/

University of Southern California Dissertations and Theses

/

The association between peer victimization and school functioning in Latino children

(USC Thesis Other)

The association between peer victimization and school functioning in Latino children

PDF

Transcript (if available)

Content THE ASSOCIATION BETWEEN PEER VICTIMIZATION

AND SCHOOL FUNCTIONING IN LATINO CHILDREN

by

Jonathan Masao Nakamoto

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PSYCHOLOGY)

August 2008

Copyright 2008 Jonathan Masao Nakamoto

ii

Table of Contents

List of Tables iii

Abstract iv

Body 1

References 42

iii

List of Tables

Table 1: Means, Standard Deviations, and Ranges for the Measures 19

Table 2: Bivariate Correlations among all of the Variables 21

Table 3: The Mediating Role of Self- and Peer-Reports of Academic 23
Engagement in the Association between Peer Victimization and
Grade Point Average (GPA)

Table 4: The Moderating Role of Friendships in the Association between 25-26
Peer Victimization and Self-Reports of Academic Engagement

Table 5: A Social Cognitive Map Co-occurrence Matrix for One 28
Classroom in the Study

iv

Abstract

This study utilized a multi-informant approach to investigate the concurrent
association between peer victimization and school functioning in a sample of 135
Latino children (55 boys; 80 girls) in the third, fourth, and fifth grades. The children
attended elementary schools in distressed urban neighborhoods. Victimization by
peers was associated with low grade point averages (GPA) and poor academic
engagement. The analyses showed academic engagement mediated the relation
between peer victimization and GPA. Moderator analyses indicated that the negative
association between peer victimization and academic engagement was exacerbated for
children with numerous friends in their classrooms. Additional moderator analyses
revealed that the negative association between victimization and engagement was
stronger for children with many aggressive friends. Overall, the results extend past
research by investigating mediators and moderators of the association between peer
victimization and school functioning in an understudied population.

1

The Association between Peer Victimization and
School Functioning in Latino Children
Introduction
Researchers have consistently found physical and verbal victimization by peers
to be associated with poor psychosocial adjustment in childhood and adolescence. In
particular, a growing number of studies have focused on the association between peer
victimization and school functioning (e.g., Buhs, Ladd, & Herald, 2006; Hoglund,
2007). Even though the number of Latino children in schools in the United States will
likely continue to increase (Fox, Connolly, & Snyder, 2005), there has not been much
research on the link between maltreatment by peers and academic outcomes in this
rapidly growing population. The deleterious consequences of peer victimization may
exacerbate the academic problems large numbers of Latinos already experience due to
the difficulties associated with learning English as a second language and living in low
socioeconomic status (SES) homes (Lutkus, Rampey, & Donahue, 2005). The overall
aim of the current study was to investigate the association between peer victimization
and school functioning in a sample of low-income Latino children attending
elementary schools in distressed urban neighborhoods.
The broad construct of school functioning includes children’s performance in
school as well as their academic engagement (Hoglund, 2007). In the school bullying
literature, researchers have assessed school performance using grade point averages
(GPA; Nishina, Juvonen, & Witkow, 2005), achievement test scores (Buhs et al.,
2006), and teacher ratings of academic functioning (Schwartz, 2000). Researchers

2

(Buhs, 2005; Hoglund, 2007) have also investigated the association between peer
victimization and measures assessing the three subtypes of academic engagement
outlined in Fredricks, Blumenfeld, and Paris’ (2004) review. Fredricks et al. (2004)
first highlighted behavioral engagement, which includes conforming to classroom
rules, completing homework on time, and abstaining from disruptive behaviors.
Emotional engagement relates to students’ affective responses to school, such as liking
or being interested in school. Finally, cognitive engagement is associated with
students’ psychological investment in their schoolwork and putting forth effort into
their studies.
The initial studies examining the link between peer victimization and school
performance were conducted with relatively homogenous samples of middle-class
children (e.g., Ladd, Kochenderfer, & Coleman, 1997). These studies showed that
higher levels of victimization were associated with lower levels of achievement. More
recently, researchers have found small but significant negative correlations between
measures of peer maltreatment and school performance with diverse samples of urban
youth (e.g., Schwartz, Gorman, Nakamoto, & Toblin, 2005). Consistent with studies
treating peer victimization as a continuous variable, researchers have found that
subgroups of victimized children have poorer academic functioning compared to their
non-victimized peers (Olweus, 1978; Schwartz, 2000).
As a counterpoint to studies showing negative associations between
victimization by peers and school functioning, prior ethnographic research suggested
that African American adolescents who do well in school can be stigmatized as

3

“acting white” and encounter more harassment than their normally achieving peers
(Fordham & Ogbu, 1986). It has been proposed that similar factors may operate in
low-income Latino communities (Ogbu & Matute-Bianchi, 1986 as cited in Fordham
& Ogbu, 1986). Empirical work, however, has not supported Fordham and Ogbu’s
conclusions. For instance, with a sample of rural African American youth, Estell,
Farmer, and Cairns (2007) found that victims of bullying had lower levels of academic
competence than their non-victimized counterparts. Graham et al. (2006) did not find
the negative association between peer victimization and school adjustment to differ
among Latino, African American, and Caucasian students attending middle schools in
urban settings. Furthermore, Schwartz’s (2000) sample, composed almost entirely of
low-income Latino and African American elementary school children, revealed a
negative association between peer victimization and academic functioning.
In addition to peer victimization’s association with poor school performance,
recent investigations have linked high levels of victimization to low academic
engagement (Buhs, 2005; Hoglund, 2007). Much research has highlighted the
importance of academic engagement as a strong predictor of school performance
(Fredricks et al., 2004). Moreover, academic engagement is a particularly important
variable given the present sample because past researchers have suggested that
increasing it could help improve the low academic achievement and high dropout rates
that are prevalent in urban schools (Fredricks, Blumenfeld, Friedel, & Paris, 2005).
My first goal in the current study was to estimate the strength of the relation
between peer victimization and school functioning in a sample of low-income Latino

4

children in late elementary school. The previously reviewed research overwhelmingly
supports my prediction that higher levels of peer victimization would be associated
with lower GPAs in the current sample of Latino children. To avoid problems
associated with shared method variance, I utilized a measure of peer victimization
based on peer-reports and GPA data from school records. In addition to GPA, I
included self- and peer-report measures of academic engagement to provide a broader
assessment of school functioning. Past research supports my prediction that peer
victimization would also be negatively related to academic engagement.
Given the evidence that peer victimization precedes academic difficulties
(Ladd et al., 1997; Schwartz et al., 2005), recent investigations have begun to examine
the mechanisms that explain the association between peer victimization and academic
achievement. A more complete understanding of these mechanisms may lead to the
development of better interventions (Juvonen, Nishina, & Graham, 2000). Researchers
have hypothesized that peer maltreatment might lead to poor school performance
through its effect on psychological adjustment (Juvonen et al., 2000; Schwartz et al.,
2005). The underlying notion behind these conceptualizations is that peer
victimization results in significant psychological distress (Hawker & Boulton, 2000)
and it is the distress that negatively impacts children’s school performance. This
proposed causal pathway is in accord with Wentzel and colleagues’ (e.g., Wentzel &
Caldwell, 1997) prior work, which viewed psychological distress as a mediating
mechanism linking negative peer relationships and poor school performance. In
specific tests of mediation, elementary and middle school students’ symptoms of

5

depression (Juvonen et al., 2000; Schwartz et al., 2005) have explained the association
between peer victimization and academic difficulties.
Academic engagement has been proposed as an important mechanism
connecting peer relationships and school performance (Wentzel, 1999), and
researchers have found some empirical support for this model. With a sample of
children in late elementary school, Furrer and Skinner (2003) found academic
engagement mediated the association between children’s sense of belonging and GPA.
Furrer and Skinner suggested that greater feelings of relatedness may increase
children’s interest in school and classroom participation because it is more fun for
children to engage in activities with individuals to whom they feel connected.
Additionally, Wentzel (2003) found a significant pathway linking sociometric status in
the sixth grade and GPA in the eighth grade through a measure assessing learning
effort. However, Wentzel’s (2003) analysis did not meet the criteria for mediation
because the inclusion of learning effort did not reduce the effect of sociometric status
on later GPA. The results of these two studies, while inconclusive, guided later
investigations proposing that victimization by peers could lead to decreased academic
engagement, which would eventually result in lower levels of academic achievement.
Some research has supported the notion that academic engagement could
operate as a mechanism linking peer victimization and school performance. With a
sample of children in late elementary school, Buhs (2005) found a significant indirect
effect of teacher ratings of peer exclusion on achievement test scores through teacher
ratings of academic engagement. On the other hand, in Buhs’ broad model, self-

6

reports of peer victimization did not have a direct effect on teacher ratings of academic
engagement. The lack of a direct effect eliminated the possibility for engagement to
mediate the relation between peer victimization and achievement. In later research,
Hoglund (2007) revealed a multi-informant measure of academic engagement
mediated the association between self-reports of physical victimization and GPA for
middle school girls. In contrast, Hoglund did not find academic engagement explained
the relation between self-reports of relational victimization and GPA for either boys or
girls.
Further research is warranted to investigate the mediating role of academic
engagement in the association between peer victimization and school performance
with different samples of children. More research is also needed that utilizes
alternative informants for peer victimization and academic engagement. Accordingly,
the second goal of the current study was to examine whether self- and peer-reports of
academic engagement mediated the association between peer-reports of victimization
and GPA. This study treats academic disengagement as a proximal outcome of peer
victimization as well as a mechanism that will likely explain the link between
victimization and low academic achievement.
A growing body of research has investigated factors that exacerbate or
ameliorate the association between victimization by peers and psychosocial
difficulties. This initial work focused on the quantity of friends as well as the presence
of a friend. Hodges, Malone, and Perry (1997) revealed that the negative association
between concurrent measures of internalizing and externalizing problems, and peer

7

victimization was attenuated for children with many friends. Furthermore, peer
maltreatment did not predict increases in internalizing and externalizing problems for
children that had a reciprocated best friend (Hodges, Boivin, Vitaro, & Bukowski,
1999). At the beginning of elementary school, friendships may play a key role in
helping children become socially integrated with their classmates, which could protect
them from later harassment (Schwartz, McFadyen-Ketchum, Pettit, & Bates, 1999).
Similarly, friends could buffer children against the negative effects of victimization
when children perceive their friends will act as protectors against bullies (Hodges et
al., 1999).
Some research intimates that the benefits associated with friendships on the
link between peer victimization and psychosocial difficulties may not be present in all
contexts. In difficult urban settings, such as the schools in the current study,
withdrawing from the peer group and having fewer friends could be relatively
adaptive. Research has indicated that “staying out of harm’s way” can be an important
coping mechanism for children in violent urban neighborhoods (Tolan, Guerra, &
Montaini-Klovdahl, 1997). In addition, corroborating their hypothesis that the nature
of peer victimization could differ across contexts, Hanish and Guerra (2000) found
that withdrawal did not correlate with peer victimization in poor urban areas with high
crime rates. In contrast, other researchers have observed modest positive associations
between withdrawal and peer victimization in middle-class settings (e.g., Boivin &
Hymel, 1997). Future large-scale research projects that collect data at schools in a
range of socioeconomic contexts are needed to assess whether the moderating effect of

8

friendships in the association between peer victimization and maladjustment differs
across settings.
Researchers have proposed that investigating the characteristics of children’s
friendships could lead to a greater understanding of the effect of friendships on
children’s development (Berndt, 1999; Hartup, 1996). Consistent with this suggestion,
some studies have investigated the moderating role of friends’ characteristics on the
link between peer victimization and psychosocial problems. Having friends with low
levels of peer victimization is associated with a weaker negative relation between
concurrent measures of internalizing and externalizing problems, and peer
victimization (Hodges et al., 1997). Furthermore, Fox and Boulton (2006) revealed
that the association between poor social skills and peer maltreatment was ameliorated
for children who had a reciprocated best friend with high levels of social preference.
Overall, these results demonstrate the possible beneficial impact of having well-
adjusted friends on the difficulties associated with peer maltreatment.
Unlike the advantages associated with friendships with well-adjusted peers,
friendships with disruptive (Berndt & Keefe, 1995) and anti-social (Laird, Pettit,
Dodge, & Bates, 1999) youth can predict future adjustment problems. Friends’
aggression may be an important variable as it can precede increases in children’s own
aggression (Werner & Crick, 2004). Moreover, Bagwell and Coie (2004) observed
that, in comparison to non-aggressive boys, aggressive boys cooperated less with their
friends and engaged in less on-task behavior with them. Friendships among antisocial

9

adolescent boys tended to be low in quality and rated as relatively less satisfying
(Dishion, Andrews, & Crosby, 1995).
The prior research investigating the advantages and disadvantages of particular
friendships informed Schwartz, Gorman, Dodge, Pettit, and Bates’ (in press) unique
investigation on moderators of the association between peer victimization and
academic functioning. With two samples of children in late elementary school,
Schwartz et al. (in press) revealed that the relation between peer victimization and
declines in GPA was ameliorated for children with numerous non-aggressive friends.
On the other hand, having greater numbers of aggressive friends exacerbated the link
between victimization by peers and increases in academic problems. Schwartz et al.’s
(in press) findings agreed with the previous research revealing the benefits of well-
adjusted friends and the potential difficulties associated with antisocial friends. One
question left unanswered by Schwartz et al.’s (in press) findings is the role of
children’s number of friends in the association between peer victimization and
academic achievement. Given the importance of academic engagement, the same
moderating effects of friendships in the association between peer maltreatment and
academic engagement deserve analysis as well.
The third goal of the current study was to investigate the moderating role of
friendship in the association between peer victimization and school functioning. I
investigated whether the number of friends related to the strength and/or direction of
the associations of peer victimization with GPA and academic engagement. The extant
literature did not allow me to form a definitive hypothesis regarding whether having

10

more friends would have an ameliorative function in the current sample of urban
Latino children. Consistent with Schwartz et al. (in press), I also tested the moderating
roles of the number of aggressive and non-aggressive friends. I anticipated that having
greater numbers of aggressive friends would exacerbate the link between peer
victimization and school functioning. Alternatively, having greater numbers of non-
aggressive friends was not likely to have negative ramifications (Schwartz et al., in
press). The analyses might help explain why not all children’s school performance
seems to suffer when they encounter peer harassment. Furthermore, investigating
these questions with a sample of Latino children residing in difficult urban
environments may help inform future research investigating the moderating role of the
socioeconomic context in which peer victimization occurs.
The role of gender is another issue warranting consideration in the current
study. Initial research found that boys experience more overt victimization than girls
while both genders encounter similar levels of relational victimization (Crick &
Grotpeter, 1996). This pattern of findings appears to hold for low-income Latino and
African American children (Storch, Nock, Masia-Warner, & Barlas, 2003). In
addition, the extant literature on relational aggression is mixed but the research
suggests that girls may engage in more of this form of aggressive behavior (Archer &
Coyne, 2005). The evidence regarding the role of gender in the link between peer
victimization and academic functioning is equivocal. Research with children in late
elementary school did not show the association between peer victimization and
academic achievement to differ across gender (Schwartz et al., 2005). In middle

11

school, some researchers found the relation was stronger for girls than boys (Graham,
Bellmore, & Mize, 2006; Hoglund, 2007). Clear evidence has yet to emerge
demonstrating that the moderating effect of friendship on the peer victimization-
academic functioning association differs across gender (Schwartz et al., in press). Due
to the current study’s small sample size, gender differences would have been difficult
to detect. Nevertheless, I cautiously investigated the role of gender in the current
analyses.
In summary, the current study explored the relation between peer victimization
and school functioning in a sample of Latino children attending elementary schools in
distressed urban neighborhoods. The study helps fill important gaps in the literature by
examining the link between peer maltreatment and academic difficulties in an
understudied population. Researchers have yet to examine the current research
questions in a sample composed of Latino children in elementary school or with a
sample of children from very low SES backgrounds. The current study also adds to the
growing body of work by investigating mediators and moderators of the relation
between peer victimization and school functioning. First, I assessed peer
victimization’s concurrent association with GPA and academic engagement. Second, I
examined whether self- and peer-reports of academic engagement mediated the
association between peer victimization and GPA. Finally, I investigated the
moderating role of children’s friendships (i.e., aggressive, non-aggressive, and total
number of friends) in the link between peer victimization and school functioning.

12

Method
Participant Recruitment
I recruited participants from two elementary schools located in urban sections
of Los Angeles County in June of 2007. Latinos comprised over 99% of both schools’
student bodies. The participating classrooms were nominally English-only
environments, but my informal observations revealed that some communication
among the students, teachers, and staff occurred in Spanish.
One of the participating schools was a charter school serving students living in
and around downtown Los Angeles. According to the charter school’s personnel, 79%
of the students qualified for the state free-lunch program and an additional 17%
qualified for the state reduced-price lunch program. Furthermore, 87% of the students’
families earned less than $30,000 per year. The area surrounding the charter school
had the highest crime rate in the city in the years prior to the data collection (Los
Angeles Police Department, 2005).
The second participating school was a four-track (i.e., children attended year
round to alleviate overcrowding) public elementary school located in a medium-sized
city south of downtown Los Angeles. The school had begun a pilot program allowing
all of the students to receive free breakfast and lunch. Prior to the start of the program,
school officials reported that over 97% of the students in the school qualified for free
or reduced-price lunch. The violent crime rate in the medium-sized city was nearly
twice the national average (Federal Bureau of Investigation, 2007).

13

At both schools, I distributed parental consent forms in English and Spanish,
depending on which language the children indicated their parents would prefer.
Regardless of whether their parents permitted them to take part in the project, all
children that returned their parental consent forms received a University of Southern
California (USC) pencil. I also conducted a raffle in each classroom and all children
that returned their parental consent forms had an opportunity to win a USC tee-shirt.
I invited all children in the third, fourth, and fifth grades at the charter school
to participate in the project. Seventy-five percent of the children returned the consent
forms, with 80% of their parents permitting them to participate. Consistent with the
standards for the field regarding the administration of peer nomination and peer rating
measures (e.g., Buhs, 2005; Schwartz, 2000), I did not administer the measures to
classrooms where less than two-thirds of the students consented to participate in the
project. The positive consent rates were 70% and 71% in the third grade classrooms
and 67% in the fourth grade classroom. Only 50% of the fifth grade classroom
returned positive consent forms.
At the public school, four fourth grade and five fifth grade classrooms going
“off track” (i.e., beginning a two-month break) were invited to participate in the
project. Seventy-one percent of the children returned the consent forms, with 74% of
their parents permitting them to participate. The positive consent rates in the
classrooms ranged from a low of 30% to a high of 75%. I administered the measures
to the two fourth grade classrooms and the two fifth grade classrooms with positive
consent rates above 66%.

14

Participants
The sample was composed of 135 third, fourth, and fifth graders. Forty-nine
children (18 boys; 31 girls) attended the charter school and 86 children (37 boys; 49
girls) attended the public school. The mean age of the participating children was 9.88
years (SD = 0.94; range = 8 - 12). With the exception of one African American child
at the charter school, Latinos constituted the entire sample. Forty-six percent of the
children reported speaking Spanish in their homes, 52% reported speaking English and
Spanish, and 2% reported speaking English in their homes. The schools classified 64%
of the participants as English language learners (ELLs). The children indicated that an
average of 5.4 individuals (SD = 2.02; range 2 - 16) lived in their homes. I asked the
initial 74 participants whether they and their parents were born in the United States.
However, this part of the procedure was discontinued after one teacher adamantly
voiced concerns about these questions. During the period of the data collection,
deportation was a legitimate concern for undocumented immigrants living in the
United States (Hing, 2006). The vast majority of the children (83%) reported being
born in the United States. On the other hand, they indicated that the majority of their
parents (Mothers = 87%; Fathers = 82%) were not born in the United States.
Measures
Peer victimization. The initial participants completed a 14-item peer
nomination inventory asking about a range of behavioral dispositions. I reduced the
length of the inventory to 10 items after I determined that the full battery of measures
consumed too much class time. The children received a list of all participating

15

students in their classrooms and were asked to nominate up to three peers who fit the
description of each item. The two third grade classrooms at the charter school had 29
participating children in total and the children in those classrooms received a list with
the names of the students in both classrooms. Two items assessed overt victimization
(i.e., “get hit, pushed, or bullied by other students,” “get beat up by other students”)
and one item assessed relational victimization (i.e., “have mean things said about
them”). I removed an additional relational victimization item (i.e., “get left out of fun
games or activities when other students are trying to hurt their feelings”) after the
initial participants had a difficult time distinguishing it from an analogous relational
aggression item. For the three peer victimization items, I calculated proportion scores
by dividing the number of nominations each child received by the number of potential
nominators in each classroom (Perry, Kusel, & Perry, 1988). The three peer
victimization items ( α = .80) were averaged to form a composite variable.
Aggression. The peer nomination inventory included two overt aggression
items (i.e., “start fights with other students by punching or pushing them,” “hit or push
other students”) and one relational aggression item (i.e., “say mean things about other
students”). When I reduced the length of the peer nomination inventory, I removed a
relational aggression item (i.e., “try to leave other students out of fun games or
activities when they are trying to hurt their feelings”). I averaged the proportion scores
for the three aggression items ( α = .90) to form a composite variable.
Friendship. To assess the children’s friendships, they were asked to nominate
their “best friend on the list” as well as “up to four other kids on the list that [they

16

were] really good friends with.” Following past research, children were only allowed
to nominate their classmates for the friendship items (Ladd et al., 1997). Prior research
has found that roughly 80% of friendship dyads occur within the context of elementary
school children’s classrooms (Kupersmidt, DeRosier, & Patterson, 1995). Due to the
small size of the two third grade classes at the charter school and the frequent
interactions between students in the two classes, I allowed them to nominate friends
from both classrooms. I classified pairs of children as friends if they reciprocally
nominated each other for any of the five friendship nominations (Criss, Pettit, Bates,
Dodge, & Lapp, 2002). I calculated the number of aggressive and non-aggressive
friends each child had by counting the number of friends that were above or below the
sample’s mean on the aggression composite variable (Schwartz et al., in press).
Peer groups. Following the social cognitive map (SCM) method used by
Farmer et al. (2002), the children were asked “Are there some kids in your class who
hang around together a lot? List the names of the kids in each of the different groups
in your class. Try to think of as many groups as you can.” Space was provided for the
children to list up to six groups with up to eight children per group. Additionally, the
children were allowed to list themselves in a group or groups and were allowed to list
their peers in multiple groups. Co-occurrence matrices for each classroom and distinct
peer groups were calculated using SCM 4.0 (Leung, 1998).
Academic engagement. As part of the peer nomination inventory, the children
were asked to nominate up to three academically engaged classmates (i.e., “pay
attention in class,” “follow the rules at school”). These two items were adapted from

17

Fredricks et al.’s (2005) behavioral engagement subscale. I averaged the proportion
scores for the two items (r = .85, p < .001) to form a composite variable. I also utilized
Fredricks et al.’s (2005) academic engagement scale to assess the children’s self-
reported academic engagement. Four items ( α = .74) from the behavioral engagement
scale (e.g., “I complete my homework on time”) were averaged to form a composite
variable. One item from Fredricks et al.’s (2005) original subscale correlated weakly
with the other four items and was removed. The reliability of the measure did not
differ substantially for the ELLs ( α = .77) and non-ELLs ( α = .67).
Grades. At the end of the marking period following the data collection, I
obtained the participating students’ grades from a review of school records. At both
schools, the teachers graded students on a 1 (far below grade level) to 4 (above grade
level) scale. The students’ grades in reading, writing, listening and speaking, math,
history, and science ( α = .91) were averaged to form a composite measure for GPA.
Procedure
I collected the data with trained research assistants in group testing sessions,
lasting 45 minutes to one hour, in the children’s classrooms. I randomly varied the
order of the peer nomination inventory and the self-report measure of academic
engagement across classrooms. I read a standardized set of instructions in English and
read each of the items aloud in English as the children completed them. A small
number of the children (i.e., less than five) had considerable difficulty understanding
some of the English items. For this small number of children, the teachers or their
fellow students provided translations of the difficult items into Spanish.

18

Missing Data
Missing data occurred because a small number of children did not complete the
self-report academic engagement measure and some children’s grades could not be
obtained from school records. Even though the amount of missing data was small (i.e.,
1%), casewise deletion would have eliminated seven of the participants from the
analyses. Based on the assumption that the missing data was missing at random (i.e.,
ignorable nonresponse), I imputed the missing data using multiple imputation (Rubin,
1987; Schafer & Graham, 2002) with PROC MI in SAS 9.1. All of the peer
nomination measures, friendship scores, and school functioning measures were
utilized to impute the missing values. I placed boundaries on the imputations forcing
the imputed values to be within a theoretically possible range for each measure (e.g.,
grades could only range from 1 to 4). To be conservative, 20 imputed datasets were
produced (Rubin, 1987). I conducted the analyses once with each imputed dataset and
the results were combined using PROC MIANALYZE, which combines the parameter
estimates and associated standard errors based on the recommendations of Rubin
(1987). Finally, I re-ran all of the analyses with only the complete cases and the results
were nearly identical to the results obtained with the imputed datasets.
Results
Descriptive Statistics
Means, standard deviations, and ranges for each measure are shown in Table 1.
The children received fewer nominations from their classmates for the peer
victimization items than for the academic engagement items. Both peer nomination

19

Table 1

Means, Standard Deviations, and Ranges for the Measures

Variable M SD Range
1. Peer victimization
a
0.08 0.07 0.00 - 0.46
2. Peer-reports of academic engagement
a
0.12 0.14 0.00 - 0.75
3. Self-reports of academic engagement 3.93 0.74 2.00 - 5.00
4. Friends 2.25 1.21 0 - 5
5. Aggressive friends 0.75 0.88 0 - 4
6. Non-aggressive friends 1.50 0.94 0 - 4
7. GPA
b
3.06 0.45 1.74 - 4.00
Note.
a
The peer victimization and peer-reports of academic engagement measures are
proportion scores based on the number of nominations each child received divided by
the number of potential nominators in each classroom.
b
GPA = Grade point average.

composite variables were positively skewed and were log transformed prior to further
analysis to reduce the potential influence of outliers. The children reported moderately
high levels of academic engagement. Their responses averaged close to a four (i.e., a
lot of the time) on the five-point scale for items assessing how frequently they behaved
in an academically engaged manner. With regards to their friendships, the children had
more non-aggressive friends than aggressive friends. Finally, the mean score for the
children’s GPAs indicated their teachers reported that their performance was, on
average, at grade level.
Bivariate Correlations
The correlations among the variables are displayed in Table 2 for the entire
sample. Gender (boys = 1, girls = 0), school (public school = 1, charter school = 0),
and ELL status (ELLs = 1, non-ELLs = 0) were coded as dichotomous variables. In
response to my first research question, peer victimization showed negative correlations

20

with GPA and the two academic engagement measures. The correlations between peer
victimization and the academic outcomes were small (Cohen, 1988) but significant.
Additionally, boys were more likely to be victimized than girls and showed lower
levels of academic engagement. ELLs had lower grades than their non-ELL peers.
Finally, the charter school’s teachers assigned significantly lower grades than the
teachers at the public school. However, ELLs comprised a larger proportion of the
students at the charter school than at the public school.
The Mediating Role of Academic Engagement
To answer my second research question, I examined the mediating roles of
self- and peer-reports of academic engagement in the association between peer
victimization and GPA. I assessed mediation based on the method recommended by
Baron and Kenny (1986). First, I estimated the strength of the association between the
predictor (i.e., peer victimization) and the mediators (i.e., self- and peer-reports of
academic engagement). Second, I estimated the strength of the association between the
predictor and the outcome (i.e., GPA). Third, I tested models with the outcome
variable regressed on both the predictor and mediators. Mediation is considered to
occur if the effect of the predictor on the mediators in the first models, the effect of the
predictor on the outcome in the second model, and the effect of the mediators on the
outcome in the third models are significant. In addition, the effect of the predictor in
the third models with the mediators needs to be reduced in comparison to its singular
effect in second model.

Table 2

Bivariate Correlations among all of the Variables

21

1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
1. Peer victimization -

2. Engagement self-reports -.31
***
-
-

3. Engagement peer-reports -.28
**
-.48
***
4. Friends -.14 -.07 -.13 -
5. Aggressive friends -.09 -.22
**
-.06 -.64
***
-
6. Non-aggressive friends -.26
**
-.11 -.23
**
-.69
***
-.11 -
7. GPA
a
-.22
*
-.36
***
-.47
***
-.04 -.01 -.06 -
8. Gender
b
-.31
***
-.19
*
-.17
*
-.12 -.25
**
-.39
***
-.12 -
9. School
c
-.17
*
-.26
**
-.09 -.05 -.16 -.09 -.38
***
-.06 -
10. ELL status
d
-.19
*
-.15 -.19
*
-.09 -.01 -.11 -.46
***
-.06 -.48
***
-
Note.
a
GPA = Grade point average.
b
Gender is coded boys = 1, girls = 0.
c
School is coded public = 1, charter = 0.
d
ELL
status is coded ELL = 1, non-ELL = 0.
*
p < .05.
**
p < .01.
***
p < .001.

22

Regression analyses revealed that peer victimization was a significant negative
predictor of self-reports ( β = -.22, p < .05) and peer-reports of academic engagement
( β = -.22, p < .05) after accounting for gender, school, and ELL status, which were
conceptualized as control variables. The effect of peer victimization on GPA did not
differ across gender, school, or ELL status and these interactions were not retained in
the models presented in Table 3. As shown in Model 1, F(4, 130) = 10.84, p < .001, R²
= .25, in Table 3, peer victimization had a small but significant negative association
with GPA after accounting for the control variables. The other two models displayed
in Table 3 revealed the effect of peer victimization on GPA was reduced to non-
significance when self-reports of academic engagement (see Model 2) and peer-
reports of academic engagement (see Model 3) were entered separately. The effects of
the academic engagement variables in Model 2, F(5, 129) = 11.64, p < .001, R² = .32,
and Model 3, F(5, 129) = 16.03, p < .001, R² = .38, were both significant and positive.
Instead of complete mediation, where the effects of peer victimization would have
been reduced to zero in Models 2 and 3, the results indicated that partial mediation had
occurred. Goodman’s (1960) test revealed that the indirect effect of peer victimization
through self-reports (Z = -1.99, p < .05) and peer-reports (Z = -2.25, p < .05) of
academic engagement were significant.
1
The results of the regression analyses

1
To assess the significance of the indirect effect, Goodman’s (1960) test was used in place of Sobel’s
(1982) test, which is recommended by Baron and Kenny (1986). Shrout and Bolger (2002) found that
Sobel’s test has a higher Type II error rate than other methods. The bootstrap method recommended by
Shrout and Bolger found the indirect effect of peer victimization on GPA through self-reports of
academic engagement to be significant (p < .05) in 18 of the 20 imputed datasets. Additionally, the
bootstrap method revealed the indirect effect through peer-reports of academic engagement was
significant (p < .05) in all 20 imputed datasets.

Table 3

The Mediating Role of Self- and Peer-Reports of Academic Engagement in the Association between Peer Victimization and
Grade Point Average (GPA)

23

Model1
a
Model 2
b
Model 3
c
Variable B SE β B SE β B SE β
Step 1
Gender
d
-0.13

0.08 -.14 -0.16 0.07 -.18
*
-0.17 0.07 -.18
*
School
e
-0.18 0.08 -.19
*
-0.12 0.08 -.13 -0.18 0.07 -.19
*
ELL status
f
-0.29

0.08

-.31
***
-0.29

0.08

-.30
***
-0.23

0.08

-.24
**
Step 2
Peer victimization -2.78

1.37

-.17
*
-1.84

1.36

-.11

-1.40

1.28

-.08
Step 3
Engagement self-reports -0.16

0.05

-.27
**
Engagement peer-reports -3.59 0.69 -.39
***

Note.
a
Model 1 estimates the effect of peer victimization on GPA after accounting for gender, school, and ELL status.
b
Model
2 shows the reduction of the effect of peer victimization after the addition of self-reports of academic engagement.
c
Model 3
shows the reduction of the effect of peer victimization after the addition of peer-reports of academic engagement.
d
Gender is
coded boys = 1, girls = 0.
e
School is coded public = 1, charter = 0.
f
ELL status is coded ELL = 1, non-ELL = 0.
*
p < .05.
**
p < .01.
***
p < .001.

24
supported my prediction that academic engagement would mediate the association
between peer victimization and GPA.
The Moderating Role of Children’s Friendships
In response to my third research question, I investigated whether the children’s
number of friendships moderated the association between peer victimization and GPA.
The variables were mean-centered prior to analysis. I tested a model with GPA
predicted from gender, school, ELL status, peer victimization, peer victimization ×
gender, friends, friends × gender, friends × peer victimization, and friends × peer
victimization × gender. For exploratory purposes I tested the same series of
interactions with school and ELL status as I did with gender. None of these
exploratory interaction terms were significant and I did not include them in the final
model. The full model was significant, F(9, 125) = 4.88, p < .001, R² = .28, but the
peer victimization × friends interaction term was not significant. A significant peer
victimization × friends interaction term would have indicated that the children’s
number of friends moderated the association between peer victimization and GPA.
Additional analyses were conducted to investigate the moderating roles of the
children’s number of aggressive and non-aggressive friends in the association between
victimization by peers and GPA. The regression models using aggressive and non-
aggressive friends were identical to the model investigating the moderating role of the
number of friends, with the exception that aggressive and non-aggressive friends were
utilized as predictors in place of friends. The full models for aggressive, F(9, 125) =

25
5.04, p < .001, R² = .27, and non-aggressive friends, F(9, 125) = 4.87, p < .001, R²
=.26, were significant. However, the peer victimization × friends interaction terms did
not reach significance in either model.
I also assessed whether the children’s number of friends moderated the
association between peer victimization and self-reports of academic engagement. The
model with academic engagement as the outcome variable had the same predictors as
the prior models with GPA as the outcome variable. I again tested the same series of
interactions with school and ELL status as I did with gender for exploratory purposes.
These exploratory interaction terms were not significant and I did not include them in
the final model (see Model 1 in Table 4). The full model was significant, F(9, 125) =
4.13, p < .001, R² = .23, and revealed a significant friends × peer victimization
interaction, which is indicative of moderation. The interaction was decomposed
following the method outlined by Aiken and West (1991). I tested models predicting
academic engagement with the same predictors as the full model but with the number
of friends fixed at low (one standard deviation below the mean), medium (the mean),
Table 4

The Moderating Role of Friendships in the Association between Peer Victimization
and Self-Reports of Academic Engagement
Model 1
a
Model 2
b
Variable B SE β B SE β
Gender
c
-0.17 0.13 -.11 -0.18 0.17 -.12
School
d
-0.26 0.14 -.17 -0.22 0.14 -.15
ELL status
e
-0.10 0.15 -.07 -0.10 0.14 -.07
Peer victimization -5.94 3.22 -.22 -6.33 3.35 -.23
Peer victimization × gender -5.82 5.02 -.14 -3.25 4.92 -.08
Friends -0.10 0.06 -.16
Friends × gender -0.01 0.10 -.01

26
Note.
a
Model 1 tests the moderating role of friends.
b
Model 2 tests the moderating
role of aggressive friends.
c
Gender is coded boys = 1, girls = 0.
d
School is coded
public = 1, charter = 0.
e
ELL status is coded ELL = 1, non-ELL = 0.
Table 4, Continued
Friends × peer victimization -7.44 2.72 -.31
**

Friends × peer victimization ×
gender
-4.02 4.44 -.09
Aggressive friends -0.18 0.10 -.21
Aggressive friends × gender -0.04 0.14 -.04
Aggressive friends × peer
victimization

-11.53 4.16 -.35
**
Aggressive friends × peer
victimization × gender

-8.19 5.70 -.18
**
p < .01.

and high (one standard deviation above the mean). The association between
victimization by peers and self-reports of academic engagement changed as the
number of friends moved from low ( β = .11, ns), to medium ( β = -.22, ns), to high ( β =
-.55, p < .01). The decomposition revealed that the negative association between peer
victimization and self-reports of academic engagement was only significant at high
levels of friends.
My last set of analyses examined the moderating role of the number of
aggressive and non-aggressive friends in the association between peer victimization
and academic engagement. The full model for aggressive friends was significant, F(9,
125) = 4.32, p < .001, R² = .23, and there was a significant aggressive friends × peer
victimization interaction (see Model 2 in Table 4). The full model for non-aggressive
friends was significant, F(9, 125) = 3.00, p < .01, R² = .18, but there was not a
significant non-aggressive friends × peer victimization interaction. To decompose the

27
significant aggressive friends × peer victimization interaction, I tested models
predicting academic engagement with the same predictors as the full model (Model 2
in Table 4) but with the number of aggressive friends fixed at low, medium, and high.
The results were consistent with the findings based on the number of friends. There
was a significant negative association between peer victimization and self-reports of
academic engagement when the number of aggressive friends was fixed at high ( β = -
.55, p < .01), but not at medium ( β = -.23, ns), or low ( β = .14, ns) levels of aggressive
friends.
Exploratory SCM Analyses
Across all of the classrooms, the SCM method produced 23 distinct peer
groups. The groups ranged in size from 2 to 11 members (M = 5.83, SD = 2.64) and
one child was not classified as a group member. A co-occurrence matrix from one
classroom is displayed in Table 5. The number of times each child was listed in a peer
group is displayed in italics on the diagonal of the matrix. Additionally, the matrix
shows the number of times each child was listed in the same peer group with every
other child in the classroom. For instance, Girl 1 (G1) was listed in the same peer
group with G2 a total of 18 times. Alternatively, G1 was only listed as being in the
same peer group with boy 2 (B2) three times. SCM 4.0 detected four distinct peer
groups in the classroom displayed in Table 5. G1 through G5 and B1 were in the first
peer group. The other peer groups were composed of G6 through G10, B2 through B7,
as well as B8 and B9, respectively. B10 received nominations as part of a peer group
with every other student in the class, but was not classified as a group member.

Table 5
A Social Cognitive Map Co-occurrence Matrix for One Classroom in the Study
28
G1 G2 G3 G4 B1 G5 G6 G7 G8 G9 G10 B2 B3 B4 B5 B6 B7 B8 B9 B10
G1 21
G2

18 19
G3 18 16

18
G4 11 9 10 15
B1 5 5 4 2 8
G5 5 4 5 5 2 7
G6 5 3 3 3 1 3 19
G7 2 1 1 1 0 1 15 19
G8 2 1 1 0 0 0 13 17 17
G9 2 3 1 2 1 1 12 15 13 17
G10

1 2 0 0 1 0 12 16 15 14 17
B2 3 3 1 4 2 0 2 0 0 1 1 14
B3 3 1 2 2 1 0 2 1 1 0 0 6 13
B4 2 0 1 1 1 0 2 1 1 0 0 5 8 11
B5 3 3 1 4 3 0 2 0 0 1 1 8 3 5 11
B6 2 3 1 1 3 1 1 0 0 1 1 9 5 2 3 10
B7 1 0 0 1 1 0 1 0 0 0 0 6 6 7 5 3 9
B8 2 1 1 0 2 0 2 1 1 0 0 1 2 2 2 1 1 12
B9 0 0 0 0 1 0 0 0 0 0 0 0 2 1 1 0 1 10 11
B10 1 2 1 4 2 2 4 3 2 3 3 3 1 1 5 1 1 1 1 10
Note. G1 = girl 1. B1 = boy 1. The matrix displays the number of times each child was listed in the same peer
group with every other child in the classroom. The number of times each was listed in a peer group is displayed in
italics on the diagonal of the matrix.

29
Discussion
The link between peer victimization and school functioning has been the focus
of a growing number of studies, and the current investigation adds to this body of
work in an important number of ways. First, the results showed that higher levels of
peer victimization were concurrently associated with low GPA and poor academic
engagement in a sample of Latino children attending elementary schools in distressed
urban neighborhoods. The analyses also revealed that academic engagement operated
as a mediating mechanism linking peer maltreatment and school performance.
Moderator analyses indicated that the negative association between victimization by
peers and academic engagement was exacerbated for children with numerous friends
in their classrooms. Further analyses revealed that the negative association between
victimization and engagement was stronger for children with many aggressive friends
but the strength of the relation between victimization and engagement was not related
to the children’s number of non-aggressive friends.
Prior research, conducted with youth in middle-class settings (e.g., Ladd et al.,
1997) and diverse urban environments (e.g., Schwartz et al., 2005), has revealed that
peer victimization is associated with poor school performance. The same negative
association has been shown to hold for Latino middle school students (Nishina et al.,
2005) and with a sample composed largely of low-income Latino children in
elementary school (Schwartz, 2000). As anticipated, the results of the present study
revealed a small but significant negative association between peer maltreatment and
GPA in a sample of Latino children in the third, fourth, and fifth grades. The

30
correlational findings in the current study are an important extension of past research
because the sample was composed of Latino elementary school children attending two
highly segregated urban schools. Furthermore, the children in the present study were
from very low-income homes and were consequently at risk for academic difficulties
(Sirin, 2005).
Consistent with past research conducted with samples comprised mostly of
middle-class Caucasian students (Buhs, 2005; Hoglund, 2007), peer victimization was
also negatively related to academic engagement. The negative relation between peer
victimization and academic engagement highlights another academic problem
associated with victimization for Latino children in urban neighborhoods. Poor
academic engagement may be a particularly important factor for low-income Latino
children already experiencing academic difficulties. The low academic engagement
displayed by victims of bullying in elementary school can have long-term
implications. Academic engagement (Buhs, 2005) and related factors such as
academic self-concept (Guay, Marsh, & Boivin, 2003) have been shown to predict
changes in academic achievement. Other prior research showed low academic
engagement in elementary school increased children’s odds of dropping out of high
school (Alexander, Entwisle, & Horsey, 1997). These negative outcomes could
partially be due to teachers’ tendency to be less supportive and less involved with
academically disengaged students (Skinner & Belmont, 1993).
The current findings provide further support for the notion that peer
victimization could have negative ramifications for children’s academic adjustment in

31
a wide range of contexts. Importantly, the current results highlight peer maltreatment
as an additional factor that may negatively impact Latino children’s school success.
Exposure to peer maltreatment could exacerbate the academic difficulties many Latino
children have due to their low SES (Sirin, 2005) and their status as ELLs (Lutkus et
al., 2005). Consistent with this suggestion, the current analyses showed peer
victimization was significantly related to lower GPA after accounting for ELL status,
which was also associated with poorer GPA.
The growing school bullying literature conflicts with Fordham and Ogbu’s
(1986) suggestion that high academic achievement may place minority youth at risk
for peer harassment. In the current study, children targeted for bullying tended to have
worse academic performance than their non-victimized peers. This finding confirms
other past research with African American and Latino youth attending elementary and
middle schools in urban settings (Nishina et al., 2005; Schwartz, 2000). It is important
to note that Fordham and Ogbu’s conclusions were based on ethnographic research
with high school students. By high school, the educational disadvantages and
stereotype threat that African American students encounter may cause many of them
to disidentify with school (Steele, 1997), which could have implications for the
connection between peer harassment and school functioning. Future empirical
research is needed to explore the link between peer victimization and school
functioning with underrepresented minority students in high school.
The results of the present study supported my hypothesis that academic
engagement would mediate the relation between peer victimization and school

32
performance. The effect of peer victimization on GPA was reduced to non-
significance when self- and peer-reports of academic engagement were separately
included in the regression model. Additionally, the indirect effects of peer
maltreatment on GPA through both academic engagement measures were significant.
The model with peer-reports of victimization, self-reports of academic engagement,
and GPA acquired from school records provided the most stringent test of mediation
because a different informant was used for the predictor, mediator, and outcome
variables.
Academic engagement makes intuitive sense as a variable that could have an
internal psychological significance (Baron & Kenny, 1986) following peer
victimization and logically explains the association between victimization and
academic achievement. Victims of bullying likely spend an inordinate amount of
energy focusing on protecting themselves and avoiding classroom bullies, which could
distract them from concentrating on their schoolwork (Buhs et al., 2006). It has also
been suggested that children interpret peer victimization as an indication they are not
attractive partners for classroom activities and they may believe they do not possess
the skills needed for success at school (Buhs, 2005). In general, victimized children
are probably more prone to dislike school and disengage from classroom activities.
The present study adds to the growing body of research showing that academic
engagement mediates the association between peer victimization and school
performance (Buhs, 2005; Hoglund, 2007) by investigating this association in an
unexamined context. The results showed that academic engagement only partially

33
mediated the association between peer victimization and school performance. Baron
and Kenny (1986) argue that partial mediation is indicative of the existence of other
mediating pathways. Consistent with this proposition, prior work has found that
psychological distress, such as symptoms of depression, could be an important
mechanism linking peer maltreatment to academic difficulties (Juvonen et al., 2000;
Schwartz et al., 2005). Overall, the accumulating findings indicate that peer
victimization could impact children’s school performance through multiple pathways.
Taken as a whole, the results of the current study add to the body of research
indicating that peer victimization can have implications for children’s school
functioning. In conjunction with enhanced academic programming, schools seeking to
improve their students’ achievement may investigate a number of school-wide
bullying programs (i.e., targeting students, parents, and teachers) that have shown
effectiveness (Olweus, 1993). Schwartz et al. (2005) suggested that interventions
should account for the connection between peer victimization and internalized distress.
These interventions could additionally focus on providing children with adaptive
coping strategies (Nishina et al., 2005). The present results indicate that incorporating
an emphasis on academic engagement into bullying programs might help break the
link between peer victimization and poor academic achievement. Given the stability of
academic achievement (Schwartz et al., 2005), academic engagement may be easier to
modify through intervention. Middle to late elementary school is an opportune time to
include an emphasis on academic engagement in bullying programs because the period

34
coincides with the beginning of the decline in the level of importance children place
on school (Fredricks & Eccles, 2002).
The results of the moderator analyses revealed that the children’s number of
friends did not relate to the strength or direction of the association between peer
victimization and GPA. The moderating effects of the number of aggressive and non-
aggressive friends in the same association were also not significant in the current
sample. In contrast, Schwartz et al. (in press) found that the negative relation between
peer maltreatment and decreases in GPA was exacerbated for children with numerous
aggressive friends and ameliorated for children with many non-aggressive friends. The
current study’s low power, due to the small sample size and non-experimental design,
makes the non-significant findings difficult to interpret (McClelland & Judd, 1993).
Another caveat is that Schwartz et al. (in press) investigated the effect of victimization
by peers on later GPA after accounting for prior GPA. The present study investigated
the moderating role of friendship on the concurrent association between peer
victimization and GPA.
The current analyses with academic engagement as the dependent variable
revealed that the number of friends was a significant moderator. In comparison to low
GPA, academic disengagement could be a more proximal outcome of peer
maltreatment. Consequently, researchers may be more likely to detect the moderating
effects of friendship in the concurrent association between peer victimization and
academic engagement. Prior studies have suggested that friends can buffer children
against the negative effects of bullying (Hodges et al., 1997), and that youth with a

35
reciprocated friend display lower levels of emotional distress (Wentzel, Barry, &
Caldwell, 2004). With the current sample, however, the negative association between
peer victimization and academic engagement was exacerbated for children with many
friends. Friendship may not operate as a protective factor in all contexts, so future
research should explore this issue by incorporating schools from a wide range of
contexts. The present analyses additionally revealed that the association between peer
victimization and academic engagement was negligible for children with few friends.
These results are consistent with Hanish and Guerra’s (2000) hypothesis that
withdrawing from the peer group could be relatively adaptive in violent urban
environments.
The final set of analyses investigated the moderating effects of the number of
aggressive and non-aggressive friends in the relation between peer victimization and
academic engagement. The number of aggressive friends was a significant moderator
of the peer victimization-academic engagement association, but the effect of non-
aggressive friends was not significant. The effect of aggressive friends indicated that
the negative relation between peer victimization and academic engagement was
exacerbated for children with high numbers of aggressive friends. The results are
consistent with Schwartz et al.’s (in press) finding that the association between peer
victimization and declines in GPA was intensified for children with many aggressive
friends. Research has shown that aggressive friends do not promote positive
adjustment conducive to school success (e.g., Hartup, 1996; Werner & Crick, 2004). It
seems unlikely that having aggressive friends would support positive behavioral

36
adjustment in any school context. The effect of the number of aggressive friends in the
present study was the same as the effect of the number of friends, which is not
surprising given the moderate correlation between the two variables. Nevertheless, in
harsh urban environments, the possibility exists that friendships have the same effect
as aggressive friendships in other contexts (e.g., Schwartz et al., in press) because the
overall aggression levels of the classrooms are elevated in these difficult settings
(Thomas & Bierman, 2006).
Even though a number of researchers have suggested that peer victimization
could have different implications in divergent settings (e.g., Hanish & Guerra, 2000;
Schwartz et al., 2005), little research has investigated the role of the context in which
peer victimization occurs. One notable study recently investigated the role of
classrooms’ ethnic diversity and found higher levels of diversity related to lower
levels of self-reported peer victimization and vulnerability (Graham, 2006). Graham
suggested that the benefits of ethnic diversity resulted from the ethnic groups sharing
power in diverse classrooms. Further research investigating contextual factors
impacting the association between peer victimization and psychosocial difficulties is
clearly warranted.
A number of bullying interventions have utilized peer-support systems to
encourage children to create a school climate that opposes bullying (Smith,
Ananiadou, & Cowie, 2003). In particular, researchers have proposed that promoting
the formation of friendships may decrease the incidence of bullying (Boulton,
Trueman, Chau, Whitehand, & Amatya, 1999). There is some empirical support that

37
befriending interventions can have a positive effect on school bullying (Mensini,
Codecasa, Benelli, & Cowie, 2003). However, Schwartz et al.’s (in press) findings
regarding the exacerbating role of aggressive friends led them to warn against
potential iatrogenic effects (Dishion, McCord, & Poulin, 1999). The results of the
current study provide more support for the claim that friendships with certain peers
might not have positive effects on bullying. Furthermore, befriending interventions are
not guaranteed to have the same beneficial impact in all settings.
The present study has a number of limitations that warrant discussion. First,
the current sample size was small in comparison to other studies investigating school
bullying, which limited the statistical power of the analyses. The study’s results should
be viewed as preliminary because of the limited sample size. Additionally, the small
sample size meant that only a limited number of peer groups were detected using the
SCM method. Future research with larger samples could investigate the impact of
children’s peer groups assessed using SCM on the correlates of peer victimization.
The small sample size was due, in part, to the relatively low overall positive consent
rate. The low consent rate could mean that the children who took part in the project
were not entirely representative of the schools’ populations. Researchers typically
have difficulty obtaining parental consent when conducting school-based research
(Fletcher & Hunter, 2003) and the problems appear to be magnified in schools with
children from low SES backgrounds (Dent et al., 1993). Given the demands of the
recent push for school accountability, it has become increasingly difficult for
investigators conducting school-based research to gain access to schools.

38
Unfortunately, schools that serve low-income minority populations tend to have the
lowest test scores. These schools are frequently under the most pressure to improve
their students’ achievement and many feel they cannot afford to dedicate time for
research projects that do not directly benefit them.
Due in part to the measurement difficulties associated with school-based
research with preadolescents, the current study did not assess the role of a number of
important factors for Latino youth, such as acculturation, generational status, and
ethnic identity. These variables can have important implications for children’s
academic outcomes. For example, researchers have shown that mothers’ acculturation
style is associated with Latino children’s pre-reading skills (Farver, Eppe, & Ballon,
2006). In adolescence, researchers found that more strongly identifying with one’s
own ethnic group related to more positive academic attitudes (Fuligni, Witkow, &
Garcia, 2005), and having a positive sense of one’s own ethnic group predicted better
school performance (Supple, Ghazarian, Frabutt, Plunkett, & Sands, & 2006). Much
less work has examined the impact of acculturation, generational status, and ethnic
identity on peer victimization. One recent study failed to uncover evidence that
generational status moderated the relation between peer victimization and
psychosocial adjustment (McKenney, Pepler, Craig, & Connolly, 2006), but further
research is necessary. Additionally, researchers have suggested that future studies
should examine the impact of generational status and country of origin on students’
peer relations with others in their own ethnic group (Bellmore, Nishina, Witkow,
Graham, & Juvonen, 2007). The role of children’s dominant language could also

39
impact peer relations in settings, such as the schools in the current study, with large
groups of ELLs and native English-speakers.
Measuring all of the variables concurrently is another major limitation of the
current study. Even though prior longitudinal research has suggested that peer
victimization precedes academic difficulties (Ladd et al., 1997; Schwartz et al., 2005),
the current models could have been specified differently with multiple waves of data.
For example, peer victimization could be an outcome of poor academic functioning as
well as a predictor of it. This model would be consistent with Egan and Perry’s (1998)
results showing reciprocal relations between peer victimization and low self-regard.
More longitudinal research is needed to investigate peer victimization and school
functioning over extended periods of time.
The utilization of a single measure of peer maltreatment based on overt and
relational victimization items (Crick & Grotpeter, 1996) was a limitation of the current
study. Past research has indicated that overt and relational victimization can have a
differential impact on academic outcomes (Hoglund, 2007). The present study also
used peer-reports of victimization while previous investigators have utilized self-
report data (Juvonen et al., 2000), teacher ratings (Schwartz, 2000), and multi-
informant latent variables (e.g., Schwartz et al., 2005) to assess peer victimization.
The use of different informants could have a potential influence on the pattern of
findings in the current study (Ladd & Kochenderfer-Ladd, 2002). Indexing academic
engagement with only items tapping behavioral engagement was another measurement

40
limitation. Subsequent research could investigate peer victimization’s association with
emotional and cognitive engagement.
Finally, Schwartz et al. (in press) warned that their results investigating the
moderating roles of aggressive and non-aggressive friends could be due to any number
of unmeasured variables. Likewise, children that formed many friendships in their
classrooms in the present study could differ in important ways from children that did
not form many friendships. These unmeasured differences could be accounting for the
observed effects of children’s friendships. For example, it is conceivable that children
forming many friendships are heavily invested in their peer group. Because of the
importance they place on the peer group, peer victimization could have a greater
negative impact on these children. In addition, given that friendships with antisocial
peers tend to be low in quality (Dishion et al., 1995), the quality of the children’s
friendships could be accounting for the moderating effect of aggressive friends in the
current sample.
In conclusion, the current study investigated the association between peer
victimization and school functioning in a sample of Latino children attending
elementary schools in poor urban neighborhoods. The present findings revealed that
peer victimization was concurrently associated with low GPA and poor academic
engagement. Moreover, the analyses revealed that academic engagement mediated the
relation between peer maltreatment and GPA. Moderator analyses provided evidence
that the connection between peer victimization and school functioning is more
nuanced than a simple main-effects model. In contrast to prior research demonstrating

41
the benefits of friends, moderator analyses indicated that the negative association
between peer victimization and academic engagement was exacerbated for children
with numerous friends in their classrooms. This interesting finding may be related to
the children’s distressed urban environment where withdrawing from the peer group
could be beneficial. Further analyses revealed that the negative association between
victimization and engagement was stronger for children with many aggressive friends.
The results highlight the need for additional research investigating the association
between peer victimization and psychosocial problems in diverse contexts.

42
References
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting
interactions. Thousand Oaks, CA: Sage.

Alexander, K. L., Entwisle, D. R., & Horsey, C. S. (1997). From first grade forward:
Early foundations of high school dropout. Sociology of Education, 70, 87-107.

Archer, J., & Coyne, S. M. (2005). An integrated review of indirect, relational, and
social aggression. Personality and Social Psychology Review, 9, 212-230.

Bagwell, C. L., & Coie, J. D. (2004). The best friendships of aggressive boys:
Relationship quality, conflict management, and rule-breaking behavior.
Journal of Experimental Child Psychology, 88, 5-24.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in
social psychological research: Conceptual, strategic, and statistical
considerations. Journal of Personality and Social Psychology, 51, 1173-1182.

Bellmore, A. D., Nishina, A., Witkow, M. R., Graham, S., & Juvonen, J. (2007). The
influence of classroom ethnic composition on same- and other-ethnicity peer
nominations in middle school. Social Development, 16, 720-740.

Berndt, T. J. (1999). Friends’ influence on students’ adjustment to school. Educational
Psychologist, 34, 15-28.

Berndt, T. J., & Keefe, K. (1995). Friends’ influence on adolescents’ adjustment to
school. Child Development, 66, 1312-1329.

Boivin, M., & Hymel, S. (1997). Peer experiences and social self-perceptions: A
sequential model. Developmental Psychology, 33, 135-145.

Boulton, M. J., Trueman, M., Chau, C., Whitehand, C., & Amatya, K. (1999).
Concurrent and longitudinal links between friendship and peer victimization:
Implications for befriending interventions. Journal of Adolescence, 22, 461-
466.

Buhs, E. S. (2005). Peer rejection, negative peer treatment, and school adjustment:
Self-concept and classroom engagement as mediating processes. Journal of
School Psychology, 43, 407-424.

43
Buhs, E. S., Ladd, G. W., & Herald, S. L. (2006). Peer exclusion and victimization:
Processes that mediate the relation between peer group rejection and children’s
classroom engagement and achievement? Journal of Educational Psychology,
98, 1-13.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.).
Hillsdale, NJ: Lawrence Earlbaum Associates.

Crick, N. R., & Grotpeter, J. K. (1996). Children’s treatment by peers: Victims of
relational and overt aggression. Development and Psychopathology, 8, 367-
380.

Criss, M. M., Pettit, G. S., Bates, J. E., Dodge, K. A., & Lapp, A. L. (2002). Family
adversity, positive peer relationships, and children’s externalizing behavior: A
longitudinal perspective on risk and resilience. Child Development, 73, 1220-
1237.

Dent, C. W., Galaif, J., Sussman, S., Stacy, A., Burtun, D., & Flay, B. (1993).
Demographic, psychological, and behavioral differences in samples of actively
and passively consented adolescents. Addictive Behaviors, 18, 51-56.

Dishion, T. J., Andrews, D. W., & Crosby, L. (1995). Antisocial boys and their friends
in early adolescence: Relationship characteristics, quality, and interactional
processes. Child Development, 66, 139-151.

Dishion, T. J., McCord, J., & Poulin, F. (1999). When interventions harm: Peer groups
and problem behavior. American Psychologist, 54, 755-764.

Egan, S. K., & Perry, D. G. (1998). Does low self-regard invite victimization?
Developmental Psychology, 34, 299-309.

Estell, D. B., Farmer, T. W., & Cairns, B. D. (2007). Bullies and victims in rural
African American youth: Behavioral characteristics and social network
placement. Aggressive Behavior, 33, 145-159.

Farmer, T. W., Leung, M.-C., Pearl, R., Rodkin, P. C., Cadwallader, T. W., & Van
Acker, R. (2002). Deviant of diverse peer groups? The peer affiliations of
aggressive elementary students. Journal of Educational Psychology, 94, 611-
620.

44
Farver, J. M., Eppe, S., & Ballon, D. (2006). Acculturation and family characteristics
that facilitate literacy development among Latino children. In M. H. Bornstein
& L. R. Cote (Eds.), Acculturation and parent-child relationships:
Measurement and development (pp. 223-248). Mahwah, NJ: Erlbaum.

Federal Bureau of Investigation. (2007). California: Offenses known to law
enforcement by state by city, 2006 [Data file]. Available from Federal Bureau
of Investigation Web site, http://www.fbi.gov

Fletcher, A. C., & Hunter, A. G. (2003). Strategies for obtaining parental consent to
participate in research. Family Relations, 52, 216-221.

Fordham, S., & Ogbu, J. U. (1986). Black students’ school success: Coping with the
‘burden of ‘acting white’”. Urban Review, 18, 176-206.

Fox, C. L., & Boulton, M. J. (2006). Friendship as a moderator of the relationship
between social skills problems and peer victimization. Aggressive Behavior,
32, 110-121.

Fox, M. A., Connolly, B. A., & Snyder, T. D. (2005). Youth indicators 2005: Trends
in the well-being of American youth (NCES 2005-050). U.S. Department of
Education, National Center for Education Statistics. Washington, DC: U.S.
Government Printing Office.

Fredricks, J. A., Blumenfeld, P., Friedel, J., & Paris, A. (2005). School engagement. In
K. A. Moore & L. H. Lippman (Eds.), What do children need to flourish:
Conceptualizing and measuring indicators of positive development (pp. 305-
321). New York, NY: Springer.

Fredricks, J. A., Blumenfeld, P. C., & Paris, A. H. (2004). School engagement:
Potential of the concept, state of the evidence. Review of Educational
Research, 74, 59-109.

Fredricks, J. A., & Eccles, J. S. (2002). Children’s competence and value beliefs from
childhood through adolescence: Growth trajectories in two male-sex-typed
domains. Developmental Psychology, 38, 519-533.

Fuligni, A. J., Witkow, M., & Garcia, C. (2005). Ethnic identity and the academic
adjustment of adolescents from Mexican, Chinese, and European backgrounds.
Developmental Psychology, 41, 799-811.

45
Furrer, C., & Skinner, E. (2003). Sense of relatedness as a factor in children’s
academic engagement and performance. Journal of Educational Psychology,
95, 148-162.

Goodman, L. A. (1960). On the exact variance of products. Journal of the American
Statistical Association, 55, 708-713.

Graham, S. (2006). Peer victimization in school: Exploring the ethnic context. Current
Directions in Psychological Science, 15, 317-321.

Graham, S., Bellmore, A. D., & Mize, J. (2006). Peer victimization, aggression, and
their co-occurrence in middle school: Pathways to adjustment problems.
Journal of Abnormal Child Psychology, 34, 363-378.

Guay, F., Marsh, H. W., & Boivin, M. (2003). Academic self-concept and academic
achievement: Developmental perspectives on their causal ordering. Journal of
Educational Psychology, 95, 124-136.

Hanish, L. D., & Guerra, N. G. (2000). Predictors of peer victimization among urban
youth. Social Development, 9, 521-543.

Hartup, W. W. (1996). The company they keep: Friendships and their developmental
significance. Child Development, 67, 1-13.

Hawker, D. S. J., & Boulton, M. J. (2000). Twenty years’ research on peer
victimization and psychosocial maladjustment: A meta-analytic review of
cross-sectional studies. Journal of Child Psychology and Psychiatry, 41, 441-
455.

Hing, B. O. (2006). Deporting our souls: Values, morality, and immigration policy.
New York: Cambridge University Press.

Hodges, E. V. E., Boivin, M., Vitaro, F., & Bukowski, W. M. (1999). The power of
friendship: Protection against an escalating cycle of peer victimization.
Developmental Psychology, 35, 94-101.

Hodges, E. V. E., Malone, M. J., & Perry, D. G. (1997). Individual risk and social risk
as interacting determinants of victimization in the peer group. Developmental
Psychology, 33, 1032-1039.

Hoglund, W. L. G. (2007). School functioning in early adolescence: Gender-linked
responses to peer victimization. Journal of Educational Psychology, 99, 683-
699.

46
Juvonen, J., Nishina, A., & Graham, S. (2000). Peer harassment, psychological
adjustment, and school functioning in early adolescence. Journal of
Educational Psychology, 92, 349-359.

Kupersmidt, J. B., DeRosier, M. E., & Patterson, C. P. (1995). Similarity as the basis
for children’s friendships: The roles of sociometric status, aggressive and
withdrawn behavior, academic achievement and demographic characteristics.
Journal of Social and Personal Relationships, 12, 439-452.

Ladd, G. W., & Kochenderfer-Ladd, B. (2002). Identifying victims of peer aggression
from early to middle childhood: Analysis of cross-informant data for
concordance, estimation of relational adjustment, prevalence of victimization,
and characteristics of identified victims. Psychological Assessment, 14, 74-96.

Ladd, G. W., Kochenderfer, B. J., & Coleman, C. C. (1997). Classroom peer
acceptance, friendship, and victimization: Distinct relational systems that
contribute uniquely to children’s school adjustment? Child Development, 68,
1181-1197.

Laird, R. D., Pettit, G. S., Dodge, K. A., & Bates, J. E. (1999). Best friendships, group
relationships, and antisocial behavior in early adolescence. Journal of Early
Adolescence, 19, 413-437.

Leung, M. C. (1998). A User Manual for SCM 4.0 (Windows version) [Computer
software]. Chapel Hill, NC: Center for Developmental Science.

Los Angeles Police Department. (2005). Statistical Digest: 2005. Retrieved January 8,
2008, from http://www.lapdonline.org/crime_maps_and_compstat/content_
basic_view/9098

Lutkus, A. D., Rampey, B. D., & Donahue, P. (2005). The nation’s report card: Trial
urban district assessment reading 2005 (NCES 2006-455). U.S. Department of
Education, National Center for Education Statistics. Washington, D.C.: U.S.
Government Printing Office.

McClelland, G. H., & Judd, C. M. (1993). Statistical difficulties of detecting
interactions and moderator effects. Psychological Bulletin, 114, 376-390.

McKenney, K. S., Pepler, D., Craig, W., & Connolly, J. (2006). Peer victimization and
psychosocial adjustment: The experiences of Canadian immigrant youth.
Electronic Journal of Research in Educational Psychology, 9, 1696-2095.

47
Menesini, E., Codecasa, E., Benelli, B., & Cowie, H. (2003). Enhancing children’s
responsibility to take action against bullying: Evaluation of a befriending
intervention in Italian middle schools. Aggressive Behavior, 29, 1-14.

Nishina, A., Juvonen, J., & Witkow, M. R. (2005). Sticks and stones may break my
bones, but names will make me feel sick: The psychosocial, somatic, and
scholastic consequences of peer harassment. Journal of Clinical Child and
Adolescent Psychology, 34, 37-48.

Olweus, D. (1978). Aggression in the schools: Bullies and whipping boys.
Washington, DC: Hemisphere Publishing Corporation.

Olweus, D. (1993). Bullying at school: What we know and what we can do.
Cambridge, MA: Blackwell.

Perry, D. G., Kusel, S. J., & Perry, L. C. (1988). Victims of peer aggression.
Developmental Psychology, 24, 807-814.

Rubin, D. B. (1987). Multiple imputation for nonresponse in surveys. New York, NY:
John Wiley & Sons, Inc.

Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art.
Psychological Methods, 7, 147-177.

Schwartz, D. (2000). Subtypes of victims and aggressors in children’s peer groups.
Journal of Abnormal Child Psychology, 28, 181-192.

Schwartz, D., Gorman, A. H., Dodge, K. A., Pettit, G. S., & Bates, J. E. (in press).
Friendships with peers who are low or high in aggression as moderators of the
link between peer victimization and declines in academic functioning. Journal
of Abnormal Child Psychology.

Schwartz, D., Gorman, A. H., Nakamoto, J., & Toblin, R. L. (2005). Victimization in
the peer group and children’s academic functioning. Journal of Educational
Psychology, 97, 425-435.

Schwartz, D., McFadyen-Ketchum, S., Dodge, K. A., Pettit, G. S., & Bates, J. E.
(1999). Early behavior problems as a predictor of later peer group
victimization: Moderators and mediators in the pathways of social risk.
Journal of Abnormal Child Psychology, 27, 191-201.

48
Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and non-experimental
studies: New procedures and recommendations. Psychological Methods, 7,
422-445.

Sirin, S. R. (2005). Socioeconomic status and academic achievement: A meta-analytic
review of research. Review of Educational Research, 75, 417-453.

Skinner, E. A., & Belmont, M. J. (1993). Motivation in the classroom: Reciprocal
effects of teacher behavior and student engagement across the school year.
Journal of Educational Psychology, 85, 571-581.

Smith, P. K., Ananiadou, K., & Cowie, H. (2003). Interventions to reduce school
bullying. The Canadian Journal of Psychiatry, 48, 591-599.

Sobel, M. E. (1982). Asymptotic confidence intervals for indirect effects in structural
equation models. In S. Leinhart (Ed.), Sociological methodology 1982 (pp.
290-312). San Francisco, CA: Jossey-Bass.

Steele, C. M. (1997). A threat in the air: How stereotypes shape the intellectual
identities and performance of women and African Americans. American
Psychologist, 52, 613-629.

Storch, E. A., Nock, M. K., Masia-Warner, C., & Barlas, M. E. (2003). Peer
victimization and social-psychological adjustment in Hispanic and African-
American children. Journal of Child and Family Studies, 12, 439-452.

Supple, A. J., Ghazarian, S. R., Frabutt, J. M., Plunkett, S. W., & Sands, T. (2006).
Contextual influences on Latino adolescent ethnic identify and academic
outcomes. Child Development, 77, 1427-1433.

Thomas, D. E., & Bierman, K. L. (2006). The impact of classroom aggression on the
development of aggressive behavior problems in children. Development and
Psychopathology, 18, 471-487.

Tolan, P. H., Guerra, N. G., & Montaini-Klovdahl, L. (1997). Staying out of harm’s
way: Coping and the development of inner-city children. In I. N. Sandler & S.
A. Wolchik (Eds.), Handbook of children’s coping with common stressors:
Linking theory, research, and interventions (pp. 453-479). New York: Plenum.

Wentzel, K. R. (1999). Social-motivational processes and interpersonal relationships:
Implications for understanding motivation at school. Journal of Educational
Psychology, 91, 76-97.

49
Wentzel, K. R. (2003). Sociometric status and adjustment in middle school: A
longitudinal study. Journal of Early Adolescence, 23, 5-28.

Wentzel, K. R., Barry, C. M., & Caldwell, K. A. (2004). Friendships in middle school:
Influences on motivation and school adjustment. Journal of Educational
Psychology, 96, 195-203.

Wentzel, K. R., & Caldwell, K. (1997). Friendships, peer acceptance, and group
membership: Relations to academic achievement in middle school. Child
Development, 68, 1198-1209.

Werner, N. E., & Crick, N. R. (2004). Maladaptive peer relationships and the
development of relational and physical aggression during middle childhood.
Social Development, 13, 495–514.

Abstract (if available)

Abstract This study utilized a multi-informant approach to investigate the concurrent association between peer victimization and school functioning in a sample of 135 Latino children (55 boys

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Mediators and moderators in the link between maternal psychological control and peer victimization for Hong Kong Chinese boys

PDF

Heterogeneity among adolescents with elevated depressive symptoms: distinct patterns of social and academic attributes

PDF

Dynamics of victimization, aggression, and popularity in adolescence

PDF

An examination of in-school and online protective factors for adolescent trajectories of online victimization and harassment over time

PDF

Emergent literacy skills and home literacy environment of Latino children who live in poverty

PDF

Friendship as a protective factor in adolescence

PDF

The home environment and Latino preschoolers' emergent literacy skills

PDF

Community violence exposure and children's subsequent rejection within the school peer group: the mediating roles of emotion dysregulation and depressive symptoms

PDF

An alternative capstone project: Evaluating the academic achievement gap for Latino English language learners in a high achieving school district

PDF

An alternative capstone project: bridging the Latino English language learner academic achievement gap in elementary school

PDF

The possible selves of Latino males participating in the Upward Bound Program

PDF

The role of social status and gender in the composition of bully-victim dyads in early adolescence

PDF

The relationship among the nurturance and monitoring dimensions of parenting, academic self-concept, and acculturation, in the academic achievement of Latino college students

PDF

Studying the impact of a career academy on Chicano/Latino students in a specific high school: a quantitative case study

PDF

Meta-accuracy as a moderator of the association between social experience and emotional adjustment during childhood

PDF

Goal orientation of Latino English language learners: the relationship between students’ engagement, achievement and teachers’ instructional practices in mathematics

PDF

The relationship between student perceptions of parental expectations, utility value, aptitude and English achievement among Asian American high school students

PDF

Untangling the developmental relations between depression and externalizing behavior among maltreated adolescents

PDF

Motivation and the meanings of health behavior as factors associated with eating behavior in Latino youth

PDF

The role of school climate in the mental health and victimization of students in military-connected schools