racial Disparities in School Support
Although school support is linked with positive outcomes
across diverse racial and ethnic groups (e.g., Garcı́a-Reid
et al. 2005; Tyler and Boelter 2008), educators have theo-
rized that Black students in particular may benefit from
teacher support to help navigate sociocultural boundaries
between school, home, and neighborhood and to cope with
experiences of discrimination at school (Gay 2002).
Research demonstrates the particular salience of supportive
teacher relationships relative to other social factors for Black
youth (Decker et al. 2007; Meehan et al. 2003), yet also
suggests that Black students experience less supportive
relationships with their teachers and less school connected-
ness relative to their White peers (Hamre and Pianta 2001;
Hughes and Kwok 2007; Furlong et al. 2011). Because race
and socioeconomic status (SES) can be conflated, some
research is beginning to assess whether racial gaps persist
even when controlling for students’ socioeconomic status
(SES; Bottiani et al. 2014; Voight et al. 2015); however, most
studies that have demonstrated racial gaps in perceived
support, climate, and connectedness have done so with
limited or no accounting for effects associated with SES.
The Role of School Context
Racial gaps in perceived support may play an important role
in the observed poorer academic, disciplinary, and mental
health outcomes among Black adolescents, yet we know
relatively little about their root causes. School racial and
ethnic diversity and SES are important aspects of the school
social context that may differentially shape adolescent per-
ceptions of support by race. Although research examining
the role of school diversity in inequalities in Black boys’
report of school support is scant overall, a handful of studies
suggest that youth of color perceive poorer racial climate and
greater discrimination in more racially/ethnically heteroge-
neous schools (Benner and Graham 2013; Seaton and Yip
2009), whereas youth of color fare better in schools with a
critical mass of same-ethnicity peers (Benner and Graham
2009, 2013). These studies examined climate perceptions
among students of color, but did not explicitly attend to racial
disparities in students’ perceptions of school support. In
addition, the studies were conducted in urban schools on the
west coast, whereas relatively little research has explored
J Youth Adolescence (2016) 45:1176–1191 1177
123
how school diversity influences student perceptions of
school in Mid-Atlantic/Southern region of the U.S., or in
suburban and rural areas, which may have larger Black
populations and lower levels of diversity overall. Impor-
tantly, to our knowledge, no studies have examined how
school SES and school diversity interact to exacerbate or
mitigate racial disparities in students’ perceptions of school
support. Black and Latino youth are more likely to attend
high-poverty schools, and high-poverty schools in turn are
associated with lower levels of teacher educational attain-
ment and higher rates of social disorder (National Center for
Education Statistics 2010). In schools marked by higher
percentage of students on free and reduced price meals and
higher concentration of Black youth, staff burnout and stress
levels are likely elevated (Bottiani et al. 2014; Collie et al.
2012); stress, in turn, has been linked to expression of
implicit prejudicial biases (Kang et al. 2014). Therefore, it is
plausible that student report of equity may be lower overall in
these schools and that disparities in students’ report of school
support might be larger in such school settings.
The Present Study
To advance our knowledge of whether racial disparities in
students’ perceptions of school support are influenced by the
racial and ethnic diversity and socioeconomic status of the
school, this study employed multi-level and multiple group
latent variable approaches to model school equity, caring, and
high expectations as a 3-dimensional model of school support.
We then examined variation in racial differences in perceived
support by school context of racial and ethnic diversity and
school SES utilizing cross-sectional, self-report data from
19,726 students in 58 Maryland high schools. We tested three
central hypotheses. First, to ascertain whether racial gaps
persisted when accounting for student SES, we tested the
hypothesis that Black students would demonstrate signifi-
cantly lower mean scores relative to White students on per-
ceived caring, high expectations, and equity (Hughes and
Kwok 2007; Furlong et al. 2011), even after accounting for
maternal educational status (as a proxy for SES). Second, we
anticipated that, in more diverse and lower SES schools,
teachers and school staff would have higher levels of stress
and burnout (Bottiani et al. 2014), which would adversely
affect their capacity to promote an equitable, emotionally
supportive climate (Jennings and Greenberg 2009). Thus, we
hypothesized that, for both Black and White students, all three
dimensions of school support would be perceived less favor-
ably in more diverse and lower SES schools. Third, in schools
with more racial and ethnic diversity and lower SES, we
anticipated that the distinctions between the numeric majority
and minority student racial, ethnic, and class groups may be
tempered. Thus, we hypothesized that racial gaps in students’
perceptions of support would be attenuated in more diverse
and lower SES schools. Together, these findings have the
potential to motivate and inform practices used by schools and
districts to promote more equitable developmental outcomes
for Black youth.
Methods
Sample
Data for this study come from students attending 58 high
schools participating in the Maryland Safe and Supportive
Schools Project (MDS3), a statewide initiative focused on
promoting school climate, student engagement, and school
safety launched in 2012. Cross-sectional data were col-
lected from adolescents in grades 9–12 via a web-based
survey administered in spring 2013. The analysis sample
was limited to Black (n = 7062) and White (n = 12,662)
adolescents only, totaling 19,726 students who were
49.9 % male, with mean age = 15.89 (SD = 1.27).
Maternal education was less than high school graduation
for 8.3 % of students; 42.3 % reported that their mothers
graduated from college. Student demographics and school
characteristics are presented in Table 1.
Procedure
High schools were invited to participate in the MDS3 ini-
tiative on a voluntary basis. Twelve of the state’s 24 dis-
tricts were approached by the Maryland State Department
of Education. Half of the schools were randomly assigned
to receive training in positive behavior supports, and the
other half were randomized to a ‘‘business as usual con-
dition’’. The data for the present study were collected early
in the implementation of the initiative, and thus interven-
tion effects were not expected. Furthermore, our analyses
did not reveal intervention effects on the study variables.
Nevertheless, a school-level intervention condition variable
was included as a control variable in the models. Anony-
mous data were collected via a waiver of active parental
consent and a youth assent process. All student participa-
tion was voluntary. The MDS3 School Climate Survey was
administered online in language arts classrooms to
approximately 25 classrooms per school, with an approxi-
mate distribution as follows: seven 9th grade classrooms
and six each of 10th, 11th, and 12th grade classrooms.
School staff administered the survey following a written
protocol. The researchers’ Institutional Review Board
approved analysis of these data. For additional information
on the project, see Bradshaw et al. (2014a).
1178 J Youth Adolescence (2016) 45:1176–1191
123
Measures
The constructs described below were measured using items
from the MDS3 Student Survey. A collaborative led by the
Johns Hopkins Center for Youth Violence Prevention
developed the survey; for additional details, see Bradshaw
et al. (2014b). Cronbach’s alphas (a) were calculated to assess the internal consistency reliability of key constructs
in the study.
Student Demographic Characteristics
Participants responded to a series of questions regarding
sociodemographic characteristics, including gender,
maternal education, grade-level (9–12), and race/ethnicity.
For race/ethnicity, participants were asked to self-identify
as either Asian/Pacific Islander, Black/African American,
Hispanic/Latino, Native American/American Indian,
Native Hawaiian, White/Caucasian, or Other (Ensminger
et al. 2000).
School Support
Twelve survey items using a four-point Likert scale were
selected from the California Healthy Kids Survey (Hanson
and Kim 2007) and the School Development School Cli-
mate Survey (Haynes et al. 2001) to assess school support.
An exploratory factor analysis (EFA) in Stata using
Table 1 Demographic characteristics
Student characteristics (N = 19,726 students) N (%)
Gender
Male 9842 (49.9)
Female 9884 (50.1)
Race/ethnicity
Black 7064 (35.8)
White 12,662 (64.2)
Maternal education
Less than high school 1636 (8.3)
Graduated from high school 5617 (28.5)
Attended some college 4094 (20.8)
Graduated from college 8379 (42.5)
Grade
Grade 9 5666 (28.7)
Grade 10 4933 (25.0)
Grade 11 4750 (24.1)
Grade 12 4377 (22.2)
Age 15.89 (1.27)
School characteristics (J = 58 schools) M (SD)
Total enrollment 1262.9 (462.9)
Percentage Black 33.7 (24.4)
Percentage White 53.2 (25.1)
Normalized generalized variance (NGV) 57.0 (20.2)
Percentage of students receiving free and reduced price meals (FARMS) 37.5 (17.8)
School categories School
(#)
Student
(#)
Enrollment M (SD) % White
M (SD)
% Black
M (SD)
%
NGV
% FARMS
Primarily White, Upper SES 13 5386 1274.49 (343.82) 85.4 (4.4) 5.7 (3.5) 31.4 (8.5) 17.8 (10.1)
Primarily White, Lower SES 2 583 874.6 (183.5) 76.2 (0.3) 13.4 (1.3) 47.6 (1.0) 44.5 (2.9)
Primarily Black, Upper SES 2 561 1218.6 (288.2) 10.7 (1.1) 75.0 (2.5) 50.0 (3.7) 30.6 (1.3)
Primarily Black, Lower SES 4 1268 1165.0 (239.9) 2.2 (1.3) 90.4 (3.0) 21.4 (6.2) 58.3 (6.1)
Racially Diverse, Upper SES 16 6188 1609.4 (460.5) 47.7 (14.8) 37.0 (15.4) 69.9 (5.7) 23.7 (7.3)
Racially Diverse, Lower SES 21 5740 1202.5 (500.4) 49.4 (136) 35.6 (11.9) 70.5 (7.1) 53.4 (8.7)
J Youth Adolescence (2016) 45:1176–1191 1179
123
principal axis factoring (PAF) with oblique rotation of
twelve items employing a separate sample (schools from
the Year 1 survey administration, N = 21,824 students,
J = 52) retained a three-factor solution (based upon
examination of the scree plot, eigenvalues [1, and parallel analysis). The three factors were caring (4 items, a = .85; items: ‘‘My teachers care about me’’, ‘‘My teachers listen
when I have something to say’’, ‘‘Students trust the
teachers’’, ‘‘Teachers respect the students’’), high expec-
tations (4 items, a = .84; items: ‘‘My teachers encourage me to work hard in my classes’’, ‘‘My teachers believe that
I can do well in school’’, ‘‘My teachers always want me to
do my best’’, ‘‘Teachers believe all students can do well if
they try’’), and equity (4 items, a = .83; items: ‘‘The school provides instructional materials that reflect my
culture’’ and ‘‘At this school, students of all races [whether
boys or girls, whether parents are rich or poor] are treated
the same’’).
School Racial and Ethnic Diversity
Racial/ethnic diversity of each school was characterized
using a normalized generalized variance (NGV) statistic
(Budescu and Budescu 2012; Simpson 1949), which can be
interpreted as the probability of randomly selecting two
individuals from a given population that belong to different
subgroups (Budescu and Budescu 2012), wherein the
higher the value, the higher the diversity of the population.
The statistic was standardized (‘‘normalized’’) to create a
relative measure of diversity allowing for direct compar-
isons across groups (bounded ratio 0 B GV B 1). Groups
included in the calculation of the statistic for all schools
were school-level percentage Black, White, Latino,
American Indian/Alaska Native, Asian, and multi-eth-
nic/multiracial, as reported in concurrent school enrollment
records.
School SES
School SES was based upon the percentage of students
receiving free or reduced price meals (FARMs), such that a
higher FARMs rate indicated a higher concentration of low
SES students. Receipt of FARMs has been shown to be
valid indicator of low household income (Ensminger et al.
2000).
Analyses
We used a multilevel approach to examine our three
hypotheses. A multi-level approach was selected because it
allowed us to test our hypothesis of school-level modera-
tion of racial inequalities by examining cross-level inter-
actions of school diversity and SES on the association
between race and student report of school support dimen-
sions, while accounting for other within-level fixed and
random effects. Moreover, because the data (from students
nested within schools) were hierarchical in nature, indi-
viduals from the same schools likely have correlated errors,
and a basic assumption of multivariate regression would
otherwise be violated (Luke 2004). Multilevel modeling
allows for correlated error structures. Prior to testing our
three hypotheses, we conducted preliminary analyses to
assess the measurement model, measurement invariance,
and missing data.
Measurement Model
We used the current study’s sample to run a confirmatory
factor analysis (CFA) in Mplus utilizing MLR estimation
and sample weights. Degree of model fit was gauged by the
Chi square statistic (v2), comparative fit index (CFI; Ben- tler 1990), non-normed fit index [NNFI, also known as the
Tucker-Lewis Index (TLI); Bentler and Bonett 1980], and
the root-mean-square error of approximation (RMSEA)
with 90 % confidence interval (RMSEA; Steiger and Lind
1980). Adequate model fit of the measurement models
were determined by Chi square test insignificance [.05, CFI [ .95, TLI [ .95, and RMSEA \ .05. We found that a three-factor model of school support provided excellent fit
to the data v2 (51) = 1083.09, p \ .001, CFI = .98, TLI = .98, RMSEA = .03 (.032–.035). With large sample
sizes, the Chi square test is known to be sensitive (Marsh
et al. 1988). Alternative fit indices based on principals of
parsimony (i.e., RMSEA) were therefore referenced to
make decisions regarding model fit (Browne and Cudeck
1992).
Measurement Invariance
We examined measurement invariance in the factor struc-
ture of the caring, equity, and high expectations model
between the Black and White student groups through a
series of configural, metric, and scalar models (Meredith
1993), fit through multiple group CFA in Mplus with MLR
estimation. In testing metric invariance, we constrained
factor loadings to be equal across groups. Scale factors
were fixed at one in one group and free in the other group.
Factor variances were free to vary across groups, and factor
means were fixed at zero in one group and free in the other
group. In testing scalar invariance, we constrained factor
loadings and thresholds to be equal across groups. Scale
factors were fixed at one in one group and free in the other
group, and factor means were fixed at zero in one group
and free in the other group. Factor variances were free to
vary across groups. Consistent with Cheung and Rensvold
(2002), measurement invariance was found through the
1180 J Youth Adolescence (2016) 45:1176–1191
123
multi-group model demonstrating adequate fit to the data,
with the difference in CFI between models at less than .01.
When comparing metric against configural models,
v2 = 33.48 (df = 9), p \ .001, DCFI = -.001, DTLI = .001, and DRMSEA = -.001. When comparing scalar against configural models, v2 = 227.74 (df = 18), p \ .001, DCFI = -.003, DTLI = .000, and DRMSEA = .001. When comparing scalar against the constrained
metric model, v2 = 189.26 (df = 9), p \ .001, DCFI = -.002, DTLI = -.001, and DRMSEA = .002. These findings supported the assumption of measurement invari-
ance by race.
Missing Data
After limiting the student sample to those who provided
adequate initial demographic information (race, age, and
gender), descriptive analyses found very little missing data
(\1 % of students were missing items). Our analyses assumed that data were missing at random (MAR;
Arbuckle and Wothke 1999). Although the amount of
missing data in the study was negligible, analyses did
suggest that Black relative to White race was associated
with missing items for the equity and caring scales. How-
ever, the association of these variables with missingness
was quite small (estimates ranged from .001–.004). Mplus
software adjusts for missingness using full-information
maximum-likelihood (FIML) estimation, which is widely
recognized as an appropriate means of handling missing
data assumed to be MAR (Schafer and Graham 2002). The
sample was weighted to represent the school-wide popu-
lation using the raking method (Battaglia et al. 2004; see
Bradshaw et al. 2014a, b). Sample weights were utilized in
all models.
Multilevel Analyses
To examine our central research questions, we estimated
two-level models using Mplus 7.11. A stepwise approach
to model building was taken, such that the multilevel
models were built one variable and one level at a time in
order to be sensitive to the stability of findings with and
without nonsignificant effects (Raudenbush and Bryk
2002). For all outcome variables, we fit linear multilevel
models. We generated standardized coefficients as an effect
size to allow readers to assess the strength of the associa-
tions identified and their practical meaning (Nieminen et al.
2013). The overall fit of the models was assessed using
Akaike and Bayesian information criteria (Akaike 1974;
Raudenbush and Bryk 2002).
The three outcome variables caring, equity, and high
expectations were modeled as latent variables using items
measured at the student level (level 1). Factor loadings of
the latent variables in the between (level 2) model were
constrained to be equal to the level 1 factor loadings to
permit assessment of cross-level interactions (school-level
moderation). Predictors included at level 1 were grade-
level, gender, maternal education, and race (dummy coded
Black relative to White). Continuous level 1 covariates
were group-mean centered to allow for assessment of
between-group differences and cross-level interactions
(Croninger 2013). At level 2, we included the percentage of
students receiving FARMs, the NGV school diversity
statistic, and MDS3 intervention condition. All continuous
level 2 variables were grand-mean centered. To examine
whether school diversity moderated discrepancies between
Black and White students’ perceptions of caring, equity,
and high expectations, we tested cross-level effects of the
hypothesized school-level predictors (i.e., school diversity
and SES) at level 2 with the level 1 effects of school
support on race. The model included the level 1 and level 2
predictor variables’ main effects and corresponding cross-
level effects on racial differences.
Post-hoc Multiple Group CFA
To further explore and illustrate identified cross-level
interactions following the multi-level analyses, a post hoc
analysis was conducted. The post hoc analysis was
designed to be exploratory in nature, to help understand the
main results, and was necessary for two reasons. First, the
diversity statistic is informative only to a certain extent.
Specifically, an estimate of .29 could reflect either a pri-
marily homogenous Black school or a primarily homoge-
nous White school. In order to avoid conflating patterns in
primarily Black and White schools, schools in the upper
range of the grand mean diversity statistic (i.e., greater than
or equal to the mean NGV .57) were categorized as ‘‘di-
verse’’, whereas those in the lower range (i.e., \.57 NGV) were broken out as either majority White or majority Black
schools; this categorization was done to permit exploration
of patterns of disparity in the resulting three different types
of schools that we refer to as Primarily White, Primarily
Black, and Racially Diverse. A second reason for our post
hoc analysis was to allow us to understand how school
racial/ethnic composition potentially intersected with
school SES. The primary analysis only yielded information
on how school SES functioned in isolation. We therefore
further assigned each of the three types of schools (Pri-
marily White, Primarily Black, and Racially Diverse) to
either the upper or lower range of the grand mean per-
centage of enrollment on free and reduced price meals
(FARMS, M = .38), resulting in a total of six mutually
exclusive groups, which we call: (1) Primarily White,
Upper SES; (2) Primarily White, Lower SES; (3) Primarily
Black, Upper SES; (4) Primarily Black, Lower SES; (5)
J Youth Adolescence (2016) 45:1176–1191 1181
123
Racially Diverse, Upper SES; and (6) Racially Diverse,
Lower SES. Descriptives for each of the six resulting
school groups, including the number of schools in each
group, is provided at the bottom of Table 1. Using these six
groups, we then conducted a multiple-group CFA with
Black race as a covariate, which provided adequate fit to
the data, v2(450) = 2826.75, p \ .001, CFI = .98, TLI = .98, and RMSEA = .04 (95 % CI .039–.042). To
permit comparisons across the six school groups, mea-
surement invariance was tested and we found the
assumption of measurement invariance held. When com-
paring metric against configural models, v2 = 144.77 (df = 45), p \ .001, DCFI = -.001, DTLI = .001, and DRMSEA = -.002. When comparing scalar against con- figural models, v2 = 560.80 (df = 90), p \ .001, DCFI = -.005, DTLI = .000, and DRMSEA = .000. When comparing scalar against the constrained metric
model, v2 = 416.03 (df = 45), p \ .001, DCFI = -.004, DTLI = -.001, and DRMSEA = .002.
Results
Figure 1 depicts the two-level latent variable model of
caring, equity, and high expectations with coefficients for
level 1 and 2 predictors and cross-level interactions
labeled. Table 2 summarizes these findings in tabular form.
As hypothesized, Black race was significantly negatively
associated with the latent variables caring (c = -.14,
p \ .001) and equity (c = -.10, p \ .001), even when accounting for maternal education, a proxy for student-
level SES. Contrary to our hypothesis, no significant dif-
ferences by race were found in latent high expectations
(c = .01, n.s.). We also observed a statistically significant positive association between maternal education level and
perceptions of school support across all three latent vari-
ables (c ranged from .03 to .05, all p \ .001). Being male was significantly positively associated with perceived car-
ing and equity (c = .05 for both, with both coefficients’ p \ .001), but was negatively associated with high expectations (c = -.03, p = .02). Interestingly, we found no linear association between grade level (9–12) and per-
ceived caring or equity, but we observed a negative asso-
ciation with high expectations such that as grade level
increased, latent perceived high expectations decreased
(c = -.04, p \ .001). Overall, these standardized associ- ations by gender, grade level, and maternal education were
relatively smaller than associations found by race for car-
ing and equity.
Our second hypothesis was that, for both Black and
White students, all three dimensions of school support
would be perceived less favorably in more diverse and
lower SES schools. Examining the level 2 main effects
reflecting the three dimensions regressed on NGV (diver-
sity) and FARMs (low SES), significant associations were
found in the predicted direction for both Black and White
students on equity (NGV c = -.16, p \ .001 and FARMs c = -.21, p \ .001). In addition, the hypothesized
Fig. 1 Dashed paths indicate non-significant p values. Except
where indicated, for all other
significant paths, p \ .001. Error terms and correlations
between latent variables are not
depicted. Care caring, High
Expect high expectations
1182 J Youth Adolescence (2016) 45:1176–1191
123
Table 2 Two-level models examining school context of diversity and socioeconomic status and student-reported caring, equity, and high expectations
Caring Equity High expectations
c p SE t c p SE t c p SE t
Student-level variables
Black race -0.14 *** .02 -6.83 -0.10 *** .03 -3.61 0.01 ns .02 0.28
Maternal education 0.05 *** .01 6.81 0.03 *** .01 4.26 0.05 *** .01 6.30
Grade level 0.01 ns .01 1.21 0.00 ns .01 -0.29 -0.04 *** .01 -4.20
Male gender 0.05 *** .02 3.44 0.05 *** .01 3.29 -0.03 ** .01 -2.42
School-level variables
NGV -0.07 ns .04 -1.85 -0.16 *** .05 -3.63 -0.08 * .04 -2.28
FARMs -0.03 ns .03 -1.06 -0.21 *** .04 -5.13 -0.01 ns .03 -0.54
Condition -0.03 ns .03 -0.89 0.01 ns .04 0.32 -0.04 ns .03 -1.65
Cross-level interactions
NGV 9 Black 0.03 ns .02 1.30 0.11 *** .03 3.25 0.05 ns .03 1.80
FARMs 9 Black 0.00 ns .02 -0.19 -0.10 *** .02 3.97 0.01 ns .02 0.56
% Between-school variance explained Caring (%) Fair treatment/inclusion (%) High expectations (%)
Within and between 36.4 39.8 13.8
Within, Between, and cross-level 48.8 46.6 11.5
AIC BIC Sample size-adjusted BIC
No covariates 547,834.00 548,189.04 548,046.03
Within and Between 547,349.47 547,870.19 547,660.44
Within, between, and cross-level 547,275.15 547,866.88 547,628.53
Coefficients are standardized. N = 19,726 students, J = 58 schools. Unadjusted ICCs, caring = .03, equity = .04, high expectations = .01.
Correlation of caring and equity = .36; correlation of equity and high expectations = .30; correlation of caring and high expectations = .47.
Correlation of FARMs and NGV = .28
ns non-significant
* p \ .05; ** p \ .01; ***p \ .001
Fig. 2 Plots of the cross-level interactions of racial differences in student report of school equity at the mean, one standard deviation (SD) below the mean, and one SD above the mean school percentage free and reduced price meals (left) and school heterogeneity diversity (right)
J Youth Adolescence (2016) 45:1176–1191 1183
123
negative association between school diversity and high
expectations was found to be significant (c = -.08, p = .02). The hypothesized associations when regressing
latent caring on school diversity and FARMs were non-
significant, although they trended in the predicted direc-
tion. There was no significant association found between
FARMs and high expectations.
Our third hypothesis was that racial gaps in students’
perceptions of support would be attenuated in more diverse
and lower SES schools. We found significant moderation
of Black–White gaps in the predicted direction by school
NGV (c = .11, p \ .001) and FARMs (c = .10, p \ .001) for equity only. As shown in Fig. 2, although student
perceptions of equity are lower overall for both Black and
White students in lower SES schools, the gap between
Black and White students is larger in higher SES schools.
Similarly, in more racially and ethnically homogenous
schools, the gap in perceived equity between Black and
White students was larger than in schools with high
diversity (although more diversity was associated overall
with lower levels of perceived equity).
Our post hoc multiple group analysis suggested the
importance of examining racial differences in student
perceptions of school equity in the context of school
diversity, predominant race, and school SES. As we note in
the Methods section, low-diversity schools were
categorized as either ‘‘primarily White’’ or ‘‘primarily
Black’’, whereas schools with NGVs higher than the mean
were classified as ‘‘racially diverse’’, creating three school
types. When further broken down by either upper or lower
SES, six mutually exclusive groups were created. Due to
the correlation between SES and racial composition of
schools (i.e., primarily Black and more racially diverse
schools tend to be lower SES), there were fewer schools in
the ‘‘Primarily Black, Upper SES’’ and the ‘‘Primarily
White, Lower SES’’ groups, whereas there were many
more schools in the ‘‘Racially Diverse, Low SES’’ group
and the ‘‘Primarily White, Upper SES’’ group. Because of
the low number of schools in some of the groups, we
caution that the results of the post hoc analysis are intended
only to further clarify trends identified in the main findings,
due to limitations of the NGV statistic. As shown in Fig. 3,
a pattern consistent with the cross-level interactions
described above was found; however, disparities by race in
latent perceptions of equity were magnified across school
types. Specifically, in schools characterized as primarily
White and upper SES, Black students’ perceptions of
equity were nearly a half a standard deviation lower than
White students’, on average (b = -.49, p \ .001). The Black–White gap was significantly larger in this school
type than in all other school types, except the White lower
SES type (Wald v2s ranged from 19.43 to 51.54, p \ .001
-0.49
-0.84
-0.75
-1.07
-0.59
-0.84
0.00
-0.48
-1.09
-1.40
-0.55
-0.91
-1.50 -1.40 -1.30 -1.20 -1.10 -1.00 -0.90 -0.80 -0.70 -0.60 -0.50 -0.40 -0.30 -0.20 -0.10 0.00
Primarily White, Upper SES
Primarily White, Lower SES
Primarily Black, Upper SES
Primarily Black, Lower SES
Racially Diverse, Upper SES
Racially Diverse, Lower SES
EQUITY ON SCHOOL TYPE BY RACE
Black White
Fig. 3 Estimates of equity on school type by race from a multiple group confirmatory factor analysis, which provided adequate fit to the
data, v2 (450) = 2826.75, p \ .001, CFI = .98, TLI = .98, and RMSEA = .04 (95 % CI .039–.042). Tests of moderation indicated
that disparity estimates in student report of school equity among
Black and White high school students significantly differed by school
type. School type is a grouping indicator that reflects both percentage
of student enrollment on FARMS and racial/ethnic composition of
student enrollment. School racial composition categories were created
by assigning schools within the upper and lower ranges of the grand
mean diversity statistic [normalized generalized variance (NGV)
mean = .57]. Schools with low diversity (i.e., \.57 NGV), were further classified as either majority White or majority Black schools,
resulting in three school types: Primarily White, Primarily Black, and
Racially Diverse. Then the three school types were further classified
as either upper or lower SES using the free and reduced price meals
mean .38 as the threshold. The process resulted in six mutually
exclusive groups. Detailed information on each of the six school
categories is provided at the bottom of Table 1
1184 J Youth Adolescence (2016) 45:1176–1191
123
for each of the four significant comparisons). It is important
to note that Black students’ perceptions of equity were
nonetheless higher in majority White, upper SES schools
relative to all other school groups. Black students’ report of
equity was lowest in schools that were majority Black and
in the lower SES range. Notably, report of equity among
Black students in diverse, upper SES range schools was
higher relative to homogenous White, lower SES schools
and homogenous Black, lower and upper SES schools. As a
result, disparities between Black and White students were
significantly smaller in diverse, upper SES schools relative
to homogenous White, lower SES schools (Wald
v2 = 17.97, p \ .001) and homogenous Black, lower and upper SES schools (Wald v2 s ranged from 4.41 to 5.33, p \ .01 for both).
Discussion
Supportive school contexts in which adults are caring, fair,
and set high expectations have been recognized as essential
to adolescents’ healthy cognitive, social, and emotional
development (Eccles and Roeser 2011). Persistent racial
disparities in students’ school outcomes (e.g., academic
performance, school discipline; Aud et al. 2012; Skiba
et al. 2011) mirror the available research on youth per-
ceptions of school support by race, which show that Black
youth perceive lower levels of supportive relationships
with adults and connectedness at school relative to their
White peers (Furlong et al. 2011; Hamre and Pianta 2001;
Hughes and Kwok 2007). Yet, there are few studies
accounting for student SES to demonstrate that racial gaps
in perceived climate are not conflated with socioeconomic
differences (Bottiani et al. 2014; Voight et al. 2015), and
none which examine the role of the school contexts in
which race interacts to shape student perceptions of school
support. Given prior research showing that students report
more positive racial climate in schools with a critical mass
of same-race peers (Benner and Graham 2013), it is pos-
sible that students perceive some enculturative benefit from
being a student of the school’s majority group; or, it is also
possible that Black and White students may become more
aware of differential treatment within schools in which
they have more exposure to students with backgrounds that
differ from their own. These considerations of race-in-
context are critical to improving our understanding of the
processes that contribute to racial disparities in school
outcomes.
Using a multi-level latent variable modeling approach,
this study explored two aspects of school context, racial
and ethnic diversity and socioeconomic status, as they
related to racial gaps in students’ perceptions of caring,
equity, and high expectations. Across three theorized
dimensions of school support, we were interested to learn
whether racial gaps persisted when accounting for student
SES, whether school diversity and school SES influenced
students’ perceptions of support overall, and whether stu-
dents’ perceptions varied differentially by race according
to the schools’ diversity and SES. Our findings demon-
strated that racial disparities were present in Black and
White students’ perceptions of caring, such that Black
students perceived significantly lower caring relative to
their White peers. This gap did not vary significantly by the
diversity of their peers or school socioeconomic context. In
contrast, no racial differences in students’ report of teacher
high expectations were identified; however, we did find
that this dimension of support decreased for both Black and
White youth as the diversity of the student body in their
school increased. Consistent with our hypotheses, we found
lower levels of perceived equity among both Black and
White students in more racially and ethnically diverse
schools. Similarly, perceived equity was lower for both
groups in lower SES schools. With regard to racial dif-
ferences, we found that Black students had lower scores on
the equity measure on average relative to their White peers,
but these gaps were moderated by school contextual vari-
ables. Specifically, White students’ perceptions of equity
were significantly lower relative to those of Black students
at schools with more diversity and low-income represen-
tation of the student body. Conversely, Black students’
perceptions of equity were significantly lower than White
students’ perceptions in schools that were primarily White
and upper SES. These findings highlight the importance of
examining person-context congruence as it relates to stu-
dents’ perceptions of school support and raise new ques-
tions about the benefits and challenges presented by the
increasing racial and ethnic diversity of our schools.
Caring
The significant finding of lower perceived teacher caring
among Black students is noteworthy in this study because it
was found even when accounting for maternal education (a
proxy for socioeconomic status) and regardless of changes
in school socioeconomic context or diversity, suggesting
further exploration of potential contributing factors is
merited. The finding that perceived caring was lower for
Black students and not malleable to changes in context of
school diversity or school SES suggests that perceived
caring may not vary in reference to the perceived treatment
of peers, but that the student–teacher relationship itself
may be a more apt target for further research and reform.
The finding adds to continued concerns regarding the
overarching differential effectiveness of schools in edu-
cating Black as compared with White students (Bingham
and Okagaki 2012; KewalRamani et al. 2007). The latent
J Youth Adolescence (2016) 45:1176–1191 1185
123
variable caring included aspects related to perceived trust,
which has been linked to defiance and behavior problems
in the classroom (Gregory and Ripski 2008; Gregory and
Weinstein 2008). Lower levels of perceived caring are also
worrisome as poorer quality relationships with teachers are
associated with a lower sense of belonging (Crouch et al.
2014), which in turn is linked to a host of negative aca-
demic, social-emotional, and behavioral outcomes (Hughes
and Kwok 2007; Resnick et al. 1998). A recent compre-
hensive review of the school as a developmental context
during adolescence highlighted that ‘‘a sense of belonging
may be especially critical for young people who must
traverse significant ethnic and racial, socioeconomic, and
sociolinguistic borders to feel fully a part of a school in
which middle-class, majority cultural norms often pre-
dominate’’ (Eccles and Roeser 2011, p. 229). Thus, Black
students may benefit especially from approaches that
enhance the quality of their relationships with their
teachers and increase their sense of belonging and con-
nection at school.
High Expectations
Contrary to our hypothesis, Black and White youth per-
ceived similar levels of success expectancies from their
teachers. This finding is inconsistent with literature high-
lighting the problem of lowered expectations and deficit
thinking among some educators towards students of color
(Garcia and Guerra 2004; McKenzie and Scheurich 2004).
Deficit thinking refers to some educators’ presumption that
within-student ‘‘deficiencies’’ (e.g., family dysfunction,
lack of valuing education, linguistic inferiority) among
students of color and low-income students are the cause of
poorer achievement among these populations. Deficit
thinking stands in contrast to strengths-based and culturally
responsive practices, which emphasize student assets and
deliberately maintain high expectations for success.
Research on deficit-thinking suggests its salience as a
barrier to supportive and positive relationships between
White teachers and students of color, and particularly
students from low-income families or identified within
disability categories (Trent et al. 1998). However, it is
possible that student-reports of teacher expectations are
less informative in addressing research questions on deficit
thinking relative to teacher-reports or observational mea-
sures, as teachers’ success expectancies may be less easily
discerned by students (which could explain why we found
gaps for caring and equity, but not high expectations). It
may also be helpful to interpret this null finding in light of
research documenting the ‘‘engagement-achievement
paradox’’ (Shernoff and Schmidt 2008), a pattern found in
multiple studies in which Black students reported higher
levels of academic success expectancies relative to peers,
despite demonstrating lower levels of academic achieve-
ment (Dotterer et al. 2009; Shernoff and Schmidt 2008).
Although there is no consensus on the mechanisms
explaining the paradox, it is nonetheless plausible to con-
sider that, if they report inflated success expectancies rel-
ative to their peers and relative to their academic
performance, Black students also may report inflated per-
ceptions of their teachers’ academic expectations of them.
Thus, additional research exploring concordance in teach-
ers’ and students’ report of academic expectations may be
informative to advance our understanding of possible bias
in teachers’ academic expectations of students of color.
Another possible explanation of the engagement-achieve-
ment paradox is that perceived discrimination may mod-
erate the association between students’ perceptions of
teachers’ success expectancies and academic outcomes, as
perceived discrimination may cause students to mistrust the
academic performance assessment process. Future research
directly examining these plausible mechanisms is needed.
With regard to our finding of poorer perceived expectations
in more racially and ethnically diverse schools, the asso-
ciation held even when accounting for the school’s per-
centage of students receiving FARMs. This finding was
consistent with our hypothesis, and aligns with the deficit
thinking and lowered expectations literature (Ferguson
2003).
Equity
Our findings indicated that students’ racial position within
the school racial/ethnic context was significantly associated
with their perceptions of school equity. As with perceived
teacher caring, we found that Black students reported sig-
nificantly lower equity on average as compared with White
students and furthermore observed that the Black–White
gap was most discordant in schools that were majority
White or majority Black (i.e., the most racially and ethni-
cally homogenous schools), however, the disparity
reversed depending on the majority race of the school.
Specifically, Black students had less favorable perceptions
of equity in majority White schools relative to White stu-
dents, whereas White students had less favorable percep-
tions of equity in majority Black schools relative to Black
students.
Over the past several decades, research has established
the importance of the interaction of students with school
contexts as a determinant of adolescent developmental
outcomes (Eccles and Roeser 2011). In applying this per-
son-context perspective to understand racial differences in
adolescents’ perceptions of school equity, it is necessary to
consider race and socioeconomic status as social position
variables whose implications vary in relation to the broader
socioeconomic and racial context in which young people
1186 J Youth Adolescence (2016) 45:1176–1191
123
find themselves (Garcı́a Coll et al. 1996). Bandura (1999)
theorized that people differentially evoke responses from
their social environment by their socially conferred status,
which is informed in large part by a person’s race and class
in relation to the majority group. So, if there is no majority
group, it seems plausible that clearly discerned hierarchies
influencing differential treatment could break down.
Yet, this does not fully explain the findings evident in
Fig. 3 (post hoc analysis), which showed that, although
gaps in perceived equity narrowed (fitting with the above),
overall perceptions of equity were still lower in more
heterogeneous schools. Another plausible explanation is
that being a student in a more diverse school setting could
raise White students’ awareness of and sensitivity to
inequitable treatment when it is present. Although in our
sample there were a number of schools with a highly
diverse student enrollment, the vast majority of the sample
schools’ teachers were White, which highlights another
possible source of students’ socially conferred status, other
than their peers. Specifically, a recent report by the
Maryland State Department of Education (MSDE) indi-
cated that district-level percentages of White teachers
ranged from 85.1 to 97.4 for the 12 districts represented in
this study (MSDE 2012). In short, it is plausible that, in
more diverse schools, there is quantitatively more inequity
experienced by students of color in their interactions with a
relatively homogenous White staff. White students’
awareness of this inequity may be increased as a result of
greater exposure to inequity, and possibly also result from
increased cross-racial/ethnic peer relationships. These
potential mechanisms merit exploration to improve our
understanding of the patterns observed in students’ per-
ceptions of equity.
Limitations and Strengths
A few limitations should be considered when interpreting
the results of this study. We relied on cross-sectional data,
which precluded our ability to examine links between
disparate school support and subsequent student disparities.
However, we did take advantage of the large sample size
and nested structure of the data to address important and
relevant contextual factors that shape differential percep-
tions of school support. Future research employing longi-
tudinal data should examine how differential perceptions of
school support can be linked to subsequent gaps in student
outcomes. An additional limitation is our reliance on stu-
dent self-report data. It is often preferable to utilize mul-
tiple informants’ report to strengthen validity and causal
hypotheses. Nonetheless, it is a strength of this study to
have insights into students’ perceptions, because such
perceptions are themselves predictive of developmental
outcomes, and may be key to understanding racial
disparities. Faculty diversity in our sample limited our
ability to examine the role of staff racial/ethnic composi-
tion. The faculty in these schools were overwhelmingly
White, with the overall sample of teachers being 84 %
White and less than 10 % Black, with little substantive
variation in teacher diversity by school. Based on prior
research suggesting that students’ perceptions of discrimi-
nation are mitigated in schools with greater school faculty
diversity (Benner and Graham 2011; Seaton and Yip 2009),
future research should examine the interaction of faculty
and student diversity as contextual influences on racial
differences in student perceptions of school support and
school equity in particular.
Another limitation of this study was the way in which
equity was operationalized. Although the results suggest
that racial bias may be a source of the disparities between
White and Black students’ perceptions of fair treatment,
the measure does not exclusively focus on culture, race,
and ethnicity; it also includes items related to gender and
socioeconomic status. This broader conceptualization of
equity might have weakened our findings on gaps in per-
ceived equity if the source of the disparity had to do with
students’ perceptions of racial bias. Furthermore, the
measure does not ask which students they feel were
unfairly treated (i.e., boys or girls? Black students or White
students?). This may seem obvious, but when looking at
lower levels of White students’ perceived equity in pre-
dominantly Black, low SES schools in Fig. 2, for example,
it is unclear whether White students feel more unfairly
treated themselves, whether they perceive less fair treat-
ment of their Black peers, or whether unfair treatment by
gender or SES should be attributed. However, our purpose
in this study was to understand disparities in Black and
White students’ perceptions of fair treatment more broadly
as an indicator of discordance in the social environment in
the school, not to assess students’ experiences of discrim-
ination per se. There are measures of discrimination that
look into these questions (unfair treatment by whom, for
what reason, e.g., see Children of Immigrants Study, Portes
and Rumbaut 2005), which we may adopt in future
research to further understand the mechanisms contributing
to our findings in this study.
Implications for Practice
Regardless of the diversity of the school context, we found
a significant, negative linear association for low-income
schools on both Black and White students’ perceptions of
equity. There may be factors within low-income schools
that lead to either higher sensitivity among students of
unfair treatment (possibly due to socialization), or it may
be that demonstrably less fair treatment by teachers is
occurring in these contexts (possibly triggered by stress).
J Youth Adolescence (2016) 45:1176–1191 1187
123
Although further research is needed to inform our knowl-
edge of the mechanisms underlying this finding, investment
in approaches that facilitate student voice in school reform
processes may be a constructive local strategy to uncover
the sources of (and remedy) perceived inequities. For
example, in the Culturally-Responsive Positive Behavioral
Interventions and Supports model (Bal et al. 2012), school
leadership employs ‘‘learning labs’’ to engage students’ and
families’ involvement and empowerment as key stake-
holders in school-wide systemic change process to promote
more positive school climate. Studies assessing the effects
of this approach are in progress.
The findings regarding perceived caring suggest that
intervention to improve Black students’ perceptions of
teacher caring may be needed and that such intervention
may do better to focus on the student–teacher relationship
itself, rather than targeting the broader social environment
or sociopolitical context of the school for reform. Given
racial and ethnic differences between a majority White
teaching workforce (Zumwalt and Craig 2005) and a
majority Black and Latino urban student population (Sable
et al. 2010), school staff in diverse schools may benefit
from on-the-job training to recognize and bridge cultural
and ecological gaps in their relationships with students
whose backgrounds differ from their own (Delpit 2006;
Gay 2010; Ladson-Billings 2009) in order to help promote
improvements in Black students’ perceptions of caring.
One approach that shows promise is the Double Check
model (Bottiani et al. 2012; Bradshaw and Rosenberg in
press; Hershfeldt et al. 2009), which uses a CARES
framework to focus on concrete skills teachers can practice
to enhance cultural Connections to curricula, Authentic
relationships, Reflective thinking, Effective communica-
tion, and Sensitivity to student culture. Although prelimi-
nary research suggests the potential of this intervention
(Bradshaw et al. 2015; Bradshaw and Rosenberg in press),
more work is needed to establish its effectiveness and to
determine effective dissemination strategies in partnership
with schools.
Conclusion
Understanding factors that contribute to disparities in per-
ceived supportive relationships is an important research
agenda within our broader efforts to identify and ultimately
eliminate disparities in schools. In this study, we found that
students’ perceptions of caring and equitable treatment
were more negative for Black students relative to their
white peers, even when accounting for student SES, and
that students’ perceptions of equity and high expectations
overall were more negative in more diverse, lower income
schools. Although perceived equity was better overall for
both Black and White students in primarily White, upper
SES schools, racial gaps in perceived equity were also
more salient in these contexts. The findings suggest that it
is necessary to examine person-context fit when exploring
racial disparities in perceived school support and point to
the need for intervention to improve perceptions of school
support for Black youth and for all students in lower
income, diverse schools. Negative interactions with adults
at school can lead to increased exposure to exclusionary
school discipline (Griffiths et al. 2012), which in turn can
have a chain of detrimental effects on developmental out-
comes (APA 2008). Interventions to engage student voice
in problem-solving, broach the topic of race (Day-Vines
et al. 2007), and enhance the quality of Black students’
relationships with their teachers have potential to shift
these dynamics to more positive trajectories.
Acknowledgments This work was funded in part by Grants from the U.S. Department of Education (IES R324A110107 and
R324A07118) and William T. Grant Foundation, awarded to
Catherine Bradshaw of Johns Hopkins University. This work was also
funded by the Institute of Education Sciences National Center for
Education Research Predoctoral Interdisciplinary Research Training
Program in the Education Sciences at the Johns Hopkins University
and the Institute of Education Sciences. We would like to thank the
Maryland State Department of Education and Sheppard Pratt Health
System for their support of this research through the Maryland Safe
and Supportive Schools Project.
Author Contributions JB conceived of the study, participated in its design and coordination and drafted the manuscript; CB participated
in the design and interpretation of the data, and helped with drafting
the manuscript; TM participated in the interpretation of the data and
helped to draft the manuscript. All authors read and approved the final
manuscript.
Funding This study was funded in part by grants from the U.S. Department of Education (IES R324A110107 and R324A07118) and
William T. Grant Foundation.
Conflicts of interest The authors report no conflict of interests.
Ethical Approval All procedures performed in studies involving human participants were in accordance with the ethical standards of
the institutional review board and with the 1964 Helsinki declaration
and its later amendments or comparable ethical standards.
Informed Consent Informed consent was obtained from all indi- vidual participants included in the study.
References
Advancement Project. (2010). Test, punish and push out: How zero
tolerance and high-stakes testing funnel youth into the school to
prison pipeline. The Advancement Project. Retrieved from http://
b.3cdn.net/advancement/d05cb2181a4545db07_r2im6caqe.pdf.
Akaike, H. (1974). A new look at the statistical model identification.
IEEE Transactions on Automatic Control, 19, 716–723.
American Psychological Association, Task Force on Resilience and
Strength in Black Children and Adolescents. (2008). Resilience
1188 J Youth Adolescence (2016) 45:1176–1191
123
in African American children and adolescents: A vision for
optimal development. Washington, DC: Author. Retrieved from
http://www.apa.org/pi/cyf/resilience.html.
Arbuckle, J., & Wothke, W. (1999). AMOS 4 user’s reference guide.
Chicago, IL: Smallwaters Corp.
Aud, S., Hussar, W., Johnson, F., Kena, G., Roth, E., Manning, E.,
et al. (2012). The condition of education 2012. NCES 2012-045:
National Center for Education Statistics. Retrieved from http://
nces.ed.gov/pubs2012/2012045_5.pdf.
Baker, J. A., Grant, S., & Morlock, L. (2008). The teacher–student
relationship as a developmental context for children with
internalizing or externalizing behavior problems. School Psy-
chology Quarterly, 23(1), 3–15. doi:10.1037/1045-3830.23.1.3.
Bal, A., Thorius, K. K., & Kozleski, E. (2012). Culturally responsive
positive behavioral support matters. Tempe, AZ: The Equity
Alliance. Retrieved from http://www.psd150.org/cms/lib2/
IL01001530/Centricity/Domain/24/CRPBIS_Matters.pdf.
Bandura, A. (1999). Social cognitive theory: An agentic perspective.
Asian Journal of Social Psychology, 2(1), 21–41. doi:10.1111/
1467-839x.00024.
Battaglia, M., Izrael, D., Hoaglin, D., & Frankel, M. (2004). Tips and
tricks for raking survey data (A.K.A. sample balancing).
Conference Papers—American Association for Public Opinion
Research. Retrieved from http://www.amstat.org/sections/srms/
Proceedings/y2004/files/Jsm2004-000074.pdf.
Benard, B. (2004). Resiliency: What we have learned. San Francisco,
CA: WestEd.
Benner, A. D., & Graham, S. (2009). The transition to high school as
a developmental process among multiethnic urban youth. Child
Development, 80, 356–376. doi:10.1111/j.1467-8624.2009.
01265.x.
Benner, A. D., & Graham, S. (2011). Latino adolescents’ experiences
of discrimination across the first 2 years of high school:
Correlates and influences on educational outcomes. Child
Development, 82(2), 508–519.
Benner, A. D., & Graham, S. (2013). The antecedents and
consequences of racial/ethnic discrimination during adolescence:
Does the source of discrimination matter? Developmental
Psychology, 49, 1602. doi:10.1037/a0030557.
Bentler, P. M. (1990). Comparative fit indexes in structural models.
Psychological Bulletin, 107(2), 238–246. doi:10.1037/0033-
2909.107.2.238.
Bentler, P. M., & Bonett, D. G. (1980). Significance tests and
goodness of fit in the analysis of covariance structures.
Psychological Bulletin, 88(3), 588–606. doi:10.1037/0033-
2909.88.3.588.
Bingham, G. E., & Okagaki, L. (2012). Ethnicity and student
engagement. In S. L. Christenson, A. L. Reschly & C. Wylie
(Eds.), Handbook of research on student engagement (pp.
65–95). New York, NY: Springer. Retrieved from www.
springer.com.
Bogart, L., Elliott, M., Kanouse, D., Klein, D., Davies, S., Cuccaro,
P., et al. (2013). Association between perceived discrimination
and racial/ethnic disparities in problem behaviors among pread-
olescent youths. American Journal of Public Health, 103(6),
1074–1081. doi:10.2105/AJPH.2012.301073.
Bottiani, J. H., Bradshaw, C. P., & Mendelson, T. (2014). Promoting
an equitable and supportive school climate in high schools: The
role of school organizational health and staff burnout. Journal of
School Psychology, 52, 567–582. doi:10.1016/j.jsp.2014.09.003.
Bottiani, J. H., Bradshaw, C. P., Rosenberg, M. S., Hershfeldt, P. A.,
Pell, K. L., & Debnam, K. D. (2012). Applying double check to
response to intervention: Culturally responsive practices for
students with learning disabilities. Insights on Learning Disabil-
ities, 9, 93–107.
Bradshaw, C. P., Debnam, K. J., Lindstrom Johnson, S., Pas, E.,
Hershfeldt, P., Alexander, A., et al. (2014a). Maryland’s
evolving system of social, emotional, and behavioral interven-
tions in public schools: The Maryland Safe and Supportive
Schools Project. Adolescent Psychiatry, 4(3), 194–206. doi:10.
2174/221067660403140912163120.
Bradshaw, C. P., Pas, E., Debnam, K., Bottiani, J., Reinke, W.,
Herman, K., & Rosenberg, M. (2015). Coaching teachers to
reduce disproportionality through cultural proficiency: Prelimi-
nary impacts of the Double Check model. Manuscript in
preparation.
Bradshaw, C. P., & Rosenberg, R. (in press). Culturally responsive
behavior management. Guilford Press: New York, NY.
Bradshaw, C. P., Waasdorp, T. E., Debnam, K. J., & Lindstrom
Johnson, S. (2014b). Measuring school climate: A focus on
safety, engagement, and the environment. Journal of School
Health, 84, 593–604. doi:10.1111/josh.12186.
Browne, M. W., & Cudeck, R. (1992). Alternative ways of assessing
model fit. Sociological Methods and Research, 21(2), 230–258.
doi:10.1177/0049124192021002005.
Budescu, D. V., & Budescu, M. (2012). How to measure diversity
when you must. Psychological Methods, 17, 215. doi:10.1037/
a0027129.
Byrd, C. M., & Chavous, T. (2011). Racial identity, school racial
climate, and school intrinsic motivation among African Amer-
ican youth: The importance of person–context congruence.
Journal of Research on Adolescence, 21(4), 849–860.
Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit
indexes for testing measurement invariance. Structural Equa-
tion Modeling, 9, 233–255. doi:10.1207/S15328007SEM09025.
Collie, R. J., Shapka, J. D., & Perry, N. E. (2012). School climate and
social-emotional learning: Predicting teacher stress, job satis-
faction, and teaching efficacy. Journal of Educational Psychol-
ogy, 104(4), 1189.
Croninger, R. G. (2013). Introduction to multilevel modeling. College
Park, MD: University of Maryland.
Crouch, R., Keys, C. B., & McMahon, S. D. (2014). Student–teacher
relationships matter for school inclusion: School belonging,
disability, and school transitions. Journal of Prevention and
Intervention in the Community, 42, 20–30. doi:10.1080/
10852352.2014.855054.
Day-Vines, N. L., Wood, S. M., Grothaus, T., Craigen, L., Holman, A.,
Dotson-Blake, K., & Douglass, M. J. (2007). Broaching the
subjects of race, ethnicity, and culture during the counseling
process. Journal of Counseling and Development, 85(4), 401–409.
Debnam, K. J., Johnson, S. L., Waasdorp, T. E., & Bradshaw, C. P.
(2014). Equity, connection, and engagement in the school
context to promote positive youth development. Journal of
Research on Adolescence, 24, 447–459.
Decker, D. M., Dona, D., & Christenson, S. L. (2007). Behaviorally
at-risk African American students: The importance of student–
teacher relationships for student outcomes. Journal of School
Psychology, 45(1), 83–109. doi:10.1016/j.jsp.2006.09.004.
Delpit, L. (2006). Other people’s children: Cultural conflict in the
classroom. New York, NY: The New Press.
Dotterer, A. M., McHale, S. M., & Crouter, A. C. (2009).
Sociocultural factors and school engagement among African
American youth: The roles of racial discrimination, racial
socialization, and ethnic identity. Applied Developmental
Science, 13(2), 61–73. doi:10.1080/10888690902801442.
Eccles, J. S., & Roeser, R. W. (2011). Schools as developmental
contexts during adolescence. Journal of Research on Adoles-
cence, 21(1), 225–241. doi:10.1111/j.1532-7795.2010.00725.x.
Elliot, A. J., & Dweck, C. S. (2005). Competence and motivation:
Competence as the core of achievement motivation. In A.
J Youth Adolescence (2016) 45:1176–1191 1189
123
J. Elliot & C. S. Dweck (Eds.), Handbook of competence and
motivation (pp. 3–12). New York, NY: Guilford Publications.
Ensminger, M. E., Forrest, C. B., Riley, A. W., Kang, M., Green, B.
F., Starfield, B., & Ryan, S. A. (2000). The validity of measures
of socioeconomic status of adolescents. Journal of Adolescent
Research, 15, 392–419. doi:10.1177/0743558400153005.
Fabelo, T., Thompson, M. D., Plotkin, M., Carmichael, D., March-
banks, M. P. III, & Booth E. A. (2011). Breaking schools’ rules:
A statewide study of how school discipline relates to students’
success and juvenile justice involvement. New York, NY;
College Station, TX: Council of State Governments Justice
Center; Public Policy Research Institute of Texas A&M
University. Retrieved from http://justicecenter.csg.org/files/
Breaking_Schools_Rules_Report_Final.pdf.
Ferguson, R. F. (2003). Teachers’ perceptions and expectations and
the Black–White test score gap. Urban Education, 38(4),
460–507. doi:10.1177/0042085903038004006.
Finn, J. D., & Zimmer, K. S. (2012). Student engagement: What is it?
Why does it matter? In S. L. Christenson, A. L. Reschly, & C.
Wylie (Eds.), Handbook of research on student engagement (pp.
97–131). New York, NY: Springer.
Furlong, M. J., O’Brennan, L. M., & You, S. (2011). Psychometric
properties of the Add Health School Connectedness Scale for 18
sociocultural groups. Psychology in the Schools, 48(10),
986–997.
Furrer, C., & Skinner, E. (2003). Sense of relatedness as a factor in
children’s academic engagement and performance. Journal of
Educational Psychology, 95(1), 148–162. doi:10.1037/0022-
0663.95.1.148.
Garcı́a Coll, C., Lamberty, G., Jenkins, R., McAdoo, H., Crnic, K.,
Wasik, B., & Vázquez Garcı́a, H. (1996). An integrative model
for the study of developmental competencies in minority
children. Child Development, 67(5), 1891–1914. doi:10.2307/
1131600.
Garcia, S. B., & Guerra, P. L. (2004). Deconstructing deficit thinking
working with educators to create more equitable learning
environments. Education and Urban Society, 36, 150–168.
Garcı́a-Reid, P., Reid, R. J., & Peterson, N. (2005). School
engagement among Latino youth in an urban middle school
context: Valuing the role of social support. Education and Urban
Society, 37(3), 257–275. doi:10.1177/0013124505275534.
Gay, G. (2002). Preparing for culturally responsive teaching. Journal
of Teacher Education, 53(2), 106–116. doi:10.1177/
0022487102053002003.
Gay, G. (2010). Culturally responsive teaching: Theory, research,
and practice. New York, NY: Teachers College Press.
Gregory, A., & Ripski, M. B. (2008). Adolescent trust in teachers:
Implications for behavior in the high school classroom. School
Psychology Review, 37(3), 337–353.
Gregory, A., Skiba, R. J., & Noguera, P. A. (2010). The achievement
gap and the discipline gap: Two sides of the same coin?
Educational Researcher, 39(1), 59–68. doi:10.3102/
0013189x09357621.
Gregory, A., & Weinstein, R. S. (2008). The discipline gap and
African Americans: Defiance or cooperation in the high school
classroom. Journal of School Psychology, 46(4), 455–475.
doi:10.1016/j.jsp.2007.09.001.
Griffiths, A.-J., Lilles, E., Furlong, M. J., & Sidhwa, J. (2012). The
relations of adolescent student engagement with troubling and
high-risk behaviors. In S. L. Christenson, A. L. Reschly & C.
Wylie (Eds.), Handbook of research on student engagement (pp.
563–584). New York, NY US: Springer. Retrieved from www.
springer.com.
Hamre, B. K., & Pianta, R. C. (2001). Early teacher–child relation-
ships and the trajectory of children’s school outcomes through
eighth grade. Child Development, 72, 625–638. doi:10.1111/
1467-8624.00301.
Hanson, T. L. & Kim, J. O. (2007). Measuring resilience and youth
development: The psychometric properties of the Healthy Kids
Survey. REL 2007-No. 34: Regional Educational Laboratory
West. Retrieved from http://ies.ed.gov/ncee/edlabs/regions/west/
pdf/REL_2007034_sum.pdf.
Haynes, N. M., Emmons, C. L., & Ben-Avie, M. (2001). The school
development program student, staff, and parent school climate
surveys. New Haven, CT: Yale Study Center.
Hershfeldt, P., Sechrest, R., Pell, K., Rosenberg, M., Bradshaw, C. P.,
& Leaf, P. J. (2009). Double-check: A framework of cultural
responsiveness applied to classroom behavior. Teaching Excep-
tional Children Plus, 6(9), 2–18.
Hughes, J., & Kwok, O. (2007). Influence of student–teacher and parent–
teacher relationships on lower achieving readers’ engagement and
achievement in the primary grades. Journal of Educational
Psychology, 99(1), 39–51. doi:10.1037/0022-0663.99.1.39.
Hughes, J. N., Luo, W., Kwok, O., & Loyd, L. K. (2008). Teacher–
student support, effortful engagement, and achievement: A
3-year longitudinal study. Journal of Educational Psychology,
100(1), 1–14. doi:10.1037/0022-0663.100.1.1.
Jennings, P. A., & Greenberg, M. T. (2009). The prosocial classroom:
Teacher social and emotional competence in relation to student
and classroom outcomes. Review of Educational Research, 79,
491–525.
Kang, Y., Gray, J. R., & Dovidio, J. F. (2014). The nondiscriminating
heart: Lovingkindness meditation training decreases implicit
intergroup bias. Journal of Experimental Psychology: Gen-
eral, 143(3), 1306.
KewalRamani, A., Gilbertson, L., Fox, M. A., Provasnik, S., &
National Center for Education Statistics, W. D. C. (2007). Status
and trends in the education of racial and ethnic minorities.
NCES 2007-039: National Center for Education Statistics.
Retrieved from http://nces.ed.gov/pubs2007/2007039.pdf.
Ladson-Billings, G. (2009). The dream-keepers: Successful teachers
of African American children (2nd ed.). San Francisco, CA:
Jossey-Bass.
Lowman, J. (1984). Mastering the techniques of teaching. San
Francisco, CA: Jossey-Bass.
Luke, D. A. (2004). Multilevel modeling. Thousand Oaks, CA: Sage.
Marsh, H. W., Balla, J. R., & McDonald, R. P. (1988). Goodness-of-
fit indexes in confirmatory factor analysis: The effect of sample
size. Psychological Bulletin, 103, 391–410. doi:10.1037//0033-
2909.103.3.391.
McKenzie, K. B., & Scheurich, J. J. (2004). Equity traps: A useful
construct for preparing principals to lead schools that are
successful with racially diverse students. Educational Adminis-
tration Quarterly, 40, 601–632.
Meehan, B. T., Hughes, J. N., & Cavell, T. A. (2003). Teacher–
student relationships as compensatory resources for aggressive
children. Child Development, 74(4), 1145–1157. doi:10.1111/
1467-8624.00598.
Meredith, W. (1993). Measurement invariance, factor analysis and
factorial invariance. Psychometrika, 58(4), 525–543. Retrieved
from http://link.springer.com/journal/11336.
MSDE. (2012). Professional staff by assignment, race/ethnicity and
gender Maryland public schools. Retrieved from http://www.
marylandpublicschools.org/NR/rdonlyres/68C794C3-C37B-
4116-85CD-B00E8FD8089B/34385/prostf14.pdf.
Nieminen, P., Lehtiniemi, H., Vähäkangas, K., Huusko, A., & Rautio,
A. (2013). Standardised regression coefficient as an effect size
index in summarizing findings in epidemiological studies.
Epidemiology, Biostatistics and Public Health, 10, 1–15.
doi:10.2427/8854.
1190 J Youth Adolescence (2016) 45:1176–1191
123
Organization for Economic Cooperation and Development (OECD).
(2008). Policy brief: Ten steps to equity in education. Retrieved
from http://www.oecd.org/edu/school/39989494.pdf.
Portes, A., & Rumbaut, R. G. (2005). Introduction: The second
generation and the children of immigrants longitudinal
study. Ethnic and Racial Studies, 28(6), 983–999.
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models
(2nd ed.). Thousand Oaks, CA: Sage.
Resnick, M. D., Bearman, P. S., Blum, R. W., Bauman, K. E., Harris,
K. M., Jones, J., & Udry, J. R. (1998). Protecting adolescents
from harm: Findings from the National Longitudinal Study of
Adolescent Health. In R. E. Muuss & H. D. Porton (Eds.),
Adolescent behavior and society: A book of readings (5th ed.,
pp. 376–395). New York, NY: McGraw-Hill.
Roeser, R. W., Eccles, J. S., & Sameroff, A. J. (2000). School as a
context of early adolescents’ academic and social-emotional
development: A summary of research findings. Elementary
School Journal, 100(5), 443–471. doi:10.1086/499650.
Sable, J., Plotts, C., & Mitchell, L. (2010). Characteristics of the 100
largest public elementary and secondary school districts in the
United States: 2008–09 (NCES 2011-301). U.S. Department of
Education, National Center for Education Statistics. Washington,
DC: U.S. Government Printing Office. Retrieved from http://
nces.ed.gov.
Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the
state of the art. Psychological Methods, 7(2), 147–177. doi:10.
1037/1082-989X.7.2.147.
Seaton, E. K., & Yip, T. (2009). School and neighborhood contexts,
perceptions of racial discrimination, and psychological well-
being among African American adolescents. Journal of Youth
and Adolescence, 38, 153–163. doi:10.1007/s10964-008-9356-x.
Shernoff, D. J., & Schmidt, J. A. (2008). Further evidence of an
engagement-achievement paradox among U.S. high school
students. Journal of Youth and Adolescence, 37(5), 564–580.
doi:10.1007/s10964-007-9241-z.
Simpson, E. H. (1949). The measurement of diversity. Nature, 163,
688. doi:10.1038/163688a0.
Skiba, R. J., Horner, R. H., Chung, C.-G., Rausch, M. K., May, S. L.,
& Tobin, T. (2011). Race is not neutral: A national investigation
of African American and Latino disproportionality in school
discipline. School Psychology Review, 40, 85–107. Retrieved
from http://www.nasponline.org/publications/spr.
Skinner, E. A., & Pitzer, J. R. (2012). Developmental dynamics of
student engagement, coping, and everyday resilience. In S.
L. Christenson, A. L. Reschly & C. Wylie (Eds.), Handbook of
research on student engagement (pp. 21–44). New York, NY:
Springer. Retrieved from www.springer.com.
Smalls, C., White, R., Chavous, T., & Sellers, R. (2007). Racial
ideological beliefs and racial discrimination experiences as
predictors of academic engagement among African American
adolescents. Journal of Black Psychology, 33(3), 299–330.
doi:10.1177/0095798407302541.
Steiger, J. H., & Lind, J. C. (1980). Statistically based tests for the
number of common factors. In Annual meeting of the psycho-
metric society. Iowa City, IA, Vol. 758.
Trent, S. C., Artiles, A. J., & Englert, C. S. (1998). From deficit
thinking to social constructivism: A review of theory, research,
and practice in special education. Review of Research in
Education, 23, 277–307. doi:10.2307/1167293.
Tyler, K. M., & Boelter, C. M. (2008). Linking black middle school
students’ perceptions of teachers’ expectations to academic
engagement and efficacy. Negro Educational Review, 59(1/2),
27–44.
U.S. Department of Education, National Center for Education
Statistics. (2010). Schools and Staffing Survey (SASS), ‘‘Public
School Teacher and Private School Teacher Data Files,’’
2007–08. Retrieved from http://nces.ed.gov/programs/coe/analy
sis/2010-figures.asp.
Voight, A., Hanson, T., O’Malley, M., & Adekanye, L. (2015). The
racial school climate gap: Within-school disparities in students’
experiences of safety, support, and connectedness. American
Journal of Community Psychology, 56, 252–267. doi:10.1007/
s10464-015-9751-x.
Wong, C. A., Eccles, J. S., & Sameroff, A. (2003). The influence of
ethnic discrimination and ethnic identification on African
American adolescents’ school and socioemotional adjustment.
Journal of Personality, 71(6), 1197–1232. doi:10.1111/1467-
6494.7106012.
Zeiders, K. H., Umaña-Taylor, A. J., & Derlan, C. L. (2012).
Trajectories of depressive symptoms and self-esteem in Latino
youths: Examining the role of gender and perceived discrimi-
nation. Developmental Psychology, 49, 951–963. doi:10.1037/
a0028866.
Zumwalt, K., & Craig, E. (2005). Teachers’ characteristics: Research
on the demographic profile. In M. Cochran-Smith & K.
M. Zeichner (Eds.), Studying teacher education: The report of
the AERA panel on research and teacher education (pp.
111–156). Mahwah, NJ: Lawrence Earlbaum Associates.
Jessika H. Bottiani is a Postdoctoral Fellow at the Curry School of Education at the University of Virginia. She holds a doctorate in
mental health and master’s from the Johns Hopkins Bloomberg
School of Public Health, and completed her training through an
Institute of Education Sciences pre-doctoral fellowship. Dr. Bottiani’s
research focuses on public health approaches to school-based violence
prevention and promotion of positive youth development, with
particular attention to racial disparities. She has provided professional
workshops, technical assistance, and consultation on topics related to
the discipline gap, culturally responsive classroom-management, and
equitable school practices.
Catherine P. Bradshaw is a Professor and the Associate Dean for Research and Faculty Development at the Curry School of Education
at the University of Virginia (UVA); she is also the Deputy Director
of the CDC-funded Johns Hopkins Center for the Prevention of Youth
Violence and Co-Director of the NIMH-funded Johns Hopkins Center
for Prevention and Early Intervention. She collaborates on research
projects examining bullying and school climate; the development of
aggressive and problem behaviors; and the design, evaluation, and
implementation of evidence-based prevention programs in schools.
She collaborates on federally supported randomized trials of school-
based prevention programs, including Positive Behavioral Interven-
tions and Supports (PBIS) and social-emotional learning curricula.
Tamar Mendelson is an Associate Professor at the Johns Hopkins Bloomberg School of Public Health in the Department of Mental
Health. Her research is aimed at developing, evaluating, and
disseminating prevention strategies to improve maternal and child
mental health, with a focus on underserved urban populations. She is
currently evaluating mindfulness-based and cognitive behavioral
approaches for improving emotional and behavioral outcomes among
urban students. She is also assessing mindfulness-based and cognitive
behavioral strategies to promote maternal mental health and prevent
postpartum depression in contexts that serve at-risk perinatal women.
J Youth Adolescence (2016) 45:1176–1191 1191
123
Journal of Youth & Adolescence is a copyright of Springer, 2016. All Rights Reserved.
- Inequality in Black and White High School Students’ Perceptions of School Support: An Examination of Race in Context
- Abstract
- Introduction
- School Support
- Equity as a Dimension of School Support
- Racial Disparities in School Support
- The Role of School Context
- The Present Study
- Methods
- Sample
- Procedure
- Measures
- Student Demographic Characteristics
- School Support
- School Racial and Ethnic Diversity
- School SES
- Analyses
- Measurement Model
- Measurement Invariance
- Missing Data
- Multilevel Analyses
- Post-hoc Multiple Group CFA
- Results
- Discussion
- Caring
- High Expectations
- Equity
- Limitations and Strengths
- Implications for Practice
- Conclusion
- Acknowledgments
- References
