Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Dynamics of victimization, aggression, and popularity in adolescence
(USC Thesis Other)
Dynamics of victimization, aggression, and popularity in adolescence
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Running head: DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY
Dynamics of Victimization, Aggression, and Popularity in Adolescence
Sarah T. Malamut
A Dissertation Presented to the
Faculty of the USC Graduate School
(DEVELOPMENTAL PSYCHOLOGY)
Doctor of Philosophy
University of Southern California
December 2019
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY ii
TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS…….……….....................................................................................1
INTRODUCTION……………………….....................................................................................2
Subtypes of Victimization.................................................................................................3
Victimization, Aggression, and Popularity…....................................................................5
Self-Identified vs. Peer-Identified Victims........................................................................6
The Current Investigation..................................................................................................7
STUDY 1.......................................................................................................................................9
METHODS....................................................................................................................................9
Participants and Procedure................................................................................................9
Measures............................................................................................................................9
RESULTS/DISCUSSION……....................................................................................................11
Descriptive Statistics .......................................................................................................11
Prospective Associations between Popularity and Victimization....................................11
Self-Reported Victimization and Aggression..................................................................12
The Moderating Role of Popularity.................................................................................12
STUDY 2……...………………………......................................................................................14
METHODS..................................................................................................................................14
Participants and Procedure...............................................................................................14
Measures...........................................................................................................................15
RESULTS/DISCUSSION……....................................................................................................15
Descriptive Statistics .......................................................................................................15
Victim Types....................................................................................................................16
Differences in Popularity, Indirect Aggression, and Bullying.........................................17
Identifying Popular, Self-Identified Victims....................................................................17
GENERAL DISCUSSION…………………………...................................................................20
Identifying Popular Victims.............................................................................................20
Self-Reported Victimization, Popularity, and Aggression...............................................21
Strengths, Limitations, & Future Directions....................................................................22
Conclusions......................................................................................................................23
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY iii
REFERENCES............................................................................................................................24
TABLES AND FIGURES...........................................................................................................28
Table 1 ............................................................................................................................28
Table 2 ............................................................................................................................29
Table 3 ............................................................................................................................30
Figure 1 ...........................................................................................................................31
Table 4 ............................................................................................................................32
Table 5 ............................................................................................................................33
Table 6 ............................................................................................................................34
APPENDIX……………..............................................................................................................35
Appendix A: Popularity Moderated by Gender...............................................................35
Appendix B: Victimization and Popularity Moderated by Gender..................................36
Appendix C: Description of Measures for Post Hoc Analyses........................................37
Appendix D: Post hoc Comparison of Victim Types......................................................38
Appendix E: Model Comparisons in LPA- Depiction of BIC Values.............................39
Appendix F: Model Fit Indices for LPA..........................................................................40
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 1
Acknowledgements
There are many people that I would like to thank for their guidance and mentorship
throughout my time as a doctoral student. First, I would like to thank my advisor, David
Schwartz. Working with David has helped me grow professionally and personally, and I am
sincerely grateful for his unwavering support. I also would like to thank my incredible labmates
over the years- I could not have asked for a more collaborative or supportive lab. I would also
like to extend my gratitude to Jo Ann Farver, Frank Manis, John Monterosso, and Jeremy
Goldbach for serving on my dissertation committee and for their support throughout the
dissertation process.
I also extend sincere appreciation to Toon Cillessen, Tessa Lansu, and Yvonne van den
Berg for their contributions to my dissertation. Without them, this project would not have been
possible. I am very grateful for the time I spent working with them.
Last but not least, I am extremely thankful for my undergraduate mentors- Hongling Xie
and Molly Dawes. They have played a substantial role in my professional development, and I
have learned so much from them over the past seven years. Their continued guidance has been
an invaluable component of my graduate career.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 2
Introduction
In this paper, we present two short-term longitudinal studies that examine the dynamics
of victimization, popularity, and aggression in adolescence. Peer victimization is associated with
an array of social, emotional, behavioral, and academic difficulties for victims (e.g., Card &
Hodges, 2008). A large proportion of this research conceptualizes victims as socially
marginalized youth with low status, who are often targeted by their more powerful and higher-
status peers. Although these youth often are victimized, this perspective does not provide a
comprehensive understanding of victimization, as high-status youth are not just perpetrators but
also targets of aggression (e.g., Dawes & Malamut 2018; Malamut, Dawes, & Xie, 2018;
Prinstein & Cillessen, 2003). In the current investigation, we explored challenges that preclude
identification of high-status (i.e., popular) victims and examined the implications of their
victimization. Specifically, we investigated whether popular victims are more likely to exhibit
increases in aggression over time.
We focused on two factors that may explain why high-status victims are often not
identified: the form of victimization assessed and the informant (self or peers). To this end, our
first research goal was to examine whether popularity was differentially related to two forms of
indirect victimization (i.e., reputational victimization and exclusion). Our second objective was
to identify a subset of youth who are popular and perceive themselves to be victimized, but are
not viewed by peers as a victim (i.e., identified via self-report but not peer report).
The lack of information regarding high-status victims’ experiences is concerning for
several reasons. Popular adolescents can be highly motivated to retain their status (e.g., Dawes &
Xie, 2014). Subsequently, if popular youth are victimized, or perceive themselves to be
victimized, then they may respond with aggression to defend their status. Therefore, our final
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 3
research goal was to examine whether the victimization of high-status youth predicted
subsequent aggression and bullying.
High-status adolescents’ experiences being victimized may be an important, understudied
phenomenon that underlies the perpetuation of aggression in the peer group, particularly as status
becomes increasingly important in adolescence (LaFontana & Cillessen, 2010). Bullying
interventions are typically less effective (or even ineffective) in adolescence as compared to
childhood (e.g., Yeager, Dahl, & Dweck, 2018). Moreover, even at ages when interventions have
had success (e.g., in childhood), they are less successful at reducing the bullying behaviors of
popular youth than youth with average or low popularity (Garandeau, Lee, & Salmivalli, 2014).
Aggression and social status are deeply intertwined, particularly in adolescence (Cillessen &
Mayeux, 2004), which may account for the ineffectiveness of bullying interventions with both
older youth as well as younger, popular youth. Furthermore, not identifying high-status victims
renders our understanding of the prevalence of victimization, along with who is at risk for
victimization, incomplete.
Subtypes of Victimization
Successfully identifying high-status victims is, in part, dependent on the type of
aggression being measured. There are several theoretical perspectives that outline why some
high-status youth will be targets of aggression (see Dawes & Malamut, 2018 for a review). High-
status adolescents may be targeted by peers who want to take down a social competitor and/or
increase their own status (e.g., Andrews, Hanish, & Santos, 2017). Given that popular
adolescents have an assortment of social resources and are dominant in the peer group (Dawes &
Malamut, 2018), youth may be more likely to use indirect forms of aggression (i.e., covert or
“behind-the-back”; Archer & Coyne, 2005) against popular peers to avoid the risk of a direct
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 4
confrontation. However, indirect victimization is typically treated as a homogenous entity,
despite support that subtypes of indirect victimization occur at different rates and are
differentially related to other characteristics (e.g., Closson, Hart, & Hogg, 2017; Prinstein &
Cillessen, 2003).
There is some evidence of positive, concurrent associations between popularity and being
the target of a rumor or having mean things said behind one’s back (reputational victimization),
but less support for positive links between popularity and experiences such as being excluded
(Closson et al., 2017). However, measures of reputational victimization are often combined with
measures of exclusion, despite that these forms of aggression serve different functions and have
unique associations with popularity (Prinstein & Cillessen, 2003), which may hinder the
identification of high-status victims. Whereas there is support of concurrent associations between
popularity and reputational victimization, prospective relations have not been studied. Therefore,
it is unclear whether popularity may actually be a risk factor for certain types of victimization.
Reputational aggression may be used to target a high-status peer (e.g., social competition)
or a low-status peer (e.g., choosing an easy target). Therefore, consistent with past research, we
expected a curvilinear association between popularity and reputational victimization (Prinstein &
Cillessen, 2003). We hypothesized that high (and low) levels of popularity would be associated
with high levels of reputational victimization over time, as youth may use this form of aggression
in attempts to damage popular youth’s social standing or reputation, or against low-status youth
to establish social norms (Dawes & Malamut, 2018; Malamut et al., 2018; Prinstein & Cillessen,
2003). On the other hand, there are aspects of popularity (e.g., social resources, centrality) that
should generally be protective against other types of aggression, such as being excluded or
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 5
neglected. Therefore, we hypothesized that high popularity would be negatively associated with
being excluded.
Victimization, Aggression, and Popularity
Whereas we hypothesized that reputational victimization and exclusion would be
differently associated with popularity, we did not hypothesize any differences in popular youth’s
use of reputational aggression and exclusion toward others. In fact, youth who are popular often
use both forms of aggression against their peers (Prinstein & Cillessen, 2003). As summarized by
Archer and Coyne (2005), different terms (i.e., indirect aggression, relational aggression, and
social aggression) have been used to describe aggression that is covert or involves using the
social group to inflict harm. However, they found relatively few conceptual or empirical
differences between indirect aggression, relational aggression, and social aggression. Based on
Archer and Coyne’s (2005) recommendations and consistent with recent research (Closson et al.,
2017), we will use indirect aggression to refer to perpetrating behavior that encompasses both
exclusion and reputational aggression.
There is some evidence that aggression and victimization may have a cyclical relation
(e.g., Ostrov & Godleski, 2013). Moreover, Ferguson, Zimmer-Gembeck, and Duffy (2016)
found that the bidirectional association between aggression and victimization was moderated by
social status. They found that popular girls who were victimized became more aggressive 7
months later. This finding suggests that popular adolescents, who are already familiar with the
rewards of social status (e.g., social resources and visibility), may be sensitive to challenges to
their social standing (i.e., victimization) and therefore may become more aggressive in an
attempt to maintain their status. Whereas Ferguson and colleagues (2016) used peer nominations
to assess victimization, self-perceived victimization is likely particularly pertinent for subsequent
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 6
aggression, especially when an adolescent is popular. Specifically, we hypothesized that self-
reported victimization, rather than peer-reported victimization, would predict increases in
aggression at high levels of popularity. Given previous findings (e.g., Ferguson et al., 2016), we
also included peer nominations of victimization as a comparison.
Self-Identified vs. Peer-Identified Victims
Another reason high-status victims are underrepresented in the literature may be related
to the low correspondence between two of the most prevalent ways of assessing victimization
(i.e., self-reports and peer nominations). The concordance between self-reports and peer
nominations is generally weak to moderate (e.g., Dawes, Chen, Farmer, & Hamm, 2017; Scholte,
Overbeek, & Burk, 2013). Recently, research has utilized person-centered analyses to address
this issue, as this method identifies individuals who are a part of groups with unique adjustment
profiles (Scholte et al., 2013). Previous research using person-centered analyses has identified
four types of victims: 1. convergent victims (self- and peer- identified), 2. self-identified victims,
3. peer-identified victims, and 4. non-victims. Self-identified victims have typically been
conceptualized as youth who are “paranoid” (e.g., Graham & Juvonen, 1998) or struggling with
internalizing symptoms or poor self-perceptions (e.g., Dawes et al., 2017; Scholte et al., 2013).
In addition to cognitive biases, discrepancies in self- and peer- reports of victimization
may also occur when youth actually do experience victimization, but for some reason do not
have a reputation in the peer group as a victim. For example, adolescents who have high status
are still targets of aggression, but may not always have reputations as victims because they
appear to be doing well socially. Nonetheless, popular youth may still believe they have been
targeted, despite not being seen as a victim. As such, the self-identified subtype of victim may
include youth with high social status. In fact, self-identified victims appear not to have the same
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 7
social difficulties as convergent and peer-identified victims (e.g., social rejection, few friends;
Scholte et al., 2013). However, research has not yet tested whether self-identified victims differ
from the other groups with regard to popularity. Furthermore, it is unknown whether certain
victim types are more likely than others to be aggressive. Self-identified victims (particularly
those with high status) may feel threatened and subsequently become more aggressive.
We attempted to replicate the four subtypes of victims identified by previous research,
with a particular interest in self-identified victims. We hypothesized that self-identified victims
would be more popular than convergent and peer-identified victims. Next, our goal was to
identify a subset of self-identified victims who also have high status. We expected that being a
self-identified victim with high popularity would predict increases in aggression and bullying.
The Current Investigation
In sum, our overarching goal was to examine the associations of victimization,
popularity, and aggression over time. Despite growing evidence that youth with high status are
also targets of aggression (e.g., Dawes & Malamut, 2018), previous research often does not
identify these types of victims, perhaps due to the form of victimization assessed or because their
peers do not view them as victims. This is concerning, as youth who believe they are being
victimized and who also have the social resources to defend their status may subsequently
increase their aggression.
The current investigation addressed this gap in two longitudinal, multi-informant studies.
In Study 1, we investigated whether: (1) popularity was differentially related to changes in
reputational victimization and exclusion over time, and (2) popularity moderated the association
between (self-reported) indirect victimization and subsequent aggression. We expected high
popularity to predict increases in reputational victimization, but decreases in being excluded.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 8
Specificity in forms of victimization likely impacts whether youth identify their high-status peers
as victims. Whereas we expected the subtype of victimization to be essential for identifying
victims with high status via peer reports, we did not have a similar hypothesis for the association
between self-reported indirect victimization and aggression. In other words, we had no reason to
expect that self-reports of being a rumor victim would be differentially related to aggression than
self-reports of being excluded, as any form of victimization would result in feeling threatened or
self-perceptions as victimized. Therefore, we hypothesized that self-reported indirect
victimization (regardless of subtype) would predict increases in indirect aggression, particularly
for youth with high levels of popularity.
In Study 2, we built on Study 1 by using person-centered analyses (latent profile analysis)
to identify victim types based on self-reports and peer nominations of being bullied (e.g., Dawes
et al., 2017; Scholte et al., 2013). Whereas Study 1 focused on indirect forms of victimization
and aggression, we identified victim types in Study 2 based on self and peer reports of bullying,
consistent with past research (Dawes et al., 2017; Scholte et al., 2013). We expected a subset of
self-identified victims to also be high in popularity, and examined whether this subtype was
more likely to bully and use indirect aggression than the other victim subtypes.
In both studies, we explored potential gender differences. There is some support that boys
and girls generally engage in indirect aggression at similar rates (e.g., Salmivalli & Kaukiainen,
2004), whereas boys appear to be more likely to bully than girls (e.g., Sentse, Kretschmer, &
Salmivalli, 2015). Moreover, aggression may be differentially related to popularity for boys and
girls (e.g., Cillessen & Mayeux, 2004). Seminal research has also highlighted the importance of
using indirect aggression to jockey for power in popular girls’ cliques (e.g., Merten, 1997).
Therefore, we considered gender as a potential moderator.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 9
Study 1
Methods
Participants and Procedure
Study 1 was completed in collaboration with a high school in the greater Los Angeles
area. In spring 2016 (T1), 659 9
th
graders were invited to participate and 413 received positive
parental consent. Participants were 376 adolescents (Mage = 14.4; 55.6% girls) in the 9
th
grade
who assented to participate and were not absent during data collection. The ethnic/racial
composition of the sample was 29.2% Latino/Hispanic, 27.3% White, 9.8% Asian/Pacific
Islander, 2.1% African American, .3% American Indian, 28.2% mixed, and 3.2% not classified.
In spring 2017 (T2), a follow-up data collection was completed when participants were in 10
th
grade. Of the participants at T1, we retained almost the full sample (n = 374). Of these, 370
participants had full data for the items relevant to the current study.
At each wave, trained graduate and undergraduate research assistants administered the
measures to participants. The research assistants read out loud the standardized instructions to
the participants and reiterated the confidentiality of their responses. Research assistants informed
participants that they could stop participating at any time, and were also available to answer any
questions. This project was approved by the university’s Internal Review Board (IRB # UP-15-
00579-CR002 “School Adjustment”).
Measures
Victimization (self-report). Participants completed a self-report questionnaire (“My Day
at School”) that assesses youth’s experiences of victimization. For the current investigation, we
focused on five items pertaining to indirect victimization (e.g., “try to keep others from liking
you”). As noted earlier, our expectation was that any perception of victimization (regardless of
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 10
subtype) would be associated with elevated aggression for popular youth. Accordingly, for our
analyses we used the averaged participants’ scores on the five items. In addition to our
theoretical basis for averaging the five items, there was also empirical support as the items had
high reliability (Cronbach’s α = .85) and an exploratory factor analysis indicated a one factor
solution.
Victimization (peer nomination). Peer nominations were also used to assess youth’s
experiences with victimization. Participants were given a random list of approximately 50
participating grademates. Participants then indicated which peers fit a series of descriptors, with
unlimited nominations. Given our interest in differentiating between victimization experiences
(e.g., Closson et al., 2017; Prinstein & Cillessen, 2003), we opted to include separate indices for
reputational victimization (“students who get mean things said about them”) and exclusion
(“students that get left out of activities, excluded, or ignored when other students are trying to
hurt their feelings”).
Aggression (peer nominations). We measured youth’s indirect aggression using peer
nominations, using the same procedure described above. Participants were asked to nominate
“students that gossip about other students” and “students that try to be mean to other students by
ignoring them or excluding them” (r = .75, p < .0001 at T1; r = .70, p < .0001 at T2).
Nominations on these items were averaged and standardized within list.
Popularity (peer nominations). Peer nominations were also used to assess participants’
popularity. Youth nominated their peers who were “most popular” and “least popular”.
Standardized nominations were calculated for each item, and a difference score was computed
(most popular – least popular) and restandardized (Cillessen & Marks, 2011).
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 11
Results and Discussion
Descriptive Statistics
Means, standard deviations, and correlations among study variables are shown in Table 1.
All continuous variables were mean centered. Girls reported higher levels of self-reported
victimization at T1 and indirect aggression at both time points than boys. There were no
significant gender differences in peer-nominated victimization (i.e., exclusion, reputational
victimization) or popularity.
Self-reported victimization (T1) was positively associated with reputational victimization
(T1 and T2), indirect aggression (T1 and T2), and popularity (T1). Moreover, popularity (T1)
was positively related to indirect aggression (T1 and T2) and reputational victimization (T1 and
T2), and negatively related to exclusion (T1 and T2). Indirect aggression and reputational
victimization, but not exclusion, were positively correlated at both time points. Indirect
aggression and both forms of peer nominated victimization were stable from T1 to T2.
Prospective Associations between Popularity and Peer-Nominated Victimization
Our first goal was to test the hypothesis that high levels of popularity would be associated
with high levels of reputational victimization, but low levels of exclusion, over time. We
conducted separate linear regressions for reputational victimization and exclusion. In each
model, the form of victimization at T2 was predicted by popularity and T1 victimization, while
controlling for gender (Table 2).
For the model predicting T2 reputational victimization, we included a quadratic
popularity term, given our hypothesis that there would be a curvilinear effect of popularity (i.e.,
an association at low and high levels of popularity). The overall model was significant, F(4, 365)
= 71.46, p < .001, R
2
= .43. The quadratic popularity term was significant (β = .19, p < .001), but
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 12
the linear popularity term was not (β = .04, p = .272). As indicated by the positive coefficient
term, both high and low levels of T1 popularity were associated with elevated T2 reputational
victimization. Next, we tested whether popularity significantly predicted exclusion. The model
predicting T2 exclusion was significant, F(3, 366) = 49.78, p < .001, R
2
= .28. There was a
negative association between T1 popularity and exclusion at T2 (β = -.10, p = .028). There were
no significant gender differences in either model (see Appendix A).
Self-Reported Victimization, Popularity, and Aggression
To examine whether self-reported indirect victimization predicts increases in indirect
aggression, along with whether this association is moderated by popularity, we conducted linear
regressions. As before, separate models were conducted with reputational victimization and
exclusion. For each set of analyses, we first examined the main effects of self-reported
victimization, popularity, and peer-nominated victimization at T1 on indirect aggression at T2,
while controlling for gender and indirect aggression at T1. The model including reputational
victimization was significant, F(5, 364) = 109.51, p < .001, R
2
= .60. Indirect aggression was
stable from T1 to T2 (β = .53, p < .001). High levels of self-reported victimization (β = .07, p =
.045), reputational victimization (β = .10, p = .017), and popularity (β = .22, p < .001) were all
independently associated with T2 indirect aggression (Table 3a, Model 1). The model including
exclusion revealed a similar pattern of findings, F(5, 364) = 107.63, p < .001, R
2
= .59 (see Table
3b, Model 1), with one exception. Exclusion did not predict indirect aggression (β = .05, p =
.168).
The Moderating Role of Popularity
Next, we examined whether T1 popularity moderated the association between T1 self-
reported victimization and T2 indirect aggression. In this model, we added two-way interactions
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 13
between self-reported victimization and popularity, and reputational victimization and
popularity, F(7, 362) = 80.46, p < .001, R
2
= .60 (see Table 3a, Model 2). The effect of self-
reported victimization on indirect aggression was qualified by popularity, (β = .09, p = .008). For
participants with low popularity at T1, self-reported indirect victimization was not associated
with indirect aggression at T2 (simple slopes test, t = -0.40, p = .69, Figure 1). However, at high
levels of T1 popularity, self-reported indirect victimization was associated with high T2 indirect
aggression (simple slopes test, t = 3.29, p = .001). The two-way interaction between popularity
and peer-reported victimization was not significant. There was also not a significant gender
moderation (see Appendix B). The same pattern of findings emerged when including exclusion
in the model (see Table 3b, Model 2)
1
.
In sum, Study 1 supports that high popularity and victimization are not mutually
exclusive (e.g., Closson et al., 2017; Dawes & Malamut, 2018; Prinstein & Cillessen, 2003). In
fact, high popularity was a risk factor for increases in reputational victimization over time.
Importantly, the association between self-perceptions of victimization and subsequent aggression
was exacerbated by popularity. As shown in Figure 1, at low levels of self-reported indirect
victimization, high and low levels of popularity were similarly associated with indirect
aggression one year later. Given past research demonstrating the strong association between
popularity and aggression (e.g., Cillessen & Mayeux, 2004; Prinstein & Cillessen, 2003), this
finding highlights the importance of self-perceived victimization in predicting future aggression.
Moreover, our findings support that some youth with high status experience victimization,
despite not being viewed by their peers as a victim.
1
We conducted sensitivity analyses, replicating our analyses with a combined measure of direct and indirect self-
reported victimization, F(7, 362) = 78.83, p < .001, R
2
= .60. Again, there was a significant interaction between self-
reported victimization and popularity (β = .07, p = .047).
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 14
Study 2
Despite the strengths of Study 1, the variable-centered analyses used do not allow us to
directly examine whether a subset of self-identified victims are in fact youth with high status. By
using person-centered analyses, however, we can first identify victim types then assess whether
some self-identified victims have high popularity and aggression. This line of inquiry can inform
whether the discordance between self-reports and peer nominations is partially driven by high-
status victims. To this end, we used latent profile analysis to identify subtypes of victims and
compared groups on popularity, aggression, and bullying. We then examined whether being a
self-identified victim with high popularity predicted increases in aggression and bullying.
Methods
Participants and Procedure
For Study 2, participants were recruited as part of the Kandinsky Longitudinal Study
(KLS), which began in 2010 to identify youth at risk for socio-emotional difficulties (Stoltz,
Cillessen, van den Berg, & Gommans, 2016). The current investigation includes data from
participants during waves 6 and 7 (i.e., years 2015 and 2016). At T1, Participants were 1004
Dutch adolescents in the first four years of secondary school (grades 7-10; 50.2% girls, Mage =
14.09, SD = 1.29) with complete data.
The head of the school provided parents with a detailed letter describing the goals and
procedure of the data collection. No parents objected to the participation of their son or daughter.
Adolescents were also informed of the details of the study and were asked for active assent at
each assessment. No students declined to participate at any stage of the assessment. This project
was approved by the Institutional Review Board of Radboud University (Protocol Number:
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 15
ECG2012-2505-038; Project Title: “Sociometry as a method to measure social relationships
among children and adolescents”).
Measures
Victimization (self-report). Participants completed the victim scale of the extended
Olweus’ Bully-Victim questionnaire (Solberg & Olweus, 2003). This scale consisted of six items
(e.g., “how often have other students ignored you”) rated on a scale ranging from 1 (“never”) to
5 (“several times a week”). This scale has typically been operationalized as a unidimensional
assessment; a pattern that is consistent with our own factor analysis. Cronbach’s ! was .69 at T1.
Victimization (peer nomination). Youth’s experiences of victimization were also
measured using peer nominations. Participants were given a list of their classmates and asked to
nominate which classmate fit various descriptors. To identify victims of bullying, participants
were asked “who in your class are bullied by others?”.
Aggression (peer nomination). As with Study 1, indirect aggression was assessed with
two peer nomination items: “who from your class say nasty things or gossip about others?” and
“who from your class exclude others or ignore others?”. Participants also reported “who from
your class bully others?” to identify perpetrators of bullying.
Popularity (peer nominations). Peer nominations were also used to assess indicators of
participants’ social standing. Participants nominated peers who were “most popular” and “least
popular”, and a composite score was calculated using the same procedure as Study 1.
Results and Discussion
Descriptive Statistics
Means, standard deviations, and correlations among study variables are presented in
Table 4. Boys were more popular and bullied at a higher rate than girls. Girls were reported to
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 16
use more indirect aggression than boys. There were no gender differences in self-reported or peer
nominated victimization.
There was a positive, but modest, correlation between self-reported victimization and
peer nominated victimization at T1. Self-reported victimization was also positively associated
with indirect aggression, although this association was small in magnitude. Peer nominated
victimization was negatively associated with all other variables (with the exception of self-
reported victimization and T2 bullying). Popularity (T1), indirect aggression (T1 and T2), and
bullying (T1 and T2) were all positively correlated.
Victim Types
To identify distinct victim groups, we conducted a series of latent profile analyses (LPA)
using tidyLPA in R (Rosenberg, Beymer, Anderson, & Schmidt, 2018). LPA compares
participants on continuous variables to assign them into mutually exclusive groups. Fit is
determined by comparing models on various statistical information criteria, with the Bayesian
information criterion (BIC) considered the most accurate indicator of the correct number of
classes (Nylund, Asparouhov, & Muthén, 2007). A comparison of BIC values indicated that a
model with varying variances and covariances fixed to zero with two or three profiles would fit
best (i.e., lowest BIC; see Appendix E and F for full model comparisons). Using a bootstrap
likelihood ratio test which tests model fit between k-1 and k models (McLachlan & Peel, 2000;
Nylund et al., 2007), we found that a three class model significantly improved model fit
compared to the two class model (p = .001). The overall entropy value for the three class model
was .90, indicating that the three groups included homogenous individuals.
Non-victims (n = 442) were low on both forms of victimization (Ms < -.36). Convergent
victims (n = 211) scored high (> ½ SD above the mean) on both self-reported and peer-reported
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 17
victimization (Ms = .62 and 1.34, respectively). Of particular interest to the current investigation
was self-identified victims. We identified a group of victims, average/high self-identified victims
(n = 351) who, on average, scored approximately a ½ SD above the mean on self-reported
victimization (M = .48), but low on peer-reported victimization (M = -.36). Of note, we did not
identify a “peer-identified” victim type. Therefore, contrary to our expectations, we did not
replicate the four victim groups identified by past research using LPA (e.g., Dawes et al., 2017;
Scholte et al., 2013)
2
. However, given our focus on self-identified victims, our analytic plan
remained the same. We also explored gender differences in victim groups, and there were
significant differences, χ
2
(2, N = 1004) = 6.19, p = .045. Boys were more likely than girls to be
convergent victims (Ns = 121 and 90, respectively).
Differences in Popularity, Indirect Aggression, and Bullying
We tested the hypothesis that self-identified victims would have elevated levels of
popularity, indirect aggression, and bullying, as compared to the other groups, by conducting
one-way ANOVAs with Tukey HSD post-hoc tests. There were significant differences between
the groups on popularity, indirect aggression, and bullying, F(2, 1001) > 5.36, ps < .005. As
hypothesized, self-identified victims were more popular and engaged in more indirect aggression
and bullying than convergent victims (Table 5). Self-identified victims were also reported as
more indirectly aggressive by peers than non-victims.
Identifying Popular, Self-identified Victims
Next, we identified a subset of self-identified victims who were popular (i.e., ½ SD above
2
Given that our victim types did not perfectly align with past research, we did a post hoc comparison of the three
victim types on an assortment of psychosocial variables (see Appendix C for full description of measures) to test
whether they were consistent with past findings. Consistent with prior investigations, convergent victims
experienced both internalizing and social difficulties, as compared to self-identified victims (Appendix D). In
general, self-identified victims did not differ from non-victims on social outcomes, but did exhibit more
internalizing symptoms than non-victims.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 18
the mean). Of the 351 self-identified victims, 117 were classified as popular victims. We
conducted independent sample t-tests to explore whether popular self-identified victims and all
other self-identified victims differed on self-reported or peer nominated victimization. There
were no significant differences in self-reported victimization between popular self-identified
victims (M = .55, SD = .69) and all other self-identified victims (M = .44, SD = .63), t = -1.37, p
= .171. There was a small, but significant, difference for peer nominated victimization, such that
popular self-identified victims (M = -.37, SD = .11) were nominated less often than all other self-
identified victims (M = -.35, SD = .10), t = 1.98, p = .049.
To test whether being a popular self-identified victim at T1 predicted increases in indirect
aggression and bullying at T2, compared to the other victim types, we conducted separate
ANCOVAs, controlling for gender and the outcome variable at T1 (Table 6). There were
significant differences between victim type on T2 indirect aggression (F(5, 714) = 58.14, p <
.001, R
2
= .29) and T2 bullying (F(5, 702) = 44.02, p < .001, R
2
= .23). Bonferroni post-hoc tests
on adjusted means revealed that popular self-identified victims (M = .32) were viewed by peers
as increasing in indirect aggression, compared to non-victims (M = .06, p = .048), convergent
victims (M = -.14, p < .001), and self-identified victims with lower status (M = -.09, p = .001).
Popular self-identified victims (M = .26) also bullied at T2 significantly more than convergent
victims (M = -.07, p = .048) and self-identified victims with lower status (M = -.09, p = .02), but
not non-victims (M = .03, p = .22).
In Study 2, we used person-centered analyses to test whether the pattern found in Study 1
(the confluence of self-reported victimization and popularity predicting aggression) may underlie
the discordance between self- and peer- reports of victimization. Contrary to our expectations,
we did not replicate the four victim types found by past research using LPA (e.g., Dawes et al.,
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 19
2017; Scholte et al., 2013). Specifically, we did not find peer-identified victims, and the sizes of
the victim types differed from past research. However, our post hoc analyses demonstrated that
the three victim types shared similar characteristics of self-identified victims, convergent
victims, and non-victims found in past research (see Footnote 2 and Appendix D).
A persistent finding in the literature is the relatively low correspondence between self-
reports and peer nominations of victimization (e.g., Scholte et al., 2013). Typically, this
discrepancy has been considered a product of some youth’s social or cognitive biases (e.g.,
Rosen, Milich, & Harris, 2007). Although this is likely true of some self-identified victims, the
results of Study 2 support that the discordance is also due to a subset of youth who are victimized
but who also are popular.
Notably, the pattern of correlations between self-reported victimization, popularity, and
aggression were considerably different in Study 2 as compared to Study 1. For example, self-
reported victimization was strongly and positively associated with both popularity and indirect
aggression in Study 1 but not Study 2. However, the overall pattern of findings (popular self-
identified victims being more aggressive) was consistent in both studies. This is noteworthy, as
the two studies included samples from very different contexts (urban Southern California and the
Netherlands). Furthermore, it highlights a benefit of using person-centered analyses, as variables
may be differentially related to one another for different individuals (Laursen & Hoff, 2006).
That is, in Study 2, self-reported victimization was not strongly associated with popularity or
aggression for the full sample, but youth with average to high levels of self-reported
victimization (and low levels of peer nominated victimization) were typically more popular and
aggressive than their peers.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 20
General Discussion
The main goals of the current investigation were to examine (1) factors that influence
identifying popular victims and (2) whether popularity moderates the association between self-
reported victimization and subsequent aggression. Taken together, the results from these studies
add to growing evidence that some youth who are victimized have high status and that
identifying these victims is dependent both on the form of victimization assessed, as well as the
informant (e.g., Dawes & Malamut, 2018). Moreover, in both studies, we found that the interplay
of self-perceived victimization and popularity predicted increases in aggression. Of note, our
findings were not moderated by gender, suggesting that the associations between victimization,
popularity, and aggression may be largely the same for boys and girls.
Identifying Popular Victims
Consistent with past findings, we found that high levels of popularity were associated
with reputational victimization but not exclusion (e.g., Closson et al., 2017; Prinstein &
Cillessen, 2003). As an extension beyond existing cross-sectional analyses, our results
demonstrated that high popularity also predicted increases in reputational victimization over
time. To our knowledge, this is one of the first studies to indicate that high popularity may be a
risk factor for certain types of victimization. This finding also suggests that one reason high-
status victims may not be identified is due to lack of specificity in measures. Reputational
victimization may be more often used against popular youth than other forms of aggression as it
is easier for the perpetrator to hide his/her identity (e.g., Closson et al., 2017). On the other hand,
popular adolescents have social resources and power, which may make it more difficult for their
peers to exclude them from activities.
We also found that self-identified victims were more popular than convergent victims.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 21
Furthermore, a third of self-identified victims could be classified as high in popularity (i.e., ½ SD
above the mean). Compared to less popular self-identified victims, popular self-identified
victims were less often nominated by peers as a victim. Together, this supports that there are
popular youth who perceive themselves to be victimized yet are not viewed by their peers as a
victim, which likely contributes to the low concordance between self- and peer- reported
victimization. In sum, we found evidence supporting that popular victims may not be readily
identified due to both the form of victimization assessed (e.g., general victimization) and the
informant (i.e., self vs. peer). Our findings further highlight youth at risk for victimization who
may typically go unnoticed (Dawes & Malamut, 2018).
Self-Reported Victimization, Popularity, and Aggression
As hypothesized in Study 1, the association between self-reported victimization and
subsequent aggression was moderated by popularity. At high levels of popularity, self-reported
victimization predicted higher levels of indirect aggression over time. Notably, there was not a
similar effect for peer nominated victimization. Moreover, in Study 2, self-identified victims
bullied more than convergent victims, and were more indirectly aggressive than convergent
victims and non-victims. These studies suggest that perceptions of being victimized are a
stronger predictor of aggression than having a reputation of being victimized. Furthermore, being
a popular self-identified victim predicted increases in bullying and indirect aggression over time.
Taken together, both studies demonstrate that youth who are successful socially (i.e.,
popular) are not immune to victimization. Our studies suggest that popular youth’s victimization
is a risk factor for subsequent indirect aggression and bullying. If youth report victimization and
have social resources (i.e., popularity), then this contributes to increases in aggression. Given
popular youth’s elevated influence (Dijkstra & Gest, 2015), their subsequent increases in
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 22
aggression may influence how aggression is viewed by the peer group. Therefore, popular
youth’s victimization may contribute to a cycle of aggression. Moreover, their victimization may
be an understudied reason why bullying interventions are less effective in adolescence and for
popular children.
Strengths, Limitations, & Future Directions
The current investigation has many methodological and theoretical strengths. First, it
consisted of two longitudinal studies with samples from two different countries, which supports
the external validity of our findings. Second, it adds to a growing call for multi-informant
approaches (e.g., Dawes et al., 2017; Scholte et al., 2013), and provides additional insight into
the discordance between self and peer reports of victimization. Third, it utilized both variable-
centered and person-centered analyses, which allowed us to examine the general impact of self-
reported victimization and popularity on aggression, as well as individual differences (Laursen &
Hoff, 2006). Lastly, it highlights a concerning pattern: self-perceived victimization predicts
increased aggression and bullying over time, especially for popular adolescents.
Still, there were limitations that should be noted. There are other factors besides
popularity that may be important in understanding the link between self-perceived victimization
and aggression. For example, recent research has highlighted the role of popularity goal (i.e.,
how much one strives to be popular) in moderating the relation between popularity and
aggression (Dawes & Xie, 2014). It is possible that only popular self-identified victims who also
value being popular will show increases in aggression. Future research should examine how
social goals influence this association.
Although we had theoretical reasons to focus on the associations between popularity,
indirect aggression, reputational victimization and exclusion in Study 1, it is also important to
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 23
consider other forms of victimization (e.g., Card & Hodges, 2008). Our sensitivity analysis in
Study 1, however, indicated that our findings replicated when using a combined measure of
indirect and direct self-reported victimization (see Footnote 1). Moreover, we found similar
associations between self-perceived victimization, popularity, and subsequent aggression in
Study 1 and Study 2, despite using different measures of self-reported victimization.
Conclusion
In conclusion, our findings highlight that at least a subset of self-identified victims are
youth who have high status. Their elevated popularity may generally preclude peers from
viewing them as victims. Our results support using measures of victimization with high
specificity, as reputational victimization, but not exclusion, was positively associated with
popularity. Furthermore, the current investigation found that self-reported victimization and
popularity uniquely interact to predict aggression. Although we focused on aggression, it is also
important to understand how high-status youth’s victimization impacts their psychosocial
adjustment (e.g., lowered self-esteem). Popularity is generally considered protective against
internalizing problems (e.g., Litwack, Wargo Aikins, & Cillessen, 2012); however, this may not
be true for popular youth who feel victimized. Our findings highlight the need to investigate
popular youth’s experiences of victimization for a comprehensive understanding of both
victimization and aggression in the peer group.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 24
References
Andrews, N. C. Z., Hanish, L. D., & Santos, C. E. (2017). Does an aggressor's target choice
matter? Assessing change in the social network prestige of aggressive youth.
Aggressive Behavior, 43, 364–374. doi:10.1002/ab.21695
Archer, J., & Coyne, S. M. (2005). An integrated review of indirect, relational, and social
aggression. Personality and Social Psychology Review, 9, 212–230.
doi:10.1207/s15327957pspr0903_2
Card, N., & Hodges, E. (2008). Peer victimization among schoolchildren: Correlations, causes,
consequences, and considerations in assessment and intervention. School Psychology
Quarterly, 23, 451–461. doi:10.1037/a0012769
Cillessen, A. H. N., & Marks, P. E. L. (2011). Conceptualizing and measuring popularity. In A.
H. N. Cillessen, D. Schwartz & L. Mayeux (Eds.), Popularity in the peer system (pp. 25– 56). New York: Guilford Press.
Cillessen, A. H. N., & Mayeux, L. (2004). From censure to reinforcement: Developmental
changes in the association between aggression and social status. Child Development,
75, 147–163. doi:10.2307/3696572
Closson, L. M., Hart, N. C., & Hogg, L. D. (2017). Does the desire to conform to peers moderate
links between popularity and indirect victimization in early adolescence? Social
Development, 26, 489–502. doi:10.1111/sode.12223
Dawes, M., Chen, C.-C., Farmer, T. W., & Hamm, J. V. (2017). Self- and peer-identified victims
in late childhood: Differences in perceptions of the school ecology. Journal of Youth and
Adolescence, 46, 2273-2288. doi:10.1007/s1096 4-017-0688-2
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 25
Dawes, M., & Malamut, S. (2018). No one is safe: Victimization experiences of high-status
youth. Adolescent Research Review, 1-21. doi:10.1007/s40894-018-0103-6
Dawes, M., & Xie, H. (2014). The role of popularity goal in early adolescents’ behaviors
and popularity status. Developmental Psychology, 50, 489-497. doi:10.1037/a0032999
Dijkstra, J. K., & Gest, S. D. (2015). Peer norm salience for academic achievement, prosocial
behavior, and bullying: Implications for adolescent school experiences. The Journal of
Early Adolescence, 35, 79-96. doi: 10.1177/0272431614524303
Ferguson, S., Zimmer-Gembeck, M. J., & Duffy, A. L. (2016). A longitudinal study of relational
aggression and victimization in early adolescence: Gender differences in the moderating
effects of social status. Journal of Relationships Research, 7, 1–11.
doi:10.1017/jrr.2016.9
Garandeau, C. F., Lee, I. A., & Salmivalli, C. (2014). Differential effects of the KiVa anti
bullying program on popular and unpopular bullies. Journal of Applied Developmental
Psychology, 35, 44–50. doi10.1016/j.appdev.2013.10.004
Graham, S., & Juvonen, J. (1998). Self-blame and peer victimization in middle school: An
attributional analysis. Developmental Psychology, 34, 587–599. doi:10.1037/0012-
1649.34.3.587
Laursen, B., & Hoff, E. (2006). Person-centered and variable-centered approaches to
longitudinal data. Merrill Palmer Quarterly, 52, 377−389.
Litwack, S. D., Aikins, J. W., & Cillessen, A. H. N. (2012). The distinct roles of sociometric and
perceived popularity in friendship: Implications for adolescent depressive affect and self-
esteem. Journal of Early Adolescence, 32, 226 –251. doi:10.1177/0272431610387142
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 26
Malamut, S. T., Dawes, M., & Xie, H. (2018). Characteristics of rumors and rumor victims in
early adolescence: Rumor content and social impact. Social Development, 1-18.
doi:10.1111/sode.12289
McLachlan, G., & Peel, D. (2000). Finite mixture models. New York: Wiley.
Merten, D. E. (1997). The meaning of meanness: Popularity, competition, and conflict among
junior high school girls. Sociology of Education, 70, 175–191. doi:10.2307/26732 07
Nylund, K., Asparouhov, T., & Muthén, B. (2007). Deciding on the number of classes in latent
class analysis and growth mixture modeling: A monte carlo simulation study. Structural
Equation Modeling: A Multidisciplinary Journal, 14, 535–569.
doi:10.1080/10705510701575396
Ojanen, T., & Findley-Van Nostrand, D. (2014). Social goals, aggression, peer preference, and
popularity: Longitudinal links during middle school. Developmental Psychology, 50,
2134-2143. doi:10.1037/a0037137
Ostrov, J.M., & Godleski, S.A. (2013). Relational aggression, victimization, and adjustment
during middle childhood. Development and Psychopathology, 25, 801–815.
doi:10.1017/S0954579413000187
Prinstein, M., & Cillessen, A. (2003). Forms and functions of adolescent peer aggression
associated with high levels of peer status. Merrill-Palmer Quarterly, 49, 310–342.
doi:10.1353/mpq.2003.0015
Rosen, P. J., Milich, R., & Harris, M. J. (2007). Victims of their own cognitions: Implicit social
cognitions, emotional distress, and peer victimization. Journal of Applied Developmental
Psychology, 28, 211–226. doi:10.1016/j.appdev.2007.02.001
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 27
Rosenberg, J. M., Beymer, P. N., Anderson, D. J., & Schmidt, J. A. (2018). tidyLPA: An R
package to easily carry out latent profile analysis (LPA) using open-source or commercial
software. Journal of Open Source Software, 3, 978. doi:10.21105/joss.00978
Salmivalli, C, & Kaukiainen, A. (2004). "Female aggression" revisited: Variable- and person-
centered approaches to studying gender differences in different types of aggression.
Aggressive Behavior, 30, 158-163. doi: 10.1002/ab.20012
Scholte, R., Burk, W., & Overbeek, G. (2013). Divergence in self- and peer-reported
victimization and its association to concurrent and prospective adjustment. Journal of
Youth and Adolescence, 42, 1789–1800. doi:10.1007/s10964-012-9896-y
Sentse, M., Kretschmer, T., & Salmivalli, C. (2015). The longitudinal interplay between
bullying, victimization, and social status: Age-related and gender differences. Social
Development, 24, 659–677. doi:10.1111/sode.12115
Solberg, M. E., & Olweus, D. (2003). Prevalence estimation of school bullying with the Olweus
Bully/victim questionnaire. Aggressive Behavior, 29, 239-268. doi:10.1002/ab.10047
Stoltz, S., Cillessen, A., van den Berg, Y., & Gommans, R. (2016). Popularity differentially
predicts reactive and proactive aggression in early adolescence. Aggressive Behavior, 42,
29-40. doi:10.1002/ab.21603
Yeager, D. S., Dahl, R. E., & Dweck, C. S. (2018). Why interventions to influence adolescent
behavior typically fail but could succeed. Perspectives on Psychological Science, 13,
101-122. doi:10.1177/1745691617722620
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 28
Tables and Figures
Table 1
Correlations, Means, and Independent Sample T-Tests (Study 1)
1
2
3
4
5
6
7
8
M
(SD)
M
(SD)boys
(N = 161)
M
(SD)girls
(N = 209)
t
1. Self-reported victimization (T1) --
1.62
(.72)
1.46
(.56)
1.73
(.81)
-3.78
***
2. Exclusion (T1) .07
-- .01
(1.01)
.11
(1.07)
-.07
(.95)
1.67
3. Reputational victimization (T1) .27
***
.51
***
-- -.01
(.98)
-.01
(.92)
-.01
(1.04)
-.022
4. Indirect aggression (T1) .27
***
-.03 .50
***
--
-.03
(.93)
-.23
(.71)
.12
(1.04)
-3.79
***
5. Popularity (T1) .17
**
-.34
***
.10
*
.62
***
-- -.03
(.97)
-.13
(.95)
.05
(.98)
-1.81
6. Exclusion (T2) .04 .53
***
.33
***
-.03 -.28
***
-- -.01
(.95)
.10
(1.04)
-.10
(.86)
1.92
7. Reputational victimization (T2) .19
***
.38
***
.64
***
.41
***
.12
*
.47
***
-- .01
(.99)
.04
(1.01)
-.01
(.98)
.46
8. Indirect aggression (T2) .30
***
-.04 .40
***
.74
***
.58
***
-.01 .51
***
-- -.02
(.96)
-.29
(64)
.19
(1.10)
-5.30
***
Note. *p < .05. **p < .01. ***p < .001.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 29
Table 2
Predicting T2 Peer-Nominated Victimization from T1 Popularity
Reputational Victimization Exclusion
b SE b SE
Gender .07 .08 .10 .09
Peer-nominated victimization T1 .56
***
.04 .46
***
.04
Popularity (linear term) T1 .05
.04 -.10
*
.05
Popularity (quadratic term) T1 .13
***
.03 -- --
Note. *p < .05. **p < .01. ***p < .001.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 30
Table 3
Predicting T2 Indirect Aggression from T1 Victimization and T1 Popularity
Panel A
Model 1 Model 2
b SE b SE
Gender -.23
***
.07 -.23
**
.07
Indirect aggression T1 .53
***
.05 .53
***
.05
Popularity T1 .22
***
.04 .21
***
.04
Self-reported victimization T1 .09
*
.05 .09
†
.05
Reputational victimization T1 .10
*
.04 .08
†
.04
Popularity X Self-reported victimization .12
**
.05
Popularity X Reputational victimization .00 .03
Panel B
Model 1 Model 2
b SE b SE
Gender -.22
**
.07 -.22
**
.07
Indirect aggression T1 .59
***
.05 .57
***
.05
Popularity T1 .21
***
.05 .22
***
.05
Self-reported victimization T1 .11
*
.05 .10
*
.05
Exclusion T1 .05 .04 .03 .04
Popularity X Self-reported victimization .13
**
.04
Popularity X Exclusion -.03 .03
Note. N=370. In Panel A, the model included peer-nominated reputational victimization. In Panel B, the model
included peer-nominated exclusion.
†
p < .10. *p < .05. **p < .01. ***p < .001.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 31
Figure 1
T1 Popularity Moderates the Association Between T1 Self-Reported Victimization and T2
Indirect Aggression
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
0.5
Low Self-Reported Victimization High Self-Reported Victimization
INDIRECT AGGRESSION (T2)
Low Popularity (T1) High Popularity (T1)
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 32
Table 4
Correlations, Means, and Independent Sample T-Tests (Study 2)
1
2
3
4
5
6
7
M
(SD)
M
(SD)boys
(N = 500)
M
(SD)girls
(N = 504)
t
1. Self-reported victimization (T1) -- -.01
(1.00)
.03
(1.01)
-.04
(.98)
1.28
2. Peer nominated victimization (T1) .23
***
-- .00
(.98)
.03
(.97)
-.02
(.99)
.74
3. Popularity (T1) -.06
-.49
***
-- .02
(1.65)
.22
(1.65)
-.19
(1.64)
3.92
***
4. Indirect aggression (T1) .06
*
-.12
***
.46
***
--
.00
(.90)
-.12
(.83)
.12
(.95)
-4.28
***
5. Bullying (T1) .05 -.09
**
.42
***
.66
***
-- .01
(1.00)
.15
(1.10)
-.14
(.86)
4.66
***
6. Indirect aggression (T2) .04 -.13
***
.44
***
.52
***
.43
***
-- .01
(.89)
-.11
(.80)
.13
(.95)
-3.68
***
7. Bullying (T2) .02 -.03 .28
***
.34
***
.47
***
.58
***
-- .01
(.98)
.18
(1.12)
-.16
(.80)
4.54
***
Note. At T1, N = 1004. Of these, 720 had data on T2 indirect aggression (356 boys, 364 girls) and 708 had data on T2 bullying (349 boys, 359).
*p < .05. **p < .01. ***p < .001.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 33
Table 5
Comparison of Study Variables by Victim Type: Means (Standard Deviations)
Victim Type
Self-identified
(n = 351)
Convergent
(n = 211)
Non-Victim
(n = 442)
F
Indirect Aggression (T1) .17
a
(1.05) -.20
b
(.70) -.03
b
(.84) 11.83
***
Bullying (T1) .13
a
(1.12) -.15
b
(.84) -.01
ab
(.96) 5.36
**
Popularity (T1) .50
a
(1.46) -1.37
b
(1.71) .30
a
(1.40) 117.84
***
Note. Means in the same row that do not share superscripts differ at p < .05 using Tukey’s HSD post-
hoc comparison.
*p < .05. **p < .01. ***p < .001.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 34
Table 6
Predicting Indirect Aggression and Bullying at T2 from Victim Type at T1
Indirect Aggression T2
(N = 720)
Bullying T2
(N = 708)
b SE b SE
Gender .11
†
.06 -.22
**
.07
Outcome variable at T1 .45
***
.03 .43
***
.04
Non-victim -.26
**
.10 -.23
*
.11
Convergent victim -.46
***
.11 -.33
**
.12
Self-identified victim (low/average popularity) -.41
***
.11 -.35
**
.12
Note. In each model, the reference group is self-identified victims (high popularity).
†
p < .10. *p < .05. **p < .01.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 35
Appendix A: Popularity Moderated by Gender
Predicting T2 Peer-Nominated Victimization from T1 Popularity with Gender Interactions
Reputational Victimization Exclusion
b SE b SE
Gender -.07 .09 -.09 .09
Peer-nominated victimization T1 .57
***
.04 .45
***
.04
Popularity (linear term) T1 .11
.06 -.16
*
.07
Popularity (quadratic term) T1 .14
**
.05 -- --
Gender X Popularity (linear term) T1 -.11 .08 .10 .09
Gender X Popularity (quadratic term) T1 -.00 .06 -- --
Note. *p < .05. **p < .01. ***p < .001.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 36
Appendix B: Victimization and Popularity Moderated by Gender
Predicting T2 Indirect Aggression from T1 Victimization and T1 Popularity
with Gender Interactions
Panel A
b SE
Gender -.23
**
.07
Indirect aggression T1 .52
***
.05
Popularity T1 .27
***
.05
Self-reported victimization T1 .07 .05
Reputational victimization T1 .07 .04
Popularity X Self-reported victimization .11
*
.05
Gender X Self-reported victimization .08 .10
Gender X Popularity -.11 .07
Gender X Self-reported victimization X Popularity .03 .10
Panel B
b SE
Gender -.22
**
.07
Indirect aggression T1 .56
***
.05
Popularity T1 .27
***
.06
Self-reported victimization T1 .08 .05
Exclusion T1 .04 .04
Popularity X Self-reported victimization .13
*
.05
Gender X Self-reported victimization .09 .10
Gender X Popularity -.12 .07
Gender X Self-reported victimization X Popularity .02 .10
Note. N=370. In Panel A, the model included peer-nominated reputational victimization. In
Panel B, the model included peer-nominated exclusion.
*p < .05. **p < .01. ***p < .001.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 37
Appendix C: Description of Measures for Post Hoc Analyses
Reciprocated Friendships
Reciprocated friendships were measured using peer nominations. Participants were asked
“Who from your class are your best friends?”. If two participants both nominated each other as a
friend, then they were classified as having a reciprocated friendship. The number of reciprocated
friendships ranged from 0 to 11.
Social Preference
Peer nominations were also used to access social preference. Participants reported which
peers they “liked most” and which peers they “liked least”. The nominations for each item were
standardized. A difference score (liked most – liked least) was calculated then restandardized.
Loneliness
Participants completed a 12-item self-report measure on loneliness (the Louvain
Loneliness Scale for Children and Adolescents). Each item was on a 4-point scale ranging from
“never” to “often”. Example items include “I feel isolated from others” and “At school I feel
alone”. Participants scores for each item were summed to create a total score.
Self-Esteem
To measure self-esteem, participants completed a 10-item self-report measure (Rosenberg
Self-Esteem Scale). Each item was on a 4-point scale ranging from “totally disagree” to “totally
agree”. Example items include “Overall I am satisfied with myself” and “I feel that I have much
to be proud of”. Participants’ scores for each item were summed to create a total score.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 38
Appendix D: Post hoc Comparison of Victim Types
Post hoc Comparison of Victim Types on Psychosocial Adjustment Variables
Victim Type
Self-identified
(n = 351)
Convergent
(n = 211)
Non-Victim
(n = 442)
F
Reciprocated Friendships 3.15
a
(1.58) 2.49
b
(1.59) 3.18
a
(1.73) 13.86
***
Social Preference .13
a
(1.56) -.85
b
(1.74) .34
a
(1.40) 44.34
***
Loneliness 17.38
a
(5.84) 18.92
b
(6.75) 14.66
c
(3.78) 52.60
***
Self-Esteem 30.81
a
(5.42) 31.20
a
(6.28) 32.99
b
(4.86) 17.52
***
Note. Means in the same row that do not share superscripts differ at p < .05 using Tukey’s HSD post-
hoc comparison.
*p < .05. **p < .01. ***p < .001.
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 39
Appendix E: Model Comparisons in LPA- Depiction of BIC Values
DYNAMICS OF VICTIMIZATION, AGGRESSION, AND POPULARITY 40
Appendix F: Model Fit Indices for LPA
LLR AIC BIC ABIC BLRT
Varying variances, covariances
fixed to zero
1 Class 2827.24 5662.48 5682.13 5669.43 N/A
2 Classes 1465.70 2949.39 2993.60 2965.01 .001
3 Classes 1304.75 2637.51 2706.27 2661.81 .001
Varying variances, varying
covariances
1 Class 2801.26 5612.53 5637.09 5621.21 N/A
2 Classes 1464.45 2950.89 3004.92 2969.98 .001
3 Classes 1304.52 2643.04 2726.54 2672.55 .001
Equal variances, covariances
fixed to zero
1 Class 2827.24 5662.48 5682.13 5669.42 N/A
2 Classes 2282.76 4579.51 4613.90 4591.66 .001
3 Classes 2282.84 4585.68 4634.80 4603.04 1.00
4 Classes 2282.84 4591.67 4655.52 4614.23 --
5 Classes 2198.72 4429.44 4508.03 4457.21 --
Equal variances, equal
covariances
1 Class 2801.26 5612.53 5637.09 5621.21 N/A
2 Classes 2801.28 5618.57 5657.86 5632.45 .997
3 Classes 2276.99 4575.98 4630.01 4595.07 --
4 Classes 2276.98 4581.97 4650.73 4606.27 --
5 Classes 2188.54 4411.08 4494.58 4440.58 --
Note. AIC = Akaike Information Criterion; BIC = Bayesian Information Criterion; ABIC = Adjusted BIC; BLRT =
Bootstrap Likelihood Ratio Test.
BLRT was not available for the one class model. BLRT discontinued once the model fit between k and k – 1 was not
statistically significant.
For each model fit indices, the bolded value represents the most parsimonious solution.
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The role of social status and gender in the composition of bully-victim dyads in early adolescence
PDF
Stigma-based peer aggression and social status in middle adolescence: the unique implications of weight-related aggression
PDF
Is affluence a developmental risk for Hong Kong Chinese adolescents?
PDF
Social status, perceived social reputations, and perceived dyadic relationships in early adolescence
PDF
Family aggression, prosocial friends, and the risk of dating and friend victimization in late adolescence
PDF
Actual and perceived social reinforcements of weight-related cognitions and behaviors in adolescent peer groups
PDF
Mediators and moderators in the link between maternal psychological control and peer victimization for Hong Kong Chinese boys
PDF
The association between peer victimization and school functioning in Latino children
PDF
Examining the longitudinal relationships between community violence exposure and aggressive behavior among a sample of maltreated and non-maltreated adolescents
PDF
Adolescent life stress and the cortisol awakening response: the moderating roles of emotion regulation, attachment, and gender
PDF
Depressive outcomes following peer victimization during adolescence: the moderating role of friends’ attributes
PDF
Interpersonal coping responses during adolescence: implications for adjustment
PDF
Popularity as a predictor of friendship affiliation in adolescence
PDF
A longitudinal study of sexual minority stress and behavioral health in sexual minority adolescents
PDF
Examining the associations of respiratory problems with psychological functioning and the moderating role of engagement in pleasurable activities during late adolescence
PDF
Emotion regulation as a mechanism linking parents’ marital aggression to adolescent behavioral problems: a longitudinal analysis
PDF
Parents' attunement: relation to adolescent problem behavior and the moderating role of marital conflict
PDF
Family aggression exposure and community violence exposure associated with brain volume in late adolescence: a comparison of automated versus manual segmentation
PDF
Meta-accuracy as a moderator of the association between social experience and emotional adjustment during childhood
PDF
Complicated relationships: the prevalence and correlates of sexual assault, dating violence victimization, and minority stress among sexual minority adolescents throughout the United States
Asset Metadata
Creator
Malamut, Sarah T.
(author)
Core Title
Dynamics of victimization, aggression, and popularity in adolescence
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Psychology
Publication Date
12/02/2019
Defense Date
06/19/2019
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
adolescence,aggression,OAI-PMH Harvest,popularity,social status,victimization
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Schwartz , David (
committee chair
), Farver, JoAnn (
committee member
), Goldbach, Jeremy (
committee member
), Manis, Frank (
committee member
), Monterosso, John (
committee member
)
Creator Email
malamutsarah@gmail.com,smalamut@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c89-239272
Unique identifier
UC11674972
Identifier
etd-MalamutSar-7961.pdf (filename),usctheses-c89-239272 (legacy record id)
Legacy Identifier
etd-MalamutSar-7961.pdf
Dmrecord
239272
Document Type
Dissertation
Rights
Malamut, Sarah T.
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
aggression
popularity
social status
victimization