[Paleopsych] Melissa Greene and Niobe Way: Self-Esteem Trajectories among Ethnic Minority Adolescents

[Premise Checker]

Self-Esteem Trajectories among Ethnic Minority Adolescents: A Growth Curve 
Analysis of the Patterns and Predictors of Change
Melissa L. Greene and Niobe Way
New York University
Journal of Research on Adolescence
Volume 15 Issue 2 Page 151  - June 2005
doi:10.1111/j.1532-7795.2005.00090.x

First, the summary from CHE 5.5.24):

A glance at the June issue of the Journal of Research on Adolescence: Race, 
gender, and self-esteem
While the self-esteem of white children, as a group, remains constant during 
their adolescence, it increases among black, Latino, and Asian-American youths 
from low-income families, and that difference is important to note because 
self-esteem has been shown to influence mental health, academic achievement, 
and social relationships, write Melissa L. Greene, a psychology instructor at 
Cornell University, and Niobe Way, an associate professor of applied psychology 
at New York University.
That was one of two major findings from their study of black, Latino, and 
Asian-American students from lower-income and working-class families at a 
public high school in New York City. The other was that "boys and girls 
experienced similar average trajectories" of self-esteem. The latter finding 
contradicts previous research, which has generally focused on white, 
middle-class children and has found different patterns of change in self-esteem 
between boys and girls.
Their study, Ms. Greene and Ms. Way say, also supports earlier findings that 
"patterns of gender difference are not identical across ethnic groups, and 
challenges the well-established assumption that boys experience higher 
self-esteem than girls through adolescence."
Their study, the authors say, suggests that cultural and contextual factors 
influence the development of self-esteem. They say that in future studies of 
adolescent psychological processes, researchers should take account of the 
varied backgrounds of subjects. In addition, they suggest, research should not 
settle for group averages, but should focus instead on individual variations.
In their study, Ms. Greene and Ms. Way also explored the effects of 
adolescents' perceptions of how family, peers, and their school environment 
affected increases in self-esteem. They say: "Each perceived context was 
significantly associated with self-esteem trajectories when examined 
independently, but family experiences emerged as most strongly related to 
changes in self-esteem."
---------------------

The current study presents a growth curve analysis of self-esteem among Black, 
Latino, and Asian American high school students. A series of hierarchical 
linear models were used to examine patterns and predictors of change in 
self-esteem over time. Results revealed an average increase in self-esteem with 
age. Although boys and girls experienced similar trajectories of self-esteem, 
ethnicity was a significant moderator of developmental change. Black 
adolescents reported higher self-esteem, while Asian American adolescents 
reported lower self-esteem, compared with their Latino peers. Latino 
adolescents experienced a sharper increase in self-esteem over time compared 
with Black adolescents. The unique and conjoint effects of adolescents' 
experiences with peers, family, and school were examined in relation to 
self-esteem trajectories. Results revealed that each perceived context was 
significantly associated with self-esteem trajectories when examined 
independently, but family experiences emerged as most strongly related to 
changes in self-esteem. Results underscore the need to examine change at the 
individual level, as well as the importance of studying the unique and conjoint 
effects of individual and contextual-level variables on developmental processes 
among ethnic minority adolescents.

The importance of self-esteem for the well-being of adolescents is underscored 
by decades of theory and research supporting its link with mental health, 
academic achievement, and social relationships (e.g., Hirsch & DuBois, 1991; 
Rosenberg, 1965; Zimmerman, Copeland Laurel, Shope, & Dielman, 1997 ). 
Furthermore, the numerous biological, physical, and cognitive changes that 
occur during adolescence highlight the importance of examining how self-esteem 
may change as well during this critical period of development. Although 
research suggests that self-esteem increases gradually during middle and late 
adolescence (e.g., McCarthy & Hoge, 1982), most studies have examined change at 
the group level, to the exclusion of the individual level (e.g., Marsh, 1989; 
McCarthy & Hoge, 1982 ). Thus, we have limited understanding of the individual 
variations in self-esteem trajectories during adolescence. Furthermore, 
existing research has focused primarily on self-esteem among White, 
middle-class adolescents (e.g., Baldwin & Hoffman, 2002 ), excluding the 
experiences of ethnically and socioeconomically diverse youth. In addition, 
while research has identified correlates of self-esteem, including family 
relationships (e.g., Coates, 1985; Luster & McAdoo, 1995; Openshaw, Thomas, & 
Rollins, 1984; Walker & Greene, 1986), peer relationships and/or friendships 
(e.g., Coates, 1985; Buhrmester & Yin, 1997), and school experiences (e.g., 
Hoge, Smit, & Hanson, 1990; Kuperminc, Leadbeater, Emmons, & Blatt, 1997 ), the 
combined influence of these factors, and the dynamic associations between 
changes in these experiences and changes in self-esteem have not been examined. 
Thus, responding to the gaps in the literature, the current study had three 
primary goals: (1) to describe the trajectories of change over time in 
self-esteem at the individual level among Black, Latino, and Asian American 
adolescents from low-income families; (2) to examine gender and ethnic 
differences in trajectories of self-esteem; (3) and to explore the unique and 
combined effects of adolescents' perceptions of family, peers, and school on 
change in self-esteem. This study is informed by ecological theory, which 
underscores the influence and interaction of multiple developmental contexts 
(Bronfenbrenner, 1979), as well as the importance of examining adolescents' 
subjective experiences of these contexts (Bronfenbrenner, 1979; Demo, Small, & 
Savin-Williams, 1987).

Self-Esteem: Definitions and Significance

Self-esteem is typically understood to be the evaluative aspect of the 
self-concept. A common definition is that of Rosenberg, who defined high 
self-esteem as "the feeling of being satisfied with oneself, believing that one 
is a person of worth" (1985, p. 210). Similarly, Harter defined self-esteem or 
self-worth as "the overall value that one places on the self as a person [...] 
" (1990, p. 67). These definitions reflect the notion of "global" or "general" 
self-esteem or self-worth. Global self-esteem has been linked with a variety of 
psychological and behavioral outcomes for children and adolescents, including 
anxiety (Bachman, 1970; Rosenberg, 1965); depression (Bachman, 1970; 
Overholser, Adams, Lehnert, & Brinkman, 1995; Rosenberg, 1965; Harter, Marold, 
& Whitesell, 1992); and suicidal ideation and behavior (Overholser, Adams, 
Lehnert, & Brinkman 1995), and is generally believed to be a good indicator of 
overall well-being (Harter, 1999; Rosenberg, 1986).

Trajectories of Change in Global Self-Esteem

Theoretical accounts of adolescent development (e.g., Erikson, 1968; Harter, 
1999; Rosenberg, 1985 ) predict that self-esteem may decline during the 
transition from childhood to early adolescence, as individuals struggle with 
their self-concept and identity in the face of the numerous physical, social, 
and cognitive changes that characterize this period (Feldman & Elliott, 1990 ). 
However, as adolescents consolidate their self-concept and begin to forge a 
sense of identity during middle and late adolescence, self-esteem should 
stabilize and/or increase. Short-term longitudinal research has supported a 
gradual increase in self-esteem during middle and late adolescence ( McCarthy & 
Hoge, 1982; Mullis, Mullis, & Normandin, 1992; O'Malley & Bachman, 1983; 
Wallace, Cunningham, & Del Monte, 1984).

Traditional research on adolescent self-esteem has focused almost exclusively 
on changes at the group, or average level, potentially ignoring individual 
variation in trajectories (Block & Robins, 1993; Hirsch & DuBois, 1991 ). 
Responding to this limitation, more recent research has used idiographic 
methods such as cluster analysis to examine trajectories of self-esteem change 
over time. These studies have supported the presence of distinct trajectories 
of self-esteem during adolescence; while some adolescents experience increases 
over time, others experience declines in self-esteem, or no change over time 
(Deihl, Vicary, & Deike, 1997; Hirsch & DuBois, 1991; Pahl, Greene, & Way, 
2000; Zimmerman et al., 1997).

While cluster analysis provides an idea of the patterns of change in a 
population, change at the individual level can be examined only through the use 
of growth curve analysis (see Rogosa & Willett, 1985; Willett, Singer & Martin, 
1998 ). Although recent research using these methods has explored change in 
domain specific areas of self-esteem or self-competence (e.g., Cole, Maxwell, 
Martin, Peeke, Seroczynski, Tram, Hoffman, Ruiz, Jacquez, & Maschman, 2001; 
Jacobs, Lanza, Osgood, Eccles, & Wigfield, 2002), to date, only a handful of 
studies have used growth curve analysis to examine changes in global 
self-esteem during adolescence. Scheier, Botvin, Griffin, and Diaz (2000) used 
growth curve analysis to examine change in self-esteem over a 4-year interval 
in a sample of White, middle-class middle-school students. Results revealed an 
average decrease in self-esteem with age, and significant variability in 
trajectories. Recently, Baldwin and Hoffman (2002) explored individual change 
in self-esteem in a sample of White, middle-class adolescents ranging in age 
from 11 to 22 years old. Results indicated a curvilinear effect of age on 
self-esteem. In addition, similar to the results of Scheier et al. (2000) , 
there was significant inter-individual variability in self-esteem trajectories. 
While these studies present a starting point for examining change at the 
individual level, they have focused exclusively on White, middle-class 
adolescents, leaving open the question of change in self-esteem among 
ethnically and socioeconomically diverse youth.

Individual-Level Moderators of Self-Esteem Trajectories

The importance of studying self-esteem trajectories among ethnic minority 
adolescents is underscored by the growing percentage of ethnic minority youth 
in the adolescent population (see McCloyd, 1998 ), as well as existing research 
indicating ethnic group differences in self-esteem. More specifically, Black 
adolescents often report higher self-esteem than their Latino and/or Asian 
American peers ( AAUW, 1991; Carlson, Uppal, & Prosser, 2000; Dukes & Martinez, 
1994; Martinez & Dukes, 1997; Pahl et al., 2000; Phinney, Cantu, & Kurtz, 1997; 
Rotheram-Borus, Dopkins, Sabate, & Lightfoot, 1996). High self-esteem among 
Black adolescents may in part result from strong ties to family and community 
that provide a sense of "group autonomy" (Chapman & Mullis, 2000 ), which may 
insulate these adolescents against the effects of racism found in the larger 
society, allowing for more positive self-esteem. Conversely, Asian American 
adolescents are often found to report the lowest self-esteem of these three 
groups (Crocker, Luhtanen, Blaine, & Broadnax, 1994; Dukes & Martinez, 1994; 
Martinez & Dukes, 1997; Way & Chen, 2000). Research indicates that Asian 
American adolescents struggle more than other ethnic minority youth with peer 
discrimination (Fisher, Wallace, & Fenton, 2000; Rosenbloom & Way, 2004), as 
well as poor perceptions of physical appearance and attractiveness (Root, 
1995). These experiences may contribute to the poor self-perceptions found 
among Asian American adolescents.

Although existing research has not examined longitudinally how trajectories of 
self-esteem differ among ethnic minority adolescents, cross-sectional research 
(AAUW, 1991), as well as research comparing trajectories among Black and White 
adolescents (Brown, McMahon, Biro, Crawford, Schreiber, Similo, Waclawiw, & 
Striegel-Moore, 1998 ), suggests that age-related differences in self-esteem 
may differ according to ethnic group. If Black adolescents report higher 
self-esteem than their peers, they may be less likely than their peers to 
indicate improvement in self-esteem with age. In other words, a "ceiling 
effect" may be evident among those who report high self-esteem. In addition, 
because of their relatively lower "starting point," Asian American adolescents 
may demonstrate the greatest age-related improvements in their self-esteem.

Similarly, gender may be an important variable that moderates self-esteem 
trajectories. Gender differences have been found consistently among White, 
middle-class samples, with boys typically reporting higher self-esteem than 
girls throughout early, middle, and late adolescence (Block & Robins, 1993; 
Kling, Hyde, Showers, & Buswell, 1999; Wilgenbusch & Merrell, 1999 ). 
Furthermore, studies that have examined change in self-esteem have found that 
these gender differences become more dramatic with age, as girls often 
experience a decline in self-esteem throughout adolescence, while their male 
peers experience improved self-esteem with age (Block & Robins, 1993; Harter, 
1993; Zimmerman et al., 1997).

Among ethnic minority adolescents, gender differences appear less clear-cut. 
For example, in a sample of Black, Latino, and White adolescents, Martinez and 
Dukes (1991) found that boys reported higher self-esteem than girls among 
Latino and White adolescents; however, among Black adolescents, the opposite 
was true, with girls reporting higher self-esteem than their male peers. 
Similarly, the declines in self-esteem found among White girls are not found 
consistently among girls from ethnic minority backgrounds. For example, while 
Hispanic and White girls experienced significant declines in self-esteem from 
elementary school through high school, Black girls continue to report high 
levels of self-esteem throughout adolescence (AAUW, 1991 ). Taken together, 
these studies suggest that the gender differences and developmental patterns 
found in White middle-class samples may not apply to adolescents of diverse 
backgrounds. The current study will examine the effects of gender and ethnicity 
on trajectories of self-esteem among Black, Latino, and Asian American 
adolescents.

Perceived Contexts and Self-Esteem

Symbolic interactionist theory (Cooley, 1902; Coopersmith, 1967; Mead, 1934; 
Rosenberg, 1979 ) asserts that the experience of the self is a function of 
perceptions of relationships with significant others. For example, the notion 
of the "looking glass self" (Cooley, 1902, p. 184 ) highlights the importance 
of perceptions and evaluations of significant others for an individual's 
self-concept and self-esteem. Family relationships, friendships, and school are 
the three primary relational contexts during adolescence, underscoring the 
importance of examining their contribution to self-esteem trajectories. 
Research has found positive associations between self-esteem and perceptions of 
family (Luster & McAdoo, 1995; Way & Robinson, 2003; Zimmerman & Maton, 1992), 
peer (Cauce, 1986; Coates, 1985) and school contexts (Fenzel, Magaletta, & 
Peyrot, 1997; Luster & McAdoo, 1995; Roeser & Eccles, 1998; Way & Robinson, 
2003 ) among ethnic minority and/or low-income adolescents. This body of 
research suggests that when adolescents perceive warmth, support, and safety in 
their family, peer, and school environments, they experience more positive 
self-perceptions as well.

However, our understanding of the contextual predictors of self-esteem has 
several significant limitations. First, longitudinal research has been limited 
and the existing research has often been short in duration (e.g., Way & 
Robinson, 2003 ), potentially obscuring any change that occurs gradually over 
time. Second, while peer and family relationships have been examined together 
(e.g., Coates, 1985; Zimmerman & Maton, 1992 ), rarely has the school context 
been included as well, precluding an understanding of the relative and combined 
effects of these three contextual-level variables on self-esteem. Third, these 
contexts are typically conceptualized as static entities, measured at only one 
point in time, rather than dynamic entities that themselves change over time 
(e.g., Way & Robinson, 2003). Finally, with a few exceptions (e.g., Way & 
Robinson, 2003 ), studies have focused on Black or African American 
adolescents, precluding an understanding of the ways in which the family, peer, 
and school contexts contribute to the self-esteem of Latino and Asian American 
adolescents.

The Current Study

The current study will investigate self-esteem trajectories in a sample of 
Black, Latino, and Asian American adolescents. A growth curve analysis using 
hierarchical linear modeling (Bryk & Raudenbush, 1987, 1992 ) will be conducted 
to examine trajectories of self-esteem, as well as individual (i.e., gender and 
ethnicity) and contextual (family, peers, and school) predictors of within- and 
between-person difference in growth trajectories. It is hypothesized that 
self-esteem will demonstrate a linear increase with age. It is predicted that 
Black adolescents will report higher self-esteem than their peers, as well as a 
flatter increase over time, while Asian Americans will experience the lowest 
levels of self-esteem, but the sharpest increases over time. No predictions are 
made regarding gender differences in self-esteem because of the inconsistency 
of gender differences in samples of ethnic minority youth. In addition, it is 
predicted that family support, friendship support, and school climate will be 
positively related to changes in self-esteem over time, and will also be 
related to individual differences in growth in self-esteem. Interactions 
between gender and ethnicity and family support, friendship support, and school 
climate will be explored, but no specific predictions are made because of the 
lack of research in this area.
  METHOD Go to: 
<http://www.blackwell-synergy.com/templates/jsp/_synergy/images/10x10_spacer.gif> 
[GO] [down] 
<http://www.blackwell-synergy.com/templates/jsp/_synergy/images/full_article_seperator2_right.gif>

Participants

Participants in the current study are part of a larger longitudinal study of 
adolescent development. All participants were students at a public high school 
on the Lower East Side of New York City. The majority of students in this 
school come from lower and working class families, with 90% of the student 
population eligible for federal assistance through the free lunch program. The 
data were collected in the fall semesters of 1996 (Time 1), 1997 (Time 2), 1998 
(Time 3), 1999 (Time 4), and 2000 (Time 5). Upon entry to the study, all 
participants were in the ninth or tenth grade. Eighty-five percent of students 
who were informed of the study in 1996 agreed to participate, and greater than 
95% of those who remained in the school each of the four additional years of 
the study chose to continue to participate.

The current study will focus only on the 205 students (53% female) who had 
complete data for a minimum of two (27%), and a maximum of five (26%) time 
points, with the majority having complete data for three or more time points 
(73%). Independent samples t-tests were used to explore any possible 
differences between students who had complete data at all five time points and 
those who did not. No significant differences were found on any predictor or 
outcome variables. Participants were racially and ethnically diverse, including 
24% Black [predominantly African American (68%) and West Indian (25%)], 48% 
Latino [predominantly Dominican (39%) and Puerto Rican (55%)], 18% Asian 
American [predominantly Chinese American (96%)], and 10% bi- or multi-racial 
[predominantly Black/Latino]. The racial/ethnic breakdown of the sample was 
comparable with the racial/ethnic breakdown of the school from which the sample 
was drawn. Eighty percent of participants were born in the United States. 
Participants tended to come from single-parent homes (73%), and have mothers 
(74%) and fathers (72%) who were not educated beyond high school.

Procedure

Students were recruited for participation through their mainstream English 
classes. Informed consent was obtained from both parents and students and 
translated into Spanish and Chinese to accommodate non-English speaking 
parents. Questionnaires were administered to all students who returned a signed 
consent form at each time point. Questionnaires were administered by members of 
a racially diverse research team and distributed during English classes or 
lunch periods. Students who had finished high school by the fifth wave of data 
collection were contacted by postcards, and completed questionnaires in a 
university research lab. Students were paid $5.00 in return for completion of 
their questionnaires at Time 1, and $10.00 at Times 2 through 5.

Measures

Self-Esteem. Self-esteem was assessed at Times 1 through 5 with the Rosenberg 
Self-Esteem Scale (RSE; Rosenberg, 1965 ), a ten-item measure of global 
self-esteem. Participants are asked to respond on a five-point Likert scale 
ranging from "strongly disagree" to "strongly agree." For the purpose of the 
present analysis, a mean score was calculated as a measure of global 
self-esteem for each participant at each time point. The RSE was developed for 
use with high-school students, and its validity and reliability have been 
demonstrated repeatedly (e.g., Hagborg, 1993; Rosenberg, 1965; Scheier et al., 
2000). Reliability was excellent in the current sample ( [alpha] >.80 at each 
time point).

Family Support. The Perceived Social Support for Family Scale (Procidano & 
Heller, 1983 ) was administered to assess perceptions of support received from 
family members at Times 1 through 5. Students were asked to respond "yes,""no," 
or "don't know" to 20 items concerning their experiences with their families. 
For the purposes of the present analysis, the total number of positive 
responses was used as a summary score of perceived family support. The authors 
report good reliability and construct validity for this measure (Procidano & 
Heller, 1983 ), and this measure has been determined to be reliable and valid 
when used in urban samples of ethnically and racially diverse adolescents 
(e.g., Tardy, 1985; Way & Leadbeater, 1999). Reliability was excellent in the 
current sample ( [alpha] >.80 at each time point).

Friendship Support. The Perceived Social Support for Friends Scale (Procidano & 
Heller, 1983 ) was administered to assess perceptions of perceived support 
received from friends at Times 1 through 5. Students were asked to respond 
"yes,""no" or "don't know" to 20 items concerning their experiences with their 
friends. For the purposes of the present analysis, the total number of positive 
responses was used as a summary score of perceived friendship support. The 
authors report good reliability and construct validity for this measure 
(Procidano & Heller, 1983 ). Additionally, this measure has been determined to 
be reliable and valid when used in urban samples of ethnically and racially 
diverse adolescents (e.g., Tardy, 1985; Way & Leadbeater, 1999). Reliability 
was excellent in the current sample ( [alpha] >.80 at each time point).

Perceived School Climate. Perceived school climate was measured at Times 1 
through 5 with a modified version of the School Climate Scale (Haynes, Emmons, 
& Comer, 1993 ). This version contains 33 items tapping three dimensions of 
school climate: student/student relations, teacher/student relations, and order 
and discipline. Students indicate their agreement with the items using a 
five-point Likert scale ranging from "strongly agree" to "strongly disagree." 
For the present analyses, a mean score was calculated for each participant as a 
measure of perceived school climate at each time point. This measure has 
demonstrated good reliability and validity previous samples (Haynes et al., 
1993) and reliability was excellent in the current sample ( [alpha] >.80 at 
each time point). For those students who had finished high school by the fifth 
wave of data collection, their reports at Time 5 of school climate were 
retrospective. However, there were no significant differences in reports of 
perceived school climate at Time 5 between those who remained in school and 
those who had graduated.

Data Analytic Method

In order to examine within- and between-person change in self-esteem, a growth 
curve analysis (see Rogosa & Willett, 1985; Willett et al., 1998) was conducted 
using Hierarchical Linear Models (HLMs) (Bryk & Raudenbush, 1987, 1992 ). A 
growth curve analysis using HLM confers numerous advantages over more 
traditional methods of investigating developmental change, such as repeated 
measures analysis of variance (see Bryk & Raudenbush, 1987, 1992). HLM uses 
empirical Bayes estimation (Strenio, Weisburg, & Bryk, 1983 ) to derive the 
final growth estimates for each participant, drawing on information at the 
within-person (i.e., level-1) and between-person (i.e., level-2) levels of 
analysis. As a result, HLM affords more precise estimates of individual growth 
over time and greater power to detect predictors of individual differences in 
change even with relatively small samples (Bryk & Raudenbush, 1987 ). This 
process also allows HLM to handle missing data, using any available data points 
to fit a growth trajectory for each participant. Finally, HLM can include 
covariates that themselves change over time. This is crucial to the current 
study, as we are interested in the dynamic relationships between changes in 
family support, friendship support, and school climate and self-esteem.
  RESULTS Go to: 
<http://www.blackwell-synergy.com/templates/jsp/_synergy/images/10x10_spacer.gif> 
[GO] [up] [down] 
<http://www.blackwell-synergy.com/templates/jsp/_synergy/images/full_article_seperator2_right.gif>

Preliminary Analyses

Because of research suggesting "within group" differences in psychological 
variables among subgroups of ethnic minority adolescents (e.g., Erkut & Tracy, 
2002; Rumbaut, 1994 ), we examined possible differences between African 
American and West Indian adolescents, as well as between Puerto Rican and 
Dominican adolescents. Independent samples t-tests were conducted to explore 
possible differences in levels of self-esteem, family support, friendship 
support, and school climate. No significant differences were found. Thus, for 
all subsequent analyses, African American and West Indian participants were 
grouped into a category labeled "Black," while Dominican and Puerto Rican 
participants were grouped together into the "Latino" category. Differences 
among Asian American participants were not explored because this subsample was 
predominantly Chinese American. Thus, for all growth models, three dummy 
variables, Black, Asian, and Bi/Multiracial were constructed to represent the 
four ethnic groups represented in the study. The Latino group was selected as 
the reference group because of previous research indicating that the greatest 
differences in self-esteem are found between Latino and non-Latino students 
(Way & Robinson, 2003).

Because participants were of different ages upon entry into the study, and 
participated in different numbers of waves of data collection, a variable (Mean 
Age) was computed to represent each participant's average age during the years 
in which he or she participated. This variable provides a check for possible 
cohort effects, as well as an additional check for attrition effects. Mean Age 
was included as a level-2 predictor in preliminary growth models, and we 
compared the fit of the full model (the model including Mean Age) to a reduced 
model. Mean Age was not a significant predictor of either the intercept or 
growth rate, and did not improve the fit of the model. Thus, we did not include 
Mean Age as a control variable in subsequent analyses.

Descriptive Analyses

Descriptive analyses were conducted to examine the distribution of the 
predictor and outcome variables for each of the five time points. Means and 
standard deviations can be seen in Table 1. Mean changes in self-esteem as a 
function of gender and ethnicity can be seen in Figures 1 and 2, respectively.

Modeling Individual Growth in Self-Esteem

Unconditional Growth Models. Unconditional growth models were constructed in 
order to examine average growth in the population, as well as the 
between-person variability in growth. Results of the linear unconditional model 
can be seen in Table 2 . The intercept of 4.02 indicates the mean level of 
self-esteem for the sample at age 16, while the significant positive slope 
coefficient indicates that self-esteem increased by .08 units per year of age 
on average. Examination of the random effects revealed significant 
heterogeneity in the intercept and slope coefficients. This result indicates 
that these coefficients should be allowed to vary at level-2, and predictors of 
inter-individual differences should be explored.

A quadratic age term ((age [-] 16)2) was added to the unconditional model in 
order to explore any possible acceleration in the growth of self-esteem (see 
Table 3). Neither the fixed nor the random effect was significant, although 
there was a trend towards significance (p<.10) for the random effect, 
suggesting some degree of inter-individual variability in the curvature of 
growth. A [chi] 2 test revealed that the addition of the quadratic term did not 
significantly improve the fit of the model. Thus, a linear model with random 
effects for both the intercept and slope was selected as best fitting growth in 
self-esteem in the population.

Demographic Predictors of Self-Esteem Trajectories. Dummy codes representing 
gender and ethnicity were entered into the level-2 equations in order to 
explore possible moderating effects of these variables on the growth 
parameters. Results of this model can be seen in Table 4 . Examination of the 
fixed effects revealed significant effects of ethnicity on the intercept. Black 
adolescents reported higher self-esteem at age 16, while Asian American 
adolescents reported lower self-esteem at age 16, compared with Latino 
adolescents. In the equation predicting the linear growth rate, there was also 
a significant effect of ethnicity. Latino adolescents experienced a steeper 
increase in their self-esteem change as compared with Black adolescents. Fitted 
growth curves as a function of ethnicity can be seen in Figure 3 . Gender was 
not a significant predictor of inter-individual differences in either the 
intercept or the slope, although there was a trend towards significance in the 
equation predicting the intercept. In this equation, the positive coefficient 
representing gender suggests that boys experienced higher self-esteem than did 
girls at age 16. However, boys and girls experienced similar rates of change in 
self-esteem as a function of age. In a second model, a series of terms 
representing the interaction of gender and ethnicity (Male [x] Black, Male [x] 
Asian American, Male [x] Bi/Multiracial) were entered the level-2 equations. 
There were no significant effects on either the intercept or the slope for any 
of the interaction terms, indicating that the ethnic differences in self-esteem 
trajectories were similar for boys and girls.

Unique Effects of Family Support, Friendship Support, and School Climate on 
Self-Esteem Trajectories. In order to examine the unique effects of the family, 
peer, and school contexts, a series of models was constructed to examine each 
context separately. Time-varying covariates were added to the level-1 equations 
in order to examine the associations between changes in the covariates and 
changes in self-esteem (Bryk & Raudenbush, 1992 ). All time-varying covariates 
were centered around each participant's unique mean averaged over time (i.e., 
group mean centering), when entered into the level-1 equations. Although these 
variables were measured across time, their coefficients were fixed (i.e., not 
allowed to vary randomly). This was done because growth curve models can only 
accurately estimate the numbers of random coefficients equal to half of the 
number of waves of data (Bryk & Raudenbush, 1992). Thus, these coefficients can 
be thought to be "varying" rather than "random" (Kreft & De Leeuw, 1998 ). In 
addition, a variable representing each participant's mean for each of the 
time-varying covariates (Mean Family Support, Mean Friendship Support, Mean 
School Climate) was entered into the level-2 equations. These variables were 
centered around the mean for the sample (i.e., grand mean centering) when 
entered into the level-2 equations. The inclusion of the mean of the 
time-varying covariates at level-2 controls for the possibility that the 
coefficients estimated at level-1 are the result of stable between-person 
differences (Bryk & Raudenbush, 1992 ). In addition, the inclusion of these 
variables can answer questions regarding the ability of these covariates to 
predict inter-individual differences in self-esteem trajectories.

Results revealed that within-person changes in self-esteem were significantly 
associated with changes in family support ( [gamma] =.02, p<.001) and 
friendship support ( [gamma] =.01, p <.05), indicating that increases in 
perceived family support and friendship support were associated with 
improvements in self-esteem. At level-2, individual differences in self-esteem 
at age 16 were predicted by mean family support ( [gamma] =.06, p<.001), and 
mean friendship support ( [gamma] =.07, p <.001). These results indicate that 
adolescents who reported higher family support and higher friendship support, 
on average, also reported higher self-esteem at age 16. A trend towards 
significance was found for mean school climate ( [gamma] =.14, p <.10), 
suggesting that adolescents who reported more positive perceptions of school 
climate on average, also reported higher self-esteem. Only mean family support 
was predictive of individual differences in the linear slope ( [gamma] = [-] 
.01, p <.05), indicating that adolescents who reported lower average levels of 
family support experienced a steeper increase in self-esteem over time compared 
with adolescents who reported higher average levels of family support.

Combined Effects of Family Support, Friendship Support, and School Climate on 
Self-Esteem Trajectories. Following the unique models, a combined model was 
constructed containing the individual- and contextual-level variables that had 
reached significance in the earlier models. Family support and friendship 
support were included as time-varying covariates at level-1 in order to examine 
the combined impact of these contexts on intra-individual change over time in 
self-esteem. At level-2, mean family support, friendship support, and perceived 
school climate were included as predictors of inter-individual differences in 
the intercept, while mean family support was also included in the equation 
predicting the linear slope. Gender and ethnicity were also included in the 
equation predicting the intercept, and ethnicity was included in the equation 
predicting the linear slope because these variables had been significant (or 
had approached significance) in the demographic model.

Results of this model can be seen in Table 4 . Family support continued to be 
significantly associated with intra-individual growth in self-esteem, 
indicating that increases in family support were associated with increases in 
self-esteem over time. Changes in friendship support were no longer 
significantly associated with intra-individual change in self-esteem once 
family support was included in the model. At level-2, ethnicity remained a 
significant predictor of the intercept and linear slope. In addition, both mean 
family support and friendship support were significant positive predictors of 
the intercept, while mean school climate demonstrated a trend towards 
significance. These results indicate that adolescents who reported higher 
levels of family and friendship support, and more positive perceptions of their 
school climate, on average, also reported higher levels of self-esteem at age 
16. After including ethnicity in the model, mean family support was no longer a 
significant predictor of the linear slope.

The final model examined possible interactions between gender and ethnicity and 
family and friendship support on within-person changes in self-esteem. Results 
of this model can be seen in the second column of Table 4 . While the effects 
of family support were robust across gender and ethnicity, there was an 
interaction between ethnicity and friendship support, with Black adolescents 
reporting a significantly stronger association between changes in friendship 
support and changes in self-esteem as compared with their Latino peers (see 
Figure 4).
  DISCUSSION Go to: 
<http://www.blackwell-synergy.com/templates/jsp/_synergy/images/10x10_spacer.gif> 
[GO] [up] [down] 
<http://www.blackwell-synergy.com/templates/jsp/_synergy/images/full_article_seperator2_right.gif>

The current study provides the first application of growth curve modeling to 
the examination of self-esteem among ethnic minority adolescents from 
low-income families. Based on theory (e.g., Harter, 1999) and previous research 
examining developmental change in self-esteem (e.g., Baldwin & Hoffman, 2002; 
Marsh, 1989 ), we predicted that adolescents in the current study would report 
improvements in self-esteem with age. This hypothesis was confirmed, as it 
appears that adolescents experience increasingly more positive self-perceptions 
with age. In addition, we predicted and found that adolescents differed 
significantly in both levels of self-esteem and the rate of change over time. 
This result is consistent with previous research using cluster analysis to 
examine self-esteem trajectories (Diehl et al., 1997; Hirsch & DuBois, 1991; 
Pahl et al., 2000; Zimmerman et al., 1997) and supports the necessity of using 
person-centered and idiographic statistical methods to examine developmental 
processes.

Although our results support earlier investigations of self-esteem change among 
White, middle-class adolescents (Baldwin & Hoffman, 2002; Marsh, 1989 ), one 
striking difference is the lack of significant gender differences in the 
current study. Previous research with White, middle-class adolescents has 
consistently found that boys and girls experience different patterns of change 
in self-esteem over time (e.g., Baldwin & Hoffman, 2002; Block & Robins, 1993; 
Harter, 1993; Zimmerman et al., 1997 ). In the current study, boys and girls 
experienced similar average trajectories. This result supports and extends 
earlier research that suggests that patterns of gender differences are not 
identical across ethnic groups (e.g., Martinez & Dukes, 1991), and challenges 
the well-established assumption that boys experience higher self-esteem than 
girls throughout adolescence.

The lack of gender differences in the current study, as well as in other 
samples of ethnic minority youth, suggests the influence of cultural and 
contextual factors in the development of self-esteem and underscores the need 
to examine mediators of these differences. Recent research has suggested that 
stable and/or higher self-esteem found among Black girls may be accounted for 
by their higher levels of perceived attractiveness and confidence in 
popularity, and lower levels of concern about weight and social 
self-consciousness relative to their White peers (Brown et al., 1998; Eccles, 
Barber, Jozefowicz Malenchuk, & Vida, 1999 ). Thus, the risk factors that may 
contribute to declines in self-esteem for White, middle-class girls may not be 
as salient and/or as intimately tied to self-esteem for Black girls. However, 
these factors have not been adequately studied among Latina and Asian American 
girls. Thus, future research should examine possible mediating variables in an 
ethnically diverse sample including Black, Latina, Asian American, and White 
adolescents.

While boys and girls experienced similar average self-esteem trajectories, 
exploration of ethnic differences in the current sample revealed significant 
differences between groups. Consistent with our predictions and previous 
research (e.g., Crocker et al., 1994; Dukes & Martinez, 1994; Pahl et al., 
2000; Phinney et al., 1997 ), Black adolescents reported higher self-esteem, 
and Asian American adolescents reported lower self-esteem than their Latino 
peers. In addition, while an increase in self-esteem with age was found for the 
sample as a whole, as predicted, there were significant ethnic differences in 
the rate of change over time. The self-esteem of Black adolescents showed a 
flatter increase over time compared with that of Latino adolescents. This may 
be because of the relatively higher self-esteem of Black youth at the early 
ages of the study, and may therefore indicate a "ceiling" for self-esteem. 
Because the self-esteem of Latino adolescents increased at a greater rate over 
time, by the end of the study, Latino youth had achieved levels of self-esteem 
comparable with their Black peers.

Unlike their Latino peers who "caught up" with their Black peers, Asian 
American adolescents continued to experience dramatically lower self-esteem 
even during the years of late adolescence. This finding is consistent with the 
small body of research that has examined self-esteem among Asian American 
adolescents (Crocker et al., 1994; Dukes & Martinez, 1994; Martinez & Dukes, 
1997 ), as well as qualitative data that indicates that Asian American 
adolescents' self-perceptions are less positive than those of their Latino and 
Black peers (Way & Pahl, 1999 ). The reasons for this finding may include the 
fact that Asian American students often report more peer discrimination than 
Latino or Black adolescents (Rosenbloom & Way, 2004). Recent research has 
linked peer discrimination with poor self-esteem among ethnic minority youth 
(Fisher et al., 2000 ). This finding may also reflect the lower priority within 
the Asian American community on enhancing self-esteem among youth than within 
the Black or Latino communities (Sodowsky, Kwan, & Pannu, 1995 ). Understanding 
why Asian American students are consistently reporting significantly lower 
self-esteem than their peers is an important next step in self-esteem research.

Alternatively, as predicted, Black adolescents were found to report higher 
self-esteem compared with their peers. It has been suggested that strong social 
support may be a factor in the relatively high self-esteem of Black 
adolescents. Research examining the use of social support has demonstrated its 
importance as a coping strategy for Black adolescents (Chapman & Mullis, 2000; 
Maton, Teti, Corns, Vieira-Baker, & Lavine, 1996; McCreary, Slavin & Berry, 
1996 ). Thus, Black adolescents may effectively use their social networks to 
cope with stressors that might otherwise affect self-esteem (e.g., Wilson, 1989 
). The strong link between social relationships and self-esteem for Black 
adolescents is supported by our finding that friendship support had a greater 
impact on changes in self-esteem among Black adolescents compared with their 
Latino peers. Previous analyses with the current data set has found self-esteem 
and friendship quality to be strongly linked among Black adolescents 
(Rosenbaum, 2000 ). Taken together, these results suggest that peer 
relationships may play a particularly important role in Black adolescents' 
sense of self. Qualitative research could aid in our understanding of the 
importance of peers relationships for the self-esteem and psychological 
well-being of Black youth.

The final goal of the current study was to explore the unique and combined 
effects of the family, peer, and school contexts on self-esteem trajectories. 
Results revealed that when adolescents experienced increased support from 
family and friends, evaluations of the self were more positive as well. In 
addition, adolescents who experienced better family and peer relationships, and 
more positive perceptions of their school, also reported higher self-esteem, 
while adolescents who reported lower levels of family support, reported greater 
increases in self-esteem over time. Additional analyses suggested that higher 
levels of family support were associated with higher levels of self-esteem, and 
therefore, a smaller increase in self-esteem over time. These results provide 
strong support for the importance of the social context for adolescents' 
self-perceptions (Cooley, 1902), and suggest that positive family, peer, and 
school environments are important contributors to psychological well-being.

When multiple contexts were examined together, the quality of family 
relationships emerged as most consistently and strongly related to self-esteem 
trajectories. This result supports previous research that has found family or 
parent relationships to be more closely tied to self-esteem and/or 
psychological well-being than peer relationships (e.g., Armsden & Greenberg, 
1987; McCreary et al., 1996; Paterson, Pryor, & Field, 1995; Way & Robinson, 
2003; Zimmerman & Maton, 1992), and highlights the ongoing importance of 
high-quality family relationships for psychological well-being during 
adolescence. The current results suggest that parents are a primary presence in 
their children's emotional lives throughout adolescence. The relatively weak 
association between school climate and self-esteem was surprising, as other 
longitudinal studies have found strong associations between school climate and 
self-esteem (Hoge et al., 1990; Way & Robinson, 2003 ). However, reports of 
perceived school climate at Time 5 were retrospective for students who had 
graduated by the fifth wave of data collection, which may have weakened the 
association between school climate and self-esteem.

While the current study adds to our existing understanding of self-esteem 
processes, and particularly, the dynamics of self-esteem among ethnic minority, 
low-income adolescents, there are several areas for further exploration. First, 
current advances in statistical methods and growth curve analysis will provide 
opportunities for greater understanding of the dynamic relationships between 
the family, peer, and school contexts, and self-esteem. For example, research 
should explore cross-domain growth models (see Sayer & Willett, 1998; Willett & 
Sayer, 1996 ), allowing for the simultaneous modeling of growth in multiple 
domains. Second, reliance on questionnaire data alone limits our understanding 
of adolescents' perceptions of themselves and their environment. Use of 
integrative methods (e.g., qualitative and quantitative data) would help to 
provide a greater understanding of the meaning of self-esteem to adolescents, 
as well as the meanings that adolescents make of the important contexts of 
their lives and how they themselves see these contexts influencing their 
feelings about themselves and their psychological adjustment. An additional 
limitation is the fact that the Time 5 school climate measure assessed both 
retrospective (i.e., for those who had already graduated) and current (i.e., 
for those who had not graduated yet) school experiences. Although additional 
analysis of the data using reports of school climate from Times 1 to 4 
suggested similar findings as those with all waves of data, the inclusion of a 
variable that includes retrospective accounts may have weakened the association 
between school climate and self-esteem. Future research should include 
additional indicators of the quality of peer relationships in order to provide 
support for the current findings.

In sum, the current results indicate that Black, Latino, and Asian American 
youth from low-income families experience an increase in self-esteem throughout 
the course of adolescence. However, our results highlight important individual 
differences in self-esteem trajectories. Our findings that trajectories 
differed as a function of ethnicity, but not gender, underscore the necessity 
of exploring psychological processes in samples that include adolescents from a 
range of backgrounds using methods that are sensitive to individual variation. 
Some of our most common assumptions about adolescent development (e.g., gender 
differences in self-esteem) will continue to be challenged as we expand our 
understanding of adolescents from diverse ethnic and socioeconomic backgrounds. 
Future research should continue to examine not only the patterns and predictors 
of mental health among diverse groups of adolescents, but also the ways in 
which families, peers, and schools can support the well-being of these young 
people. Only then will we have the tools necessary to help adolescents develop 
to their fullest and healthiest potentials.

  Acknowledgements Go to: 
<http://www.blackwell-synergy.com/templates/jsp/_synergy/images/10x10_spacer.gif> 
[GO] [up] [down] 
<http://www.blackwell-synergy.com/templates/jsp/_synergy/images/full_article_seperator2_right.gif>

This research is based on the Doctoral Dissertation of Melissa L. Greene in the 
Department of Psychology, New York University. The authors would like to thank 
Niall Bolger, Diane Hughes, Joanna Sattin, Esther Marron, and Benjamin Williams 
for their feedback and support. In addition, we would like to thank the 
National Science Foundation, The National Institute of Mental Health, and The 
William T. Grant Foundation for their generous support of this project.
  References Go to: 
<http://www.blackwell-synergy.com/templates/jsp/_synergy/images/10x10_spacer.gif> 
[GO] [up] 
<http://www.blackwell-synergy.com/templates/jsp/_synergy/images/full_article_seperator2_right.gif>
•   American Association of University Women (AAUW) (1991). Shortchanging 
girls, shortchanging America: Full data report. Washington, DC: American 
Association of University Women Educational Foundation.
•   Armsden, G. C., & Greenberg, M. T. (1987). The inventory of parent and peer 
attachment: Individual differences and their relationship to psychological 
well-being in adolescence. Journal of Youth and Adolescence, 16, 427 [-] 453.
[ISI Abstract]
•   Bachman, J. J. (1970). Youth in transition (Vol. 2). Ann Arbor: University 
of Michigan Institute for Social Research, Survey Research Center.
•   Baldwin, S. A., & Hoffman, J. P. (2002). The dynamics of self-esteem: A 
growth-curve analysis. Journal of Youth and Adolescence, 2, 101 [-] 113.
[CrossRef Abstract]
•   Block, J., & Robins, R. W. (1993). A longitudinal study of consistency and 
change in self-esteem from early adolescence to early adulthood. Child 
Development, 64, 909 [-] 923.
[ISI Abstract] [CSA Abstract]
•   Bronfenbrenner, U. (1979). The ecology of human development. Cambridge, MA: 
Harvard University Press.
•   Brown, K. M., McMahon, R. P., Biro, F. M., Crawford, P., Schreiber, G. B., 
Similo, S., Waclawiw, M., & Striegel-Moore, R. (1998). Changes in self-esteem 
in black and white girls between the ages of 9 and 14 years. Journal of 
Adolescent Health, 23, 7 [-] 19.
[CrossRef Abstract] [ISI Abstract] [CSA Abstract]
•   Bryk, A. S., & Raudenbush, S. W. (1987). Application of hierarchical linear 
models to assessing change. Psychological Bulletin, 101, 147 [-] 158.
[CrossRef Abstract] [ISI Abstract]
•   Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models. 
Newbury Park, CA: Sage.
•   Buhrmester, D., & Yin, J. (1997, March). A longitudinal study of friends' 
influence on adolescents' adjustment. Paper presented at the Meeting for the 
Society for Research on Child Development, Washington, D.C.
•   Carlson, C., Uppal, S., & Prosser, E. C. (2000). Ethnic differences in 
processes contributing to the self-esteem of early adolescent girls. Journal of 
Early Adolescence, 20, 44 [-] 67.
[ISI Abstract]
•   Cauce, A. M. (1986). Social networks and social competence: Exploring the 
effects of early adolescent friendships. American Journal of Community 
Psychology, 14, 607 [-] 628.
[CrossRef Abstract] [ISI Abstract] [CSA Abstract]
•   Chapman, P. L., & Mullis, R. L. (2000). Racial differences in adolescent 
coping and self-esteem. The Journal of Genetic Psychology, 161, 152 [-] 160.
[ISI Abstract] [CSA Abstract]
•   Coates, D. L. (1985). Relationships between self-concept measures and 
social network characteristics for Black adolescents. Journal of Early 
Adolescence, 5, 319 [-] 338.
•   Cole, D. A., Maxwell, S. E., Martin, J. M., Peeke, L. G., Seroczynski, A. 
D., Tram, J. M., Hoffman, K. B., Ruiz, M. D., Jacquez, F., & Maschman, T. 
(2001). The development of multiple domains of child and adolescent 
self-concept: A cohort sequential longitudinal design. Child Development, 72, 
1723 [-] 1746.
[Synergy Abstract] [ISI Abstract] [CSA Abstract]
•   Cooley, C. H. (1902). Human nature and the social order. New York: 
Scribner.
•   Coopersmith, S. (1967). The antecedents of self-esteem. San Francisco: W.H. 
Freeman.
•   Crocker, J., Luhtanen, R., Blaine, B., & Broadnax, S. (1994). Collective 
self-esteem and psychological well-being among White, Black, and Asian college 
students. Personality and Social Psychology Bulletin, 20, 503 [-] 513.
[ISI Abstract] [CSA Abstract]
•   Deihl, L. M., Vicary, J. R., & Deike, R. C. (1997). Longitudinal 
trajectories of self-esteem from early to middle adolescence and related 
psychosocial variables among rural adolescents. Journal of Research on 
Adolescence, 7, 393 [-] 411.
[ISI Abstract] [CSA Abstract]
•   Demo, D. H., Small, S. A., & Savin-Williams, R. C. (1987). Family relations 
and the self-esteem of adolescents and their parents. Journal of Marriage and 
the Family, 49, 705 [-] 715.
[ISI Abstract] [CSA Abstract]
•   Dukes, R. L., & Martinez, R. (1994). The impact of ethgender on self-esteem 
among adolescents. Adolescence, 29, 105 [-] 115.
[ISI Abstract] [CSA Abstract]
•   Eccles, J. S., Barber, B. L., Jozefowicz Malenchuk, O., & Vida, M. (1999). 
Self-evaluations of competence, task values, and self-esteem. In N. G. Johnson, 
M. C. Roberts, & J. Worell (Eds.), Beyond appearance: A new look at adolescent 
girls (pp. 53 [-] 83). Washington: American Psychological Association.
•   Erikson, E. H. (1968). Identity: Youth and crisis. New York: W.W. Norton & 
Company.
•   Erkut, S., & Tracy, A. J. (2002). Predicting adolescent self-esteem from 
participation in school sports among Latino subgroups. Hispanic Journal of 
Behavioral Sciences, 24, 409 [-] 429.
[ISI Abstract] [CSA Abstract]
•   Feldman, S. S., & Elliott, G. R. (1990). At the threshold: The developing 
adolescent. Cambridge, MA: Harvard University Press.
•   Fenzel, L. M., Magaletta, P. R., & Peyrot, M. F. (1997). The relationship 
of school strain to school functioning and self-worth among urban, African 
American early adolescents. Psychology in the Schools, 34, 279 [-] 288.
[CrossRef Abstract] [ISI Abstract]
•   Fisher, C. B., Wallace, S. A., & Fenton, R. E. (2000). Discrimination 
distress during adolescence. Journal of Youth and Adolescence, 29, 679 [-] 695.
[CrossRef Abstract] [ISI Abstract] [CSA Abstract]
•   Hagborg, W. J. (1993). The Rosenberg self-esteem scale and Harter's 
self-perception profile for adolescents: A concurrent validity study. 
Psychology in the Schools, 30, 132 [-] 136.
[ISI Abstract]
•   Harter, S. (1990). Causes, correlates, and the functional role of global 
self-worth: A life-span perspective. In R. J. Sternberg, & J. Kolligian (Eds.), 
Competence considered (pp. 67 [-] 98). New Haven, CT: Yale University Press.
•   Harter, S. (1993). Causes and consequences of low self-esteem in children 
and adolescents. In R. F. Baumeister (Ed.), Self-esteem: The puzzle of low 
self-regard (pp. 87 [-] 116). New York: Plenum.
•   Harter, S. (1999). The construction of the self. New York: Guilford.
•   Harter, S., Marold, D.B, & Whitesell, N. R. (1992). A model of psychosocial 
risk factors leading to suicidal ideation in young adolescents. Development and 
Psychopathology, 4, 167 [-] 188.
[ISI Abstract]
•   Haynes, N., Emmons, C., & Comer, J. P. (1993). Elementary and middle-school 
climate survey. Unpublished manuscript, Yale University Child Study Center.
•   Hirsch, B. J., & DuBois, D. L. (1991). Self-esteem in early adolescence: 
The identification and prediction of contrasting longitudinal trajectories. 
Journal of Youth and Adolescence, 20, 53 [-] 72.
[CrossRef Abstract] [ISI Abstract] [CSA Abstract]
•   Hoge, D. R., Smit, E. K., & Hanson, S. L. (1990). School experiences 
predicting changes in self-esteem of sixth- and seventh-grade students. Journal 
of Educational Psychology, 82, 117 [-] 127.
[CrossRef Abstract] [ISI Abstract]
•   Jacobs, J. E., Lanza, S., Osgood, D. W., Eccles, J. S., & Wigfield, A. 
(2002). Changes in children's self-competence and values: Gender and domain 
differences across grades one through twelve. Child Development, 73, 509 [-] 
527.
[Synergy Abstract] [ISI Abstract] [CSA Abstract]
•   Kling, K. C., Hyde, J. S., Showers, C. J., & Buswell, B. N. (1999). Gender 
differences in self-esteem: A meta-analysis. Psychological Bulletin, 125, 470 
[-] 500.
[CrossRef Abstract] [ISI Abstract] [CSA Abstract]
•   Kreft, I., & De Leeuw, J. (1998). Introducing multilevel models. Thousand 
Oaks, CA: Sage.
•   Kuperminc, G., Leadbeater, B. J., Emmons, C., & Blatt, S. J. (1997). 
Perceived school climate and problem behaviors in middle-school students: The 
protective function of a positive educational environment. Journal of Applied 
Developmental Science, 1, 76 [-] 88.
[CSA Abstract]
•   Luster, T., & McAdoo, H. P. (1995). Factors related to self-esteem among 
African American youths: A secondary analysis of the High/Scope Perry preschool 
data. Journal of Research on Adolescence, 5, 451 [-] 467.
[ISI Abstract] [CSA Abstract]
•   Marsh, H. W. (1989). Age and sex effects in multiple dimensions of 
self-concept: Preadolescence to early adulthood. Journal of Educational 
Psychology, 81, 417 [-] 430.
[CrossRef Abstract] [ISI Abstract]
•   Martinez, R. O., & Dukes, R. L. (1991). Ethnic and gender differences in 
self-esteem. Youth and Society, 22, 318 [-] 338.
[ISI Abstract]
•   Martinez, R. O., & Dukes, R. L. (1997). The effects of ethnic identity, 
ethnicity, and gender on adolescent well-being. Journal of Youth and 
Adolescence, 26, 503 [-] 516.
[CrossRef Abstract] [ISI Abstract] [CSA Abstract]
•   Maton, K. I., Teti, D. M., Corns, K. M., Vieira-Baker, C. C., & Lavine, J. 
R. (1996). Cultural specificity of support sources, correlates and contexts: 
Three studies of African-American and Caucasian youth. American Journal of 
Community Psychology, 24, 551 [-] 587.
[ISI Abstract] [CSA Abstract]
•   McCarthy, J. D., & Hoge, D. R. (1982). Analysis of age effects in 
longitudinal studies of adolescent self-esteem. Developmental Psychology, 18, 
372 [-] 379.
[CrossRef Abstract] [ISI Abstract]
•   McCloyd, V. C. (1998). Changing demographics in the American population: 
Implications for research on minority children and adolescents. In V. C. 
McLloyd & L. Steinberg (Ed.), Studying minority adolescents conceptual, 
methodological, and theoretical issues (pp. 3 [-] 20). Mahwah, NJ: Lawrence 
Erlbaum Associates.
•   McCreary, M. L., Slavin, L. A., & Berry, E. J. (1996). Predicting problem 
behavior and self-esteem among African American adolescents. Journal of 
Adolescent Research, 11, 216 [-] 234.
[ISI Abstract] [CSA Abstract]
•   Mead, G. H. (1934). Mind, self and society. Chicago: University of Chicago 
Press.
•   Mullis, A. K., Mullis, R. L., & Normandin, D. (1992). Cross-sectional and 
longitudinal comparisons of adolescent self-esteem. Adolescence, 27, 51 [-] 61.
[ISI Abstract]
•   O'Malley, P. M., & Bachman, J. G. (1983). Self-esteem: Change and stability 
between ages 13 and 23. Developmental Psychology, 19, 257 [-] 268.
[CrossRef Abstract]
•   Openshaw, D.K, Thomas, D. L., & Rollins, B. C. (1984). Parental influence 
of adolescent self-esteem. Journal of Early Adolescence, 4, 259 [-] 274.
[CSA Abstract]
•   Overholser, J. C., Adams, D. M., Lehnert, K. L., & Brinkman, D. C. (1995). 
Self-esteem deficits and suicidal tendencies among adolescents. Journal of the 
American Academy of Child and Adolescent Psychiatry, 34, 919 [-] 928.
[CrossRef Abstract] [ISI Abstract]
•   Pahl, K., Greene, M., & Way, N. (2000, March). Self-esteem trajectories 
among urban, low-income, ethnic minority high school students. Poster presented 
at the Meeting of the Society for Research on Adolescence, Chicago, IL.
•   Paterson, J., Pryor, H., & Field, J. (1995). Adolescent attachment to 
parents and friends in relation to aspects of self-esteem. Journal of Youth and 
Adolescence, 24, 365 [-] 375.
[CrossRef Abstract] [ISI Abstract] [CSA Abstract]
•   Phinney, J. S., Cantu, C. L., & Kurtz, D. A. (1997). Ethnic and American 
identity as predictors of self-esteem among African American, Latino, and White 
adolescents. Journal of Youth and Adolescence, 26, 165 [-] 185.
[CrossRef Abstract] [ISI Abstract] [CSA Abstract]
•   Procidano, M., & Heller, K. (1983). Measure of perceived social support 
from friends and family. American Journal of Community Psychology, 11, 1 [-] 
24.
[CrossRef Abstract] [ISI Abstract]
•   Roeser, R., & Eccles, J. (1998). Adolescents' perceptions of middle-school: 
Relation to longitudinal changes in academic and psychological adjustment. 
Journal of Research on Adolescence, 8, 123 [-] 158.
[ISI Abstract] [CSA Abstract]
•   Rogosa, D. R., & Willett, J. B. (1985). Understanding correlates of change 
by modeling individual differences in growth. Psychometrika, 50, 203 [-] 228.
[ISI Abstract]
•   Root, M. P. P. (1995). The psychology of Asian American women. In H. 
Landrine (Ed.), Bringing cultural diversity to feminist psychology: Theory, 
research, and practice (pp. 265 [-] 301). Washington, D.C.: American 
Psychological Association.
•   Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: 
Princeton University Press.
•   Rosenberg, M. (1979). Conceiving the self. New York: Basic Books.
•   Rosenberg, M. (1985). Self-concept and psychological well-being in 
adolescence. In R. L. Leahy (Ed.), The development of the self (pp. 205 [-] 
246). New York: Academic Press.
•   Rosenberg, M. (1986). Self-concept from middle childhood through 
adolescence. In J. Suls, & A. G. Greenwald (Eds.), Psychological perspectives 
on the self (pp. 107 [-] 136). Hillsdale, NJ: Lawrence Erlbaum Associates.
•   Rosenbaum, G. (2000). An investigation of the ecological factors associated 
with friendship quality in urban, low-income, racial and ethnic minority 
adolescents. Unpublished doctoral dissertation, New York University
•   Rosenbloom, S. R., & Way, N. (2004). Experiences of discrimination among 
African American, Asian American, and Latino adolescents in an urban high 
school. Youth and Society, 35, 420 [-] 451.
[CrossRef Abstract] [ISI Abstract] [CSA Abstract]
•   Rotheram-Borus, M. J., Dopkins, S., Sabate, N., & Lightfoot, M. (1996). 
Personal and ethnic identity, values and self-esteem among Black and Latino 
adolescent girls. In B. J. R. Leadbeater, & N. Way (Eds.), Urban girls: 
Resisting stereotypes, creating identities (pp. 35 [-] 52). New York: New York 
University Press.
•   Rumbaut, R. G. (1994). The crucible within: Ethnic identity, self-esteem, 
and segmented assimilation among children of immigrants. International 
Migration Review, 28, 748 [-] 794.
[ISI Abstract] [CSA Abstract]
•   Sayer, A. G., & Willett, J. B. (1998). A cross-domain model for growth in 
adolescent expectancies. Multivariate Behavioral Research, 33, 509 [-] 543.
[ISI Abstract]
•   Scheier, L. M., Botvin, G. J., Griffin, K. W., & Diaz, T. (2000). Dynamic 
growth models of self-esteem and adolescent alcohol use. Journal of Early 
Adolescence, 20, 178 [-] 209.
[ISI Abstract]
•   Sodowsky, G. R., Kwan, K. K., & Pannu, R. (1995). Ethnic identity of Asians 
in the United States. In J. G. Ponterotto, J. M. Casas, L. A. Suzuki, & C. A. 
Alexander (Eds.), Handbook of multicultural counseling (pp. 123 [-] 154). 
Thousand Oaks, CA: Sage.
•   Strenio, J. L. F., Weisberg, H. I., & Bryk, A. S. (1983). Empirical Bayes 
estimation of individual growth curves parameters and their relationship to 
covariates. Biometrics, 39, 71 [-] 86.
[ISI Abstract]
•   Tardy, C. H. (1985). Social support measures. American Journal of Community 
Psychology, 13, 187 [-] 202.
[CrossRef Abstract] [ISI Abstract]
•   Walker, L. S., & Greene, J. W. (1986). The social context of adolescent 
self-esteem. Journal of Youth and Adolescence, 15, 315 [-] 322.
[ISI Abstract]
•   Wallace, J. R., Cunningham, T. F., & Del Monte, V. (1984). Change and 
stability in self-esteem between late childhood and early adolescence. Journal 
of Early Adolescence, 4, 253 [-] 257.
•   Way, N., & Chen, L. (2000). The characteristics, quality, and correlates of 
friendships among African American, Latino, and Asian American Adolescents from 
low-income families. Journal of Adolescent Research, 15, 274 [-] 301.
[ISI Abstract] [CSA Abstract]
•   Way, N., & Leadbeater, B. J. (1999). Pathways toward educational 
achievement among African American and Puerto Rican adolescent mothers: 
Reconsidering the role of family support. Development and Psychopathology, 11, 
349 [-] 364.
[CrossRef Abstract] [ISI Abstract]
•   Way, N., & Pahl, K. (1999). Friendship patterns among urban adolescent 
boys: A qualitative account. In M. Kopala, & L. A. Suzuki (Eds.), Using 
qualitative methods in psychology (pp. 145 [-] 161). Thousand Oaks, CA: Sage.
•   Way, N., & Robinson, M. G. (2003). A longitudinal study of the effects of 
family, friends, and school experiences on the psychological adjustment of 
ethnic minority, low-SES adolescents. Journal of Adolescent Research, 18, 324 
[-] 326.
[CrossRef Abstract] [ISI Abstract] [CSA Abstract]
•   Wilgenbusch, T., & Merrell, K. W. (1999). Gender differences in 
self-concept among children and adolescents: A meta-analysis of 
multidimensional studies. School Psychology Quarterly, 14, 101 [-] 120.
[ISI Abstract]
•   Willett, J. B., & Sayer, A. G. (1996). Cross-domain analysis of change over 
time: Combining growth modeling and covariance structure analysis. In G. A. 
Marcoulides, & R. E. Schumacker (Eds.), Advanced structural equation modeling 
(pp. 125 [-] 157). Hillsdale: NJ: Erlbaum.
•   Willett, J. B., Singer, J. D., & Martin, N. C. (1998). The design and 
analysis of longitudinal studies of development and psychopathology in context: 
Statistical models and methodological recommendations. Development and 
Psychopathology, 10, 395 [-] 426.
[CrossRef Abstract] [ISI Abstract]
•   Wilson, M. N. (1989). Child development in the context of the Black 
extended family. American Psychologist, 44, 380 [-] 383.
[CrossRef Abstract] [ISI Abstract] [CSA Abstract]
•   Zimmerman, M. A., Copeland Laurel, A., Shope, J. T., & Dielman, T. E. 
(1997). A longitudinal study of self-esteem: Implications for adolescent 
development. Journal of Youth and Adolescence, 26, 117 [-] 141.
[CrossRef Abstract] [ISI Abstract] [CSA Abstract]
•   Zimmerman, M. A., & Maton, K. I. (1992, March). Self-esteem, social 
support, and life stress: A regression analysis among male African American 
adolescents. Paper presented at The Society for Research on Adolescence, 
Washington, D.C.

Correspondence
Requests for reprints should be sent to Melissa L. Greene, Department of 
Psychiatry, Weill Medical College of Cornell University, White Plains, New York 
10605. E-mail: mlg2004 at med.cornell.edu

Image Previews

TABLE 1
Predictor and Outcome Variable Means (and Standard Deviations)


TABLE 2
Unconditional Growth Models for Self-Esteem


TABLE 3
Demographic Predictors of Self-Esteem Trajectories


TABLE 4
Combined Effects of Family Support, Friendship Support, and School Climate on 
Trajectories of...


FIGURE 1
Mean changes in self-esteem as a function of age and gender.


FIGURE 2
Mean changes in self-esteem as a function of age and ethnicity.

FIGURE 3
Fitted growth curves for self-esteem as a function of ethnicity.

FIGURE 4
Intraindividual change in self-esteem as a function of friendship support and 
ethnicity.