School Psychology Quarterly

© 2018 American Psychological Association 2018, Vol. 33, No. 2, 223–234

1045-3830/18/$12.00 http://dx.doi.org/10.1037/spq0000249

Bullying Involvement, Teacher–Student Relationships,

and Psychosocial Outcomes
Francis L. Huang, Crystal Lewis, Daniel R. Cohen, Sara Prewett, and Keith Herman
University of Missouri

Students involved in bullying experience mental health issues and negative psychosocial outcomes. Few
studies have investigated how teacher–student relationships (TSRs) may buffer the negative outcomes
experienced by students involved in bullying. To investigate the moderating role of TSRs with bullying
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

involvement status and psychosocial outcomes, we used data from 691 middle school students, 85
This document is copyrighted by the American Psychological Association or one of its allied publishers.

teachers, and 6 schools in one urban district. We used both student- and teacher-reported outcomes and
regression models included baseline measures (i.e., depression, concentration problems, emotional
regulation problems, behavioral engagement) taken 8 months earlier. Regardless of bullying involve-
ment, student-reported TSR had a beneficial association for all outcomes controlling for baseline
measures and student demographic variables. However, bully/victims with low TSRs experienced a
heightened risk for depressive symptoms suggesting increased attention to this subgroup of students.

Impact and Implications

Although teacher–student relationships (TSRs) have shown a beneficial relationship with various
academic outcomes, improved TSRs are also associated with a reduction in problematic psychosocial
outcomes for students. However, student-reported TSRs may also vary by a student’s bullying
involvement with bullies having poorer relationships with their teachers. Bully victims with low
TSRs experience an increased risk of depression and warrant further attention. Given the variability
of how much training teachers receive on the provision of social support for students (Pavri, 2004),
administrators and school psychologists should consider coordinated efforts that foster a school
culture that supports the importance of TSRs.

Keywords: teacher–student relationships, bullying depression, concentration problems, bully/victims

Supplemental materials: http://dx.doi.org/10.1037/spq0000249.supp

School bullying is generally defined as chronic, intentional acts detrimental effects associated with being a bully and/or a victim
of aggression directed toward a student who has less status or (Hawker & Boulton, 2000; Nansel et al., 2001).
power (Espelage & Swearer, 2003; Olweus, 1993). Based on the Based on their bullying involvement status, adolescents can be
School Crime Supplement to the National Crime Victimization classified as bullies, victims, bully/victims, and noninvolved1 stu-
Survey, 22% of students ages 12 through 18 in the U.S. have been dents with each involved group sharing varied psychosocial ad-
victims of some form of bullying (Lessne & Cidade, 2015). Bul- justment difficulties (Espelage & Swearer, 2003; Forero, McLel-
lying is recognized as a serious concern for educators, parents, and lan, Rissel, & Bauman, 1999; Veenstra et al., 2005). Considerable
students as a result of a host of risk factors and long lasting evidence has shown that bullying victimization is associated with
lower school engagement (Mehta, Cornell, Fan, & Gregory, 2013),
academic achievement (Nansel et al., 2001; Strøm, Thoresen,
Wentzel-Larsen, & Dyb, 2013), greater levels of depression
(Bond, Carlin, Thomas, Rubin, & Patton, 2001; Fekkes, Pijpers,
Francis L. Huang, Crystal Lewis, Daniel R. Cohen, Sara Prewett, and
Keith Herman, Department of Educational, School, and Counseling Psy-
Fredriks, Vogels, & Verloove-Vanhorick, 2006), higher levels of
chology, University of Missouri. suicide ideation (Copeland et al., 2013; Brunstein Klomek, Mar-
This research reported here was supported by the Institute of Education rocco, Kleinman, Schonfeld, & Gould, 2007), and poorer mental
Sciences, U.S. Department of Education, through Grant R305A130143 to and physical health (Benedict, Vivier, & Gjelsvik, 2014; Rigby,
the University of Missouri (PI: Keith Herman). The opinions expressed are 2000, 2001; Slee, 1995). Students who bully others have generally
those of the authors and do not represent views of the Institute or the U.S. shown higher levels of truancy (Forero et al., 1999) and lower
Department of Education. levels of school bonding (Haynie et al., 2001). Individuals cate-
Correspondence concerning this article should be addressed to Francis L.
Huang, Department of Educational, School, and Counseling Psychology,
University of Missouri, 16 Hill Hall, Columbia, MO 65211. E-mail: 1
Some make a distinction between bystanders who witness bullying and
huangf@missouri.edu noninvolved students (Haynie et al., 2001; C. Wang et al., 2015).

223
 224 HUANG, LEWIS, COHEN, PREWETT, AND HERMAN

gorized as both bullies and victims (e.g., bully/victims or aggres- bullying itself (Baldry & Farrington, 2005; Rothon, Head,
sive victims) represent a high risk group who exhibit increased Klineberg, & Stansfeld, 2011).
levels of conduct problems, school disengagement, concentration
problems, and a large number of (if not the most severe) psycho-
logical and psychosomatic symptoms (Forero et al., 1999; Ju- Teacher–Student Relationships and
vonen, Graham, & Schuster, 2003; Kumpulainen, Räsänen, & Psychosocial Outcomes
Puura, 2001). Bully/victims are also characterized by higher rates
Supportive relationships with adults and peers may moderate the
of problem behavior, depressive symptoms, lower self-control and
effects of bullying with regard to academic achievement (Konishi,
social competence, and poorer school functioning compared with
Hymel, Zumbo, & Li, 2010; Rothon et al., 2011; Strøm et al.,
other students (Haynie et al., 2001).
2013), yet little research has investigated this effect with regard to
The challenges faced by students involved in bullying are well
bullying involvement, TSRs, and psychosocial outcomes. The
recognized (Hawker & Boulton, 2000) though less research has
majority of TSR studies have focused on academically oriented
been conducted with regard to the potential benefits of positive
measures rather than psychosocial outcomes (e.g., O’Connor, Col-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

teacher–student relationships (TSRs) in relation to these negative

lins, & Supplee, 2012; Reddy, Rhodes, & Mulhall, 2003). Psycho-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

outcomes. Over the years, various studies have shown the benefits
social outcomes (e.g., engagement, emotional regulation, ability to
of positive TSRs with academic achievement, affective/behavioral
concentrate) are important in themselves and are directly linked to
outcomes, and engagement (Cornelius-White, 2007; Roorda,
academic achievement, grade promotion, and may protect
Koomen, Spilt, & Oort, 2011). Although a supportive classroom
against future academic failure (Cornelius-White, 2007; Dun-
environment and positive TSRs have been shown to be associated
can et al., 2007; Gregory & Korth, 2016; Li-Grining, Votruba-
with reduced bullying involvement (Di Stasio, Savage, & Burgos,
Drzal, Maldonado-Carreño, & Haas, 2010; Stipek & Miles,
2016; Gregory et al., 2010), less has been written about its poten-
2008; Willson & Hughes, 2009).
tial to serve as a buffer or protective factor with regard to negative
psychosocial outcomes brought about by bullying involvement Positive TSRs may protect against several forms of psychoso-
(Rigby, 2000). Protective factors refer to influences (e.g., TSRs) cial maladjustment (e.g., depression, low self-esteem, disruptive
that may modify or ameliorate an individual’s response to some behavior, and anxiety) which students involved in bullying often
risk factor (e.g., bullying involvement) that leads to a negative experience (e.g., Haynie et al., 2001; Juvonen et al., 2003; Yen et
outcome (e.g., depression; Rutter, 1985). Teacher–student relation- al., 2010). Psychosocial maladjustments may lead to poor school
ships thus have the potential to offset some of the stressors expe- engagement and lower academic outcomes (Graham, Bellmore, &
rienced by students involved with bullying. Mize, 2006).
Students who struggle with internalizing problems (e.g., depres-
sion) often show a greater dependency on their teachers compared
Teacher–Student Relationships and to their peers (Henricsson & Rydell, 2004) and students who are
Bullying Involvement overly dependent on their teachers have an increased likelihood of
Teachers play a critical role creating a supportive environment being bullied (Troop-Gordon & Kopp, 2011). However, strong
by engaging students, establishing relationships, managing the teacher support has been associated with reductions in depression
classroom, serving as positive role models for prosocial behaviors, (Colarossi & Eccles, 2003; Murray & Greenberg, 2000) and other
and enforcing school rules (Di Stasio et al., 2016; LaRusso, Ro- forms of internalizing behavior (Pianta & Stuhlman, 2004; Rueger,
mer, & Selman, 2008). Teachers set the overall tone of the class- Malecki, & Demaray, 2008). Students who perceived greater sup-
room and other than the students themselves, teachers are a port from their teachers were less depressed and had lower mis-
school’s key resource against bullying and victimization (Rodkin conduct problems (Reddy et al., 2003). A safe and supportive
& Hodges, 2003). When teachers are able to establish positive school environment may function as a protective factor for stu-
relationships with students who are difficult to teach and exhibit dents at risk for depression (Langille, Rasic, Kisely, Flowerdew, &
behavior problems, those students tend to have reduced aggressive Cobbett, 2012).
behaviors and greater school engagement (Doumen, Buyse, Col- Positive TSRs are also associated with decreases in teacher-
pin, & Verschueren, 2011). reported externalizing behavior and aggression (Benhorin & Mc-
A few recent studies have shown that positive TSRs were Mahon, 2008; Murray & Greenberg, 2000). Nurmi and Kiuru
associated with less victimization related to bullying and peer (2015) emphasized that child characteristics and behaviors could
aggression (Di Stasio et al., 2016; Elledge et al., 2016; Murray- evoke certain responses from their teachers and when teachers and
Harvey & Slee, 2010; Obsuth et al., 2016; Troop-Gordon & Kopp, students have a conflictual relationship, students’ disruptive be-
2011). In addition, a negative environment marked by poor or haviors are likely to persist though positive TSRs may reduce
highly conflictual TSRs, may promote peer aggression (Doumen et disruptive behaviors and improve prosocial behaviors(Nurmi &
al., 2011; Reavis, Keane, & Calkins, 2010; Troop-Gordon & Kiuru, 2015). Additionally, adolescents with a history of teacher
Kopp, 2011) and students with low TSRs may also be more conflict showed lower incidences of maladaptive behavior patterns
vulnerable to bullying (Shin & Kim, 2008). Although several when they perceived a positive, trustworthy relationship with their
studies have focused on bullying prevention (Ttofi & Farrington, teacher (Gregory & Ripski, 2008). Students with positive TSRs
2011) and some have focused on whether TSRs may buffer chil- tended to be more engaged in school, have higher academic
dren from the risk of peer aggression (Elledge et al., 2016; Troop- achievement, and were more motivated (Gregory & Korth, 2016;
Gordon & Kopp, 2011), fewer studies have investigated factors Roorda et al., 2011; M. T. Wang & Eccles, 2013; Wentzel, Battle,
(e.g., social supports) that may reduce the negative effects of Russell, & Looney, 2010). Teacher–student relationships then may
 BULLYING AND TEACHER–STUDENT RELATIONSHIPS 225

ameliorate some of the negative outcomes associated with bullying (Fraser, Kirby, & Smokowski, 2004). The presence of supportive
involvement. adults has been shown to strongly predict a child’s emotional
well-being and school-based relationships have been shown to be
The Current Study more important than familiar support in middle childhood (Oberle,
Schonert-Reichl, Guhn, Zumbo, & Hertzman, 2014). Additionally,
Given that students involved in bullying experience various although TSRs are dyadic in nature, where both students and
negative psychosocial outcomes (e.g., Haynie et al., 2001) and teachers have their own valid perception of the relationship (Brink-
positive TSRs are associated with improved psychosocial out- worth, McIntyre, Juraschek, & Gehlbach, 2017), we focused on the
comes (e.g., Reddy et al., 2003; Roorda et al., 2011), we investi- students’ perception of their relationship with their teachers be-
gated the moderating role of TSRs using a longitudinal sample of cause their point of view was important considering we were
middle school students. We used a 9-month cohort study with investigating student psychosocial outcomes. Finally, we focused
baseline measurements taken in the fall with follow-up measure- on middle school students and the middle school years mark a
ments in the spring. We specifically asked: particularly vulnerable time for youth (Akos, Rose, & Orthner,
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

2015; Feldlaufer, Midgley, & Eccles, 1988) with an increasing

This document is copyrighted by the American Psychological Association or one of its allied publishers.

1. Are there differences in TSRs based on student bullying number of students being bullied (Bradshaw, Sawyer, & O’Brennan,
involvement? 2007; Nansel et al., 2001). Changes in school characteristics from
primary to middle school (e.g., larger, impersonal classes) may
2. Do students involved in bullying (i.e., victims, bullies,
contribute to increased bullying victimization (Pellegrini, 2002;
bully/victims) have poorer psychosocial outcomes in the
Wigfield, Lutz, & Laurel Wagner, 2005) and a decline in school
spring (i.e., depression, concentration problems, emo-
engagement (Busteed, 2013). Although TSRs are important at all
tional regulation problems, behavioral engagement) com-
points in a student’s education, middle schools are often structured
pared with noninvolved students while controlling for
in a way that impedes the formation of strong ties with teachers
baseline (fall) measures?
(Reddy et al., 2003) and the quality of TSRs often declines in
3. Is the quality of teacher–student relationship associated middle school (Lynch & Cicchetti, 1997).
with beneficial psychosocial outcomes for all students
(e.g., higher TSRs is associated with a reduction in de- Method
pression)?

4. Are the negative outcomes experienced by students in- Participants

volved in bullying moderated by the quality of TSRs
(e.g., higher TSR is associated with a greater reduction in The study utilized data collected as part of a 9-month long
depression for students involved in bullying)? cluster randomized control trial (CRCT) evaluating the effects of a
classroom management program in middle school classrooms. The
Given the focus on TSRs and bullying, answering the initial current sample represented the first three of four annual cohorts
question if students involved in bullying differed on TSRs is that will be recruited by the end of the study. Six schools (all the
important as bullies and bully/victims tend to have poorer TSRs middle schools from one urban school district) were recruited to
(Cohen’s d ⬃ 0.40 to 0.50) compared with victims, bystanders, participate in the study and all math and reading teachers were
and noninvolved students (C. Wang, Swearer, Lembeck, Collins, invited to participate. Approximately 73% of teachers were re-
& Berry, 2015). The current study is distinct in several ways. First, cruited and 27% declined citing lack of time as a reason for
even though a large number of studies have been conducted on the nonparticipation. Each teacher selected one of their classrooms,
association of bullying with mental health measures using cross typically their most challenging classroom in terms of behavior
sectional data (see Rothon et al., 2011 for a list of studies), few management, as the target of the study and all students in the class
studies have used data spanning more than one time period (e.g., were invited to participate. Data were provided only for students
Bond et al., 2001; Fekkes et al., 2006; Kumpulainen, Räsänen, & who provided parent consent and child assent. Seventy-six percent
Henttonen, 1999; Troop-Gordon & Kopp, 2011) raising the issue of parents consented and 100% of students assented to participate.
of the directionality of effects (e.g., Does bullying cause depres- Participants were drawn from a total of 939 students (who were
sion or are depressed students bullied?). Second, we focus not only present for the entire school year) and 86 teachers (sixth to eighth
on victims of bullying but also on the bullies and bully/victims, grade) from six schools in an urban Missouri school district.
together with TSR, as prior research does not support the view that Students (female ⫽ 50%, Grade 6 ⫽ 40%, Grade 7 ⫽ 35%, Grade
adolescents can simply be classified as victims or bullies (Espelage 8 ⫽ 25%) were 67% Black, 17% White, and 16% identified as
& Swearer, 2003; Veenstra et al., 2005). The bully/victims in some other race/ethnicity. Teachers were primarily female (80%)
particular represent a particularly high-risk population and often and 71% White and 26% Black (with 3% reporting another race/
have psychiatric problems, demonstrated poor adjustment across ethnicity). The six schools had an average enrollment size of
social/emotional dimensions, and remain involved with bullying approximately 700 students (range ⫽ 412 to 885), with an average
for longer periods of time (Kumpulainen et al., 1999; Nansel et al., of 66% of students eligible for free or reduced price lunch (FRPL;
2001). Third, we focus on the potential of TSRs as a moderator of a commonly used proxy for socioeconomic status), and had a
adverse psychosocial outcomes related to bullying involvement. A predominantly Black student enrollment (M ⫽ 76%). The race/
protective factor acts as compensatory factor or a moderator which ethnicity breakdown of the student sample was representative of
changes the intensity of the effect risk factor for a set of outcomes the school district as a whole in 2015.
 226 HUANG, LEWIS, COHEN, PREWETT, AND HERMAN

Of the 939 students, 14% (n ⫽ 135) appeared in the dataset for each student whereas students completed paper self-
more than once (e.g., in the succeeding school year, the student assessments.
was with another teacher in the next grade level who was also
participating in the study). To avoid issues of sample nonindepen-
dence where the same student’s data is analyzed more than once, Measures
we used a random number generator to select only one case from Dependent and independent variables consisted of both student
the duplicates to be included in the analytic sample. and teacher reported scales and were collected in the fall and
As part of the CRCT, teachers rated each of the participating spring of each school year. Reliabilities of the scales in the spring
students on several measures which included two items indicating are presented using coefficient omega which overcomes many of
student bullying involvement in the fall and spring of the school the issues related to using Cronbach’s alphas (Dunn, Baguley, &
year (see Procedures and Measures sections). We selected only Brunsden, 2014). All student demographic data were obtained
students who were not involved in bullying in the fall (i.e., neither from the school district (see Table 1 for descriptives).
a bully nor a victim) to form our analytic sample, similar to what Depression (student reported). Eight of the nine questions of
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

was done by Fekkes et al. (2006), which resulted in a reduced the Patient Health Questionnaire (PHQ9) for Teens (Kroenke,
This document is copyrighted by the American Psychological Association or one of its allied publishers.

sample of 718 students.2 By limiting the sample to the noninvolved Spitzer, & Williams, 2001) scale were asked. One question deemed
students in the fall, we can assess the change in psychosocial sensitive was excluded (i.e., “Thoughts that you would be better off
outcomes that may result from bullying involvement in the spring.
dead, or of hurting yourself in some way?”). Responses on this scale
Compared with the original sample, the analytic sample had
ranged from 0 ⫽ not at all to 4 ⫽ nearly every day. This measure had
slightly more girls (female ⫽ 54%) and fewer Black students
an internal consistency score of ␻ ⫽ .82.
(64%). Of the 718 students, 60% were eligible for FRPL, 10%
Concentration problems (teacher reported). This scale
were identified as gifted and talented, and 8% were identified as
from Teacher Observation of Classroom Adaptation-Checklist
having a disability (see Table 1).
(TOCA-C; Koth, Bradshaw, & Leaf, 2009) was composed of the
mean of teacher-reported responses from the TOCA-C subscale
Procedures comprised of seven items such as “Student pays attention” and
The study was completed with approval from the University of “Completes assignments.” Response options ranged from 1 ⫽
Missouri Institutional Review Board. Data were collected in Oc- almost always to 6 ⫽ never and certain items were reverse coded
tober (fall) and May (spring) of each school year from both so that higher scores reflected more negative outcomes. The inter-
students and teachers. Teachers completed online survey ratings nal consistency of the scale was ␻ ⫽ .97.
Emotional regulation problems (teacher reported). This
scale was composed of the mean of teacher-reported responses
Table 1 from the TOCA-C (Koth et al., 2009) subscale comprised of five
Descriptive Statistics (n ⫽ 691) items such as “impulsive” and “easily frustrated”. Response op-
tions ranged from 1 ⫽ almost always to 6 ⫽ never. The internal
Variable n % M SD Range
consistency of the scale was ␻ ⫽ .87.
Bullying status (S) Behavioral engagement (student reported). The scale (Skin-
Noninvolved 563 81.5 ner, Furrer, Marchand, & Kindermann, 2008) was composed of the
Victim 32 4.6
Bully 77 11.1
mean of five items including items such as “When I’m in class, I
Bully/victim 19 2.8 listen very carefully” and “I try hard to do well in class.” Response
Race/ethnicity options ranged from 1 ⫽ not at all true to 4 ⫽ very true. The
White 155 22.4 internal consistency of the scale was ␻ ⫽ .78.
Black 503 72.8
Student engagement in bullying (teacher reported). In both
Other race/ethnicity 33 4.8
Female 377 45.4 fall and spring, as part of the TOCA-C (Koth et al., 2009), teachers
Eligible for FRPL 416 60.2 were asked to indicate for students in their class if a student bullied
With a disability 54 7.8 others and if a student was bullied by others. Response options for
Identified as gifted/talented 72 10.4 both items ranged from 1 ⫽ never to 6 ⫽ almost always. Students
Grade level
6 297 43.0 who received a rating of never or rarely (1 or 2) were classified as
7 243 35.2 not bullies or not victims following guidelines of Solberg and
8 151 21.9 Olweus (2003) which considers the frequency of occurrence of the
Scales activity. As a result, a student bullying involvement status variable
Student teacher relationship (S) 3.66 .95 1 to 5
Depression (F) 5.35 4.85 0 to 24 was formed consisting of victims of bullying, bullies, bully/vic-
Depression (S) 5.52 5.01 0 to 24 tims, and not involved students (i.e., student was neither a bully
Concentration problems (F) 2.60 1.16 1 to 6 nor bullied). In the spring, the majority of students were not
Concentration problems (S) 2.60 1.26 1 to 5.6 involved (82%), with 11% categorized as bullies, 5% as victims,
Emotional regulation (F) 1.98 .85 1 to 5.4
and 3% as bully/victims.
Emotional regulation (S) 2.14 .92 1 to 5.2
Behavioral engagement (F) 3.47 .47 1 to 4
Behavioral engagement (S) 3.38 .49 1 to 4 2
As a check, we also ran the analyses with all the students, regardless of
Note. FRPL ⫽ free or reduced price lunch; (S) ⫽ taken in the spring; bullying involvement status at Time 1 (n ⫽ 894). Results, available in the
(F) ⫽ taken in the fall. online appendix, were consistent with the results presented.
 BULLYING AND TEACHER–STUDENT RELATIONSHIPS 227

Although the identification of bullying involvement was ob- perspective of a student’s bullying involvement (Leff, Freedman,
tained through the use of single items without a presentation of a Macevoy, & Power, 2011).
formal definition of bullying, other surveys, such as the nationally Perceived teacher–student relationship scale (student
representative and commonly cited Youth Risk Behavioral Sur- reported). The scale (Roeser, Midgley, & Urdan, 1996) was
veillance System (Centers for Disease Control & Prevention composed of five items (e.g., “The teacher really cares about
[CDC], 2015) also use a single item, answerable by a yes or a no, students as individuals,” “The teacher treats students with re-
spect”) which asked students about their relationship with their
to measure bullying involvement and without providing a bullying
teacher. Responses ranged from 1 ⫽ not at all true to 5 ⫽ very true
definition based on the frequency of occurrence. Frequency is an
and were averaged to form the scale. Scale score reliability was
important aspect of bullying and as noted by the CDC (2017) itself,
␻ ⫽ .88. The scale contained items that corresponded to student
bullying is not an isolated incident and is repeated multiple times perceptions of trust, caring, and regard for adolescent perspectives
or is highly likely to be repeated. A study by Rønning et al. (2009), (Brinkworth et al., 2017).
which specifically compared teacher, student, and parent reports of Covariates. We also included gender, race/ethnicity (i.e.,
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

bullying, indicated that teacher-reported frequent bullying (com- White, Black, Other), and grade level as dummy-code variables
This document is copyrighted by the American Psychological Association or one of its allied publishers.

pared with only sometimes bullied and reports from other infor- (with White, boys, and the sixth grade as the reference groups). In
mants) was the strongest predictor of future psychiatric caseness, addition, a student’s FRPL status, disability status, and gifted
a marker for major depressive episodes (Østergaard et al., 2010). education status were included as dummy codes. Although not the
With regard to the use of definitions, two large experimental focus of the study, a dummy coded intervention status variable
studies have also shown that the presence or absence of a defini- (1 ⫽ treatment, 0 ⫽ control) was added to account for differences
tion did not affect bullying prevalence in both reports of students that may have resulted from the random assignment of teachers to
in middle (Huang & Cornell, 2016) and high (Huang & Cornell, the primary study conditions.3
2015) school.
Analytic Strategy
Previous studies have indicated that teachers can be a valid
source of a student’s bullying behavior (Card & Hodges, 2008; Of the 718 students in the analytic sample, only a small fraction
Leff, Patterson, Kupersmidt, & Power, 1999; Pellegrini & Bartini, had some missing data (3.8%) resulting in a reduced sample size
2000). In a systematic review of 26 school-based interventions to of 691 students with complete data. The variables with the largest
prevent bullying, Vreeman and Carroll (2007) identified several proportion of missing data were for behavioral engagement (2.4%)
studies that used teacher-reported bullying measures as evaluation and teacher–student relationships (2.0%) in the spring. To test if
outcomes. Some studies also focused on adolescent behavior such data were missing completely at random (MCAR), we used Little’s
as teacher-rated aggression or student social interactions (Fast, (1988) MCAR test. Results indicated that missing data were
Fanelli, & Salen, 2003; Tierney & Dowd, 2000). More recently, a MCAR, ␹2(203) ⫽ 163.3, p ⫽ .98, allowing the use of listwise
large (n ⫽ 12,344) randomized control effectiveness trial (Waas- deletion which would not bias results.
dorp, Bradshaw, & Leaf, 2012), with support from the Institute of To answer the first question investigating differences in TSRs
Education Sciences, the National Institute of Mental Health, and by bullying involvement status, we used a teacher-fixed effect
(Murnane & Willett, 2011) regression model using cluster robust
the CDC, used teacher-reports of bullying behavior using the
standard errors. Although ANOVAs are typically used to detect
TOCA-C. Another study noted a “substantial overlap” (Jansen
group differences with continuous outcome variables, our analysis
et al., 2014, p. 475) between teacher- and student-reported victim-
accounted for observation nonindependence due to the clustering
ization (⬃75% agreement) though noted differences may be due to
of observations which can lead to misestimated standard errors (F.
the perspectives of the reporters (Rønning et al., 2009). Even when Huang, 2016). To account for the possibility of inflated Type I
comparing victimization in adolescents using self-reports, diaries, errors due to multiple comparisons, we used a Benjamini-Hochberg
and peer nominations, the agreement of the different reporting (1995) correction procedure as suggested by the What Works Clear-
methods vary as each method was likely tapping into a different inghouse (Institute of Education Sciences, 2014).
aspect of bullying victimization (Pouwels, Lansu, & Cillessen, For the psychosocial outcomes, a series of multiple linear re-
2016). gression models were used to assess the association of the dummy-
In a large, statewide study conducted in 2017 with 85,762 coded bullying involvement variable (with not involved students
middle schoolers and 12,661 teachers in 410 public schools (Cor- as the reference group) with the outcome measures (i.e., depres-
nell et al., 2017), both teachers and students were asked to assess sion, concentration problems, emotional regulation problems, and
bullying at their school using the prevalence of teasing and bully- behavioral engagement) in the spring. All outcomes measures were
ing (PTB) scale (Konold et al., 2014). Teachers and staff tended to transformed to z scores so that the bullying involvement coeffi-
report that bullying was less of a problem (36%) at school com- cients could be interpreted as standardized mean differences using
pared with students (46%) and the correlation between student- Cohen’s (1992) effect size guidelines (i.e., 0.20 ⫽ small, 0.50 ⫽
and teacher-reported school-level PTB was r ⫽ .69. Another medium, 0.80 ⫽ large).
statewide study, but with high school students, has shown that the The first set of models included the bullying status variable in
correspondence with teacher and student bullying reports was r ⫽ the spring while controlling for commonly used student demo-
.61–.65 (Cornell, 2014). Though teacher-report measures may not
be sensitive to certain types of bullying (e.g., relational bullying, 3
This variable was included in all models but was not statistically
playground bullying), teacher reports provide a different but valid significant for any model and is not shown.
 228 HUANG, LEWIS, COHEN, PREWETT, AND HERMAN

graphic variables (i.e., gender, race/ethnicity, grade level, disabil- (n ⫽ 77) and bully/victims (n ⫽ 19) had the lowest TSRs (Ms ⫽
ity status, FRPL status, and gifted education status). In addition, a 3.34) compared with victims (n ⫽ 32; M ⫽ 3.69) and noninvolved
baseline psychosocial measure taken in the fall was included students (n ⫽ 563; M ⫽ 3.72). However, among all the compar-
which accounted for preexisting conditions. The models answered isons (which accounted for the nesting and multiple contrasts),
the second research question if there were differences in psycho- only the mean difference between bullies and noninvolved stu-
social outcomes associated with bullying involvement while con- dents was statistically significant (p ⬍ .01) reflecting a small to
trolling for a host of covariates including baseline scores. The moderate effect size (d ⫽ 0.36).
second set of models include the standardized measure of the
student-rated perceived TSR scale allowing the interpretation of Difference in Psychosocial Outcomes by Bullying
the regression coefficients as standardized betas. Statistically sig-
Involvement
nificant student-reported TSRs which result in improved psycho-
social outcomes would indicate the main/promotive effects of Based on regression models using the four spring psychosocial
TSRs (i.e., the benefit is for all student groups regardless of outcomes (see Table 2) and including the measures taken in the fall
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

bullying involvement status). The final set of models added an as covariates (Time 1 measure) for depression, concentration prob-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

interaction term using the TSR scale and bullying involvement lems, emotional regulation problems, and behavioral engagement,
status. A statistically significant interaction term would indicate a victims of bullying had higher concentration (d ⫽ 0.29, p ⬍ .05)
differential effect of TSRs (i.e., a protective factor or a buffering and emotional regulation problems (d ⫽ 0.40, p ⬍ .001) compared
effect) depending on the bullying involvement status of the stu- with the noninvolved group. Bullies also had higher concentration
dent. problems (d ⫽ 0.51, p ⬍ .001), emotional regulation problems
All models investigating psychosocial outcomes included school (d ⫽ 0.74, p ⬍ .001), and lower behavioral engagement
fixed effects to account for the clustering of respondents within (d ⫽ ⫺0.33, p ⬍ .01) compared with the reference group. Finally,
schools. Unlike multilevel models, the fixed effects model com- the bully victim had high levels of concentration (d ⫽ 0.71, p ⬍
pletely accounted for all observed and unobserved group-level .001) and emotional regulation problems (d ⫽ 1.36, p ⬍ .001)
variables that may bias results (F. Huang, 2016; Murnane & when compared with the students who were not involved in
Willett, 2011). A cohort fixed effect was also included to account bullying. For all groups, the differences in the depression outcome
for any differences that may be attributed to the different school were not statistically significant (all ps ⬎ .05). Model R2s ranged
years in which the data were collected. Cluster robust standard from .34 (for behavioral engagement) to .60 for concentration
errors (Cameron & Miller, 2015) at the teacher level were used to problems.
properly estimate standard errors and to account for any additional
nesting effects. Regression diagnostics were performed and no The Association of Teacher–Student Relationships and
influential observations (Cohen, Cohen, West, & Aiken, 2003)
Psychosocial Outcomes
were detected, multicollinearity was not a problem, and residuals
for all models were approximately normally distributed (all |skew- Adding the student-reported TSR variable in the next sets of
ness| ⬍ 1). All data management and analyses were done using R models indicated how much TSRs contributed to the outcomes
3.4 (R Core Team, 2017). over and above all the other variables already included in the
model (see Table 2). For depression, concentration problems, and
Results emotional regulation problems, the association of TSR was statis-
tically significant with higher TSRs associated with better out-
Difference in TSR by Bullying Involvement comes (␤ ⫽ ⫺0.09, ps ⬍ .01) with small improvements in R2. The
coefficient of TSR for behavioral engagement was larger and
Descriptively, comparison of TSR mean scores in the spring statistically significant (␤ ⫽ 0.27, p ⬍ .001) with a sizable increase
(see Figure 1) by bullying involvement status indicates that bullies in R2 (Rchange
2
⫽ .07).

4.5

4.0
Relationship Scale
Teacher-Student

3.72 3.69
3.5
3.34 3.34

3.0

2.5

Bu l ly* Bul ly/Vi cti m No ni nvo lved * Victim

Figure 1. Comparison of spring teacher-student relationship scores by bullying involvement status (n ⫽ 691).
ⴱ
Only the difference between the bully and noninvolved groups are statistically significant (p ⬍ .01). Error bars
represent the 95% confidence interval. Analysis accounting for clustering and multiple comparisons.
 BULLYING AND TEACHER–STUDENT RELATIONSHIPS 229

Table 2
Regression Results for Spring Psychosocial Outcomes (n ⫽ 691)

Emotional regulation
Depression Concentration problems problems Behavioral engagement
Variables w/T1 w/TSR w/T1 w/TSR w/T1 w/TSR w/T1 w/TSR

Victims1 .02 (.11) .01 (.11) .29ⴱ (.12) .29ⴱ (.12) .40ⴱⴱⴱ (.12) .40ⴱⴱⴱ (.12) .05 (.12) .09 (.12)
Bullies1 .05 (.11) .02 (.12) .51ⴱⴱⴱ (.08) .47ⴱⴱⴱ (.08) .74ⴱⴱⴱ (.10) .70ⴱⴱⴱ (.10) ⫺.33ⴱⴱ (.12) ⫺.24ⴱ (.12)
Bully/victims1 .20 (.16) .17 (.15) .71ⴱⴱⴱ (.20) .68ⴱⴱⴱ (.21) 1.36ⴱⴱⴱ (.21) 1.33ⴱⴱⴱ (.21) ⫺.05 (.21) .04 (.18)
Fall measure .12ⴱⴱⴱ (.01) .11ⴱⴱⴱ (.01) .56ⴱⴱⴱ (.03) .55ⴱⴱⴱ (.03) .65ⴱⴱⴱ (.04) .65ⴱⴱⴱ (.04) 1.16ⴱⴱⴱ (.08) 1.01ⴱⴱⴱ (.08)
Female .13 (.07) .12 (.07) ⫺.07 (.05) ⫺.09 (.05) .04 (.05) .02 (.05) .07 (.07) .12 (.06)
Black2 ⫺.11 (.08) ⫺.12 (.08) .19ⴱⴱ (.06) .19ⴱⴱ (.06) .06 (.08) .05 (.08) .08 (.10) .11 (.09)
Other race2 .05 (.15) .04 (.15) .06 (.09) .06 (.08) .01 (.11) .01 (.11) ⫺.17 (.18) ⫺.15 (.17)
With a disability .00 (.10) .03 (.11) ⫺.05 (.09) ⫺.02 (.09) .1 (.10) .12 (.10) .03 (.12) ⫺.03 (.11)
⫺.03 (.08) ⫺.03 (.08) ⫺.05 (.06) ⫺.06 (.06) ⫺.17ⴱ (.07) ⫺.17ⴱ (.07)
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Identified as gifted .06 (.10) .05 (.10)

⫺.05 (.05) ⫺.07 (.05) ⫺.17ⴱ (.07) ⫺.13ⴱ (.06)
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Eligible for FRPL .04 (.07) .03 (.08) .02 (.05) .01 (.05)
Grade 73 ⫺.04 (.08) ⫺.04 (.08) .01 (.08) .00 (.08) .05 (.10) .04 (.09) ⫺.02 (.07) ⫺.01 (.07)
Grade 83 ⫺.08 (.06) ⫺.09 (.06) .07 (.09) .07 (.09) .08 (.09) .08 (.09) ⫺.12 (.09) ⫺.13 (.08)
TSR scale ⫺.09ⴱⴱⴱ (.03) ⫺.09ⴱⴱ (.03) ⫺.09ⴱⴱ (.03) .27ⴱⴱⴱ (.04)
R2 .35 .36 .60 .61 .55 .56 .34 .41
Note. All outcomes and TSR scale are standardized. Cluster robust standard errors in parenthesis. Models accounted for school and time fixed effects.
Intervention status included as a predictor. T1 ⫽ time 1 variable (baseline) of the outcome measure taken in the fall; FRPL ⫽ free or reduce price lunch;
TSR ⫽ teacher–student relationship scale.
1
Noninvolved students ⫽ reference group. 2 White ⫽ reference group. 3 Grade 6 ⫽ reference group.
ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01. ⴱⴱⴱ p ⬍ .001.

The Moderating Role of TSR and Psychosocial suggesting virtually no association of TSR and depression for
Outcomes bullies. The relationship however of TSR with bully/victims is
much more pronounced and TSR functions as a protective factor
For the final set of models, the interaction terms between TSR for bully/victims. However, bully/victims with poor TSR may also
and bullying involvement status were added to test for the mod- be at risk for higher levels of depression.
erating relationship. For three out of the four outcomes, the addi-
tional variables did not result in an improvement in model fit nor
Discussion
were any of the added coefficients statistically significant (all ps ⬎
.05, not shown). For depression, statistically significant interaction Although studies have investigated the role of TSRs in relation
terms were found (see Table 3). As coefficients for interactions are to bullying involvement and peer aggression (e.g., Elledge et al.,
challenging to interpret on their own, we present the results of the 2016; Troop-Gordon & Kopp, 2011), studies have generally not
moderation effects visually in Figure 2. investigated the potential of TSRs in reducing the negative out-
As seen in Figure 2, depression decreased for both victims and comes associated with bullying itself. Descriptively, our results
noninvolved students as TSR increased indicating the beneficial suggest that bullies and bully/victims had the poorest relationships
association of TSR. In contrast, the slope for bullies is almost flat with teachers and results were similar to another middle school
study that compared TSRs among bullying-involved students (C.
Wang et al., 2015). However, in our study, only student-reported
Table 3 TSR differences of bullies versus noninvolved students were sta-
Moderating Role of Teacher–Student Relationships and Bullying tistically significant, probably as a result of the large standard
Status With Depression (n ⫽ 691) errors that bully/victims had with TSRs (see Figure 1) due to the
smaller number of bully/victims in our sample.
Variables Depression
Teachers should be cognizant that TSRs perceived by students
1
Victims .01 (.11) as being positive may function as both a protective and promotive
Bullies1 .08 (.11) factor as it relates to psychosocial outcomes. Teachers may have a
Bully/victims1 .06 (.16) harder time developing relationships with challenging students
Fall measure .11ⴱⴱⴱ (.01)
TSR scale ⫺.12ⴱⴱⴱ (.03) who pick on or victimize other students (as indicated by having the
Victim ⫻ TSR .06 (.11) lowest TSR scores), however, the bullies and the bully/victims, not
Bully ⫻ TSR .21ⴱ (.10) just the victims, may benefit from better TSRs. As Nurmi and
Bully/victim ⫻ TSR ⫺.29ⴱ (.12) Kiuru (2015) have illustrated, when students develop good rela-
R2 .36
tionships with their teachers, the level of student disruptive behav-
Note. TSR ⫽ teacher–student relationship scale. Model includes all prior iors tended to decrease and prosocial behaviors increased.
covariates used (not shown). All outcomes and TSR scale are standardized. Study results also indicated that compared with noninvolved
Cluster robust standard errors in parenthesis. Models accounted for school
and time fixed effects. Intervention status included as a predictor.
individuals, students with bullying involvement had more concen-
1
Noninvolved students ⫽ reference group. tration and emotion regulation problems, and lower engagement.
ⴱ
p ⬍ .05. ⴱⴱⴱ p ⬍ .001. This is consistent with previous findings that detail the psycho-
 230 HUANG, LEWIS, COHEN, PREWETT, AND HERMAN
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Figure 2. Moderating relationship of teacher-student relationships and bullying status with depression (n ⫽
691). Both depression and teacher-student relationships are shown in standard deviation units. See the online
article for the color version of this figure.

logical, social, and behavioral consequences of bullying involve- may better perform their role in reducing bullying and the mal-
ment not just for the victims (Copeland et al., 2013; Haynie et al., adaptive outcomes related to it.
2001). These psychosocial outcomes, though highly relevant on Several interventions, some relatively simple (e.g., Gehlbach et
their own, have a strong predictive relationship with future aca- al., 2016) and some more intensive (e.g., Anderson, Christenson,
demic success and warrant considerable attention (Duncan et al., Sinclair, & Lehr, 2004; Mikami, Gregory, Allen, Pianta, & Lun,
2007; Li-Grining et al., 2010). Not surprisingly as many other 2011) show promise in fostering better relationships between
studies have noted the benefits of TSR and engagement (Roorda et students and teachers. For example, MyTeachingPartner-Secondary
al., 2011), the behavioral engagement outcome had the strongest (Mikami et al., 2011), a teacher professional development program, is
relationship with TSR. As bullies had lower behavioral engage- designed to promote TSRs among middle and high school students.
ment compared with noninvolved students, positive TSRs may With MTP-S, teachers are coached to inquire about students’ extra-
help reduce this risk as positive TSRs may raise engagement levels curricular interests (to build relationships) and teachers are encour-
for all students. This is also especially salient in middle schools, aged to incorporate these interests in their teaching material to im-
regardless of bullying involvement status, where engagement has prove engagement. Although MTP-S is an intensive intervention,
been known to wane (Busteed, 2013). Gehlbach et al.’s (2016) brief treatment (used with ninth graders)
Our finding that the quality of student-reported TSRs acted as a focuses on highlighting similarities between student and teacher in-
moderator for self-reported depression is particularly notable. terests (with information collected through a get-to-know-you survey)
TSRs moderated depressive symptomatology for bully/victims; as a means of building positive relationships.
specifically, as positive TSR increased, depressive symptoms de-
creased more sharply for students in the bully/victim category
Limitations
compared to all other students (see Figure 2). However, depression
symptoms were much higher for bully/victims with poor TSRs. A Several limitations should be kept in mind when interpreting
possible reason may be that bully/victims are at greater risk for results. First, our sample consisted primarily of Black students
poorer psychosocial outcomes and existing literature supports this with a high level of students eligible for FRPL. Thus, the study
notion (Copeland et al., 2013; Holt et al., 2015; Kumpulainen et findings may not generalize to the broader public but focuses on
al., 1999; Swearer, Song, Cary, Eagle, & Mickelson, 2001) which disadvantaged students. Second, the use of teacher reports in
may make them more amenable to the effects of having a positive middle school may underestimate prevalence rates as teachers may
relationship with their teacher. not observe bullying behaviors that may occur in other classes
These findings and previous literature describing the beneficial (Craig, Henderson, & Murphy, 2000) or in other places (e.g.,
association of positive TSRs with various outcomes (e.g., Roorda bathrooms, school cafeteria, playground) where there may be less
et al., 2011) suggest that the enhancement of relationships between adult supervision (Pellegrini & Bartini, 2000). Although other
teachers and students may be meaningful for students experiencing methods of bullying involvement identification have been sug-
a variety of psychosocial risk factors. As risk factors increase, gested (Cornell & Huang, 2015), there is no existing gold standard
access to positive TSRs may be particularly critical. If teachers with regard to the measurement of bullying involvement (F. L.
have a better awareness of TSR and are provided support for Huang & Cornell, 2016; Ladd & Kochenderfer-Ladd, 2002). How-
building positive relationships with challenging students, teachers ever, our results are consistent with the pattern of identification of
 BULLYING AND TEACHER–STUDENT RELATIONSHIPS 231

bullies and victims suggested by Wienke Totura, Green, Karver, Bond, L., Carlin, J. B., Thomas, L., Rubin, K., & Patton, G. (2001). Does
and Gesten (2009) where teachers may report more bullies and less bullying cause emotional problems? A prospective study of young
victims compared with student reports. Third, specific modalities teenagers. British Medical Journal, 323, 480 – 484. http://dx.doi.org/10
of bullying (e.g., social, verbal) were not explored and the rela- .1136/bmj.323.7311.480
tionship of TSR and psychosocial outcomes may differ based on Bradshaw, C. P., Sawyer, A. L., & O’Brennan, L. M. (2007). Bullying and
peer victimization at school: Perceptual differences between students
the type of bullying. In addition, teachers may not observe other
and school staff. School Psychology Review, 36, 361–382.
covert forms of bullying such as cyberbullying (Card & Hodges, Brinkworth, M. E., McIntyre, J., Juraschek, A. D., & Gehlbach, H. (2017).
2008). Fourth, we consider TSRs from the point of view of the Teacher–student relationships: The positives and negatives of assessing
student, considering that student outcomes were of primary inter- both perspectives. Journal of Applied Developmental Psychology. Ad-
est. Teachers themselves may have their own view of their rela- vance online publication. http://dx.doi.org/10.1016/j.appdev.2017.09
tionship with their students which do not necessarily have to match .002
a student’s perception. Finally, though our primary interest was on Brunstein Klomek, A., Marrocco, F., Kleinman, M., Schonfeld, I. S., &
student relationships with teachers, support from other adults such Gould, M. S. (2007). Bullying, depression, and suicidality in adoles-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

as parents (which we did not have data for) have also shown to cents. Journal of the American Academy of Child & Adolescent Psychi-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

function as a protective factor for middle school students (Rueger, atry, 46, 40 – 49. http://dx.doi.org/10.1097/01.chi.0000242237.84925.18
Chen, Jenkins, & Choe, 2014). Busteed, B. (2013). The school cliff: Student engagement drops with each
school year. Retrieved from http://www.gallup.com/opinion/gallup/
170525/school-cliff-student-engagement-drops-school-year.aspx
Conclusion and Implications
Cameron, A. C., & Miller, D. L. (2015). A practitioner’s guide to cluster-
The current study found differences in both teacher–student robust inference. The Journal of Human Resources, 50, 317–372. http://
relationships as well as psychosocial outcomes for students that dx.doi.org/10.3368/jhr.50.2.317
were classified as bullies, victims, and bully/victims compared Card, N. A., & Hodges, E. V. E. (2008). Peer victimization among
with those with no involvement in bullying. In addition, results schoolchildren: Correlations, causes, consequences, and considerations
in assessment and intervention. School Psychology Quarterly, 23, 451–
indicated that for bully/victims, student-reported TSRs functioned
461. http://dx.doi.org/10.1037/a0012769
as a protective factor for depressive symptomatology and TSR had
Centers for Disease Control. (2017). Featured topic: Bullying Research.
a promotive association for emotional regulation, behavioral en- Retrieved from https://www.cdc.gov/violenceprevention/youthviolence/
gagement, and concentration problems. These findings highlight bullyingresearch/index.html
the need to include efforts to support positive relationships be- Centers for Disease Control and Prevention. (2015). Youth risk behavior
tween teachers and students as a means to facilitate the social and surveillance system. Retrieved from http://www.cdc.gov/healthyyouth/
emotional well being of students, particularly for those at greater data/yrbs/index.htm
risk. Given the variability in how much training teachers receive Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155–159.
on the provision of social supports for students (Pavri, 2004), http://dx.doi.org/10.1037/0033-2909.112.1.155
administrators, school psychologists, school social workers, coun- Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple
selors, and other school personnel charged with supporting the regression/correlation analysis for the behavioral sciences. Mahwah,
psychosocial well-being of students should consider coordinated NJ: Erlbaum.
Colarossi, L. G., & Eccles, J. S. (2003). Differential effects of support
efforts to foster a school culture that places greater value on the
providers on adolescents’ mental health. Social Work Research, 27,
importance of TSRs, especially in secondary schools.
19 –30. http://dx.doi.org/10.1093/swr/27.1.19
Copeland, W. E., Wolke, D., Angold, A., Costello, E. J., Copeland, W. E.,
References Angold, A., & Wolke, D. (2013). Adult psychiatric outcomes of bullying
Akos, P., Rose, R. A., & Orthner, D. (2015). Sociodemographic modera- and being bullied by peers in childhood and adolescence. Journal of the
tors of middle school transition effects on academic achievement. The American Medical Association Psychiatry, 70, 419 – 426. http://dx.doi
Journal of Early Adolescence, 35, 170 –198. http://dx.doi.org/10.1177/ .org/10.1001/jamapsychiatry.2013.504
0272431614529367 Cornelius-White, J. (2007). Learner-centered teacher–student relationships
Anderson, A. R., Christenson, S. L., Sinclair, M. F., & Lehr, C. A. (2004). are effective: A meta-analysis. Review of Educational Research, 77,
Check & Connect: The importance of relationships for promoting en- 113–143. http://dx.doi.org/10.3102/003465430298563
gagement with school. Journal of School Psychology, 42, 95–113. Cornell, D. (2014, December). School climate and safety in Virginia high
http://dx.doi.org/10.1016/j.jsp.2004.01.002 schools: Perceptions of students and teachers. Keynote presentation for
Baldry, A. C., & Farrington, D. P. (2005). Protective factors as moderators Strengthening Connections Climate Forum. Virginia Center for School
of risk factors in adolescence bullying. Social Psychology of Education, and Campus Safety, Midlothian, VA.
8, 263–284. http://dx.doi.org/10.1007/s11218-005-5866-5 Cornell, D., & Huang, F. (2015). School counselor use of peer nominations
Benedict, F. T., Vivier, P. M., & Gjelsvik, A. (2014). Mental health and to identify victims of bullying. Professional School Counseling, 18,
bullying in the United States among children aged 6 to 17 years. Journal 191–205. http://dx.doi.org/10.5330/2156-759X-18.1.191
of Interpersonal Violence, 30, 782–795. http://dx.doi.org/10.1177/ Cornell, D., Huang, F., Konold, T., Jia, Y., Malone, M., Burnette, A., . . .
0886260514536279 Meyer, J. P. (2017). Tech. Rep. No. of the Virginia Secondary School
Benhorin, S., & McMahon, S. D. (2008). Exposure to violence and ag- Climate Survey 2017: Results for 6th, 7th, and 8th grade students and
gression: Protective roles of social support among urban African Amer- teachers. Charlottesville, VA: University of Virginia.
ican youth. Journal of Community Psychology, 36, 723–743. http://dx Craig, W. M., Henderson, K., & Murphy, J. G. (2000). Prospective teach-
.doi.org/10.1002/jcop.20252 ers’ attitudes toward bullying and victimization. School Psychology
Benjamini, Y., & Hochberg, Y. (1995). Controlling the false discovery International, 21, 5–21. http://dx.doi.org/10.1177/0143034300211001
rate: A practical and powerful approach to multiple testing. Journal of Di Stasio, M. R., Savage, R., & Burgos, G. (2016). Social comparison,
the Royal Statistical Society Series B. Methodological, 57, 289 –300. competition and teacher–student relationships in junior high school
 232 HUANG, LEWIS, COHEN, PREWETT, AND HERMAN

classrooms predicts bullying and victimization. Journal of Adolescence, of cross-sectional studies. Journal of Child Psychology and Psychiatry,
53, 207–216. http://dx.doi.org/10.1016/j.adolescence.2016.10.002 41, 441– 455. http://dx.doi.org/10.1111/1469-7610.00629
Doumen, S., Buyse, E., Colpin, H., & Verschueren, K. (2011). Teacher– Haynie, D. L., Nansel, T., Eitel, P., Crump, A. D., Saylor, K., Yu, K., &
child conflict and aggressive behaviour in first grade: The intervening Simons-Morton, B. (2001). Bullies, victims, and bully/victims: Distinct
role of children’s self-esteem. Infant and Child Development, 20, 449 – groups of at-risk youth. The Journal of Early Adolescence, 21, 29 – 49.
465. http://dx.doi.org/10.1002/icd.725 http://dx.doi.org/10.1177/0272431601021001002
Duncan, G. J., Dowsett, C. J., Claessens, A., Magnuson, K., Huston, A. C., Henricsson, L., & Rydell, A.-M. (2004). Elementary school children with
Klebanov, P., . . . Japel, C. (2007). School readiness and later achieve- behavior problems: Teacher-child relations and self-perception. A pro-
ment. Developmental Psychology, 43, 1428 –1446. http://dx.doi.org/10 spective study. Merrill-Palmer Quarterly, 50, 111–138. http://dx.doi
.1037/0012-1649.43.6.1428 .org/10.1353/mpq.2004.0012
Dunn, T. J., Baguley, T., & Brunsden, V. (2014). From alpha to omega: A Holt, M. K., Vivolo-Kantor, A. M., Polanin, J. R., Holland, K. M., DeGue,
practical solution to the pervasive problem of internal consistency esti- S., Matjasko, J. L., . . . Reid, G. (2015). Bullying and suicidal ideation
mation. British Journal of Psychology, 105, 399 – 412. http://dx.doi.org/ and behaviors: A meta-analysis. Pediatrics, 135, e496 – e509. http://dx
10.1111/bjop.12046 .doi.org/10.1542/peds.2014-1864
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Elledge, L. C., Elledge, A. R., Newgent, R. A., Cavell, T. A., Newgent, Huang, F. (2016). Alternatives to multilevel modeling for the analysis of
This document is copyrighted by the American Psychological Association or one of its allied publishers.

R. A., Elledge, L. C., & Elledge, A. R. (2016). Social risk and peer clustered data. Journal of Experimental Education, 84, 175–196. http://
victimization in elementary school children: The protective role of dx.doi.org/10.1080/00220973.2014.952397
teacher–student relationships. Journal of Abnormal Child Psychology, Huang, F. L., & Cornell, D. G. (2015). The impact of definition and
44, 691–703. http://dx.doi.org/10.1007/s10802-015-0074-z question order on the prevalence of bullying victimization using student
Espelage, D., & Swearer, S. (2003). Research on bullying and victimiza- self-reports. Psychological Assessment, 27, 1484 –1493. http://dx.doi
tion: What have we learned and where do we go from here? School .org/10.1037/pas0000149
Psychology Review, 32, 365–383. Huang, F. L., & Cornell, D. G. (2016). Question order affects the mea-
Fast, J., Fanelli, F., & Salen, L. (2003). How becoming mediators affects surement of bullying victimization among middle school students. Ed-
aggressive students. Children & Schools, 25, 161–171. http://dx.doi.org/ ucational and Psychological Measurement, 76, 724 –740. http://dx.doi
10.1093/cs/25.3.161
.org/10.1177/0013164415622664
Fekkes, M., Pijpers, F. I. M., Fredriks, A. M., Vogels, T., & Verloove-
Institute of Education Sciences. (2014). WWC procedures and standards
Vanhorick, S. P. (2006). Do bullied children get ill, or do ill children get
handbook: What Works Clearinghouse. Retrieved from http://ies.ed.gov/
bullied? A prospective cohort study on the relationship between bullying
ncee/wwc/documentsum.aspx?sid⫽
and health-related symptoms. Pediatrics, 117, 1568 –1574. http://dx.doi
Jansen, P. W., Verlinden, M., Berkel, A. D., Mieloo, C. L., Raat, H.,
.org/10.1542/peds.2005-0187
Hofman, A., . . . Tiemeier, H. (2014). Teacher and peer reports of
Feldlaufer, H., Midgley, C., & Eccles, J. S. (1988). Student, teacher, and
overweight and bullying among young primary school children. Pedi-
observer perceptions of the classroom environment before and after the
atrics. Advance online publication. http://dx.doi.org/10.1542/peds.2013-
transition to junior high school. The Journal of Early Adolescence, 8,
3274
133–156.
Juvonen, J., Graham, S., & Schuster, M. A. (2003). Bullying among young
Forero, R., McLellan, L., Rissel, C., & Bauman, A. (1999). Bullying
adolescents: The strong, the weak, and the troubled. Pediatrics, 112,
behaviour and psychosocial health among school students in New South
1231–1237. http://dx.doi.org/10.1542/peds.112.6.1231
Wales, Australia: Cross sectional survey. British Medical Journal, 319,
Konishi, C., Hymel, S., Zumbo, B. D., & Li, Z. (2010). Do school bullying
344 –348.
and student-teacher relationships matter for academic achievement? A
Fraser, M., Kirby, L., & Smokowski, P. (2004). Risk and resilience in
childhood. In M. Fraser (Ed.), Risk and resiliency in childhood: An multilevel analysis. Canadian Journal of School Psychology, 25, 19 –39.
ecological perspective (pp. 13– 66). Washington, DC: NASW Press. http://dx.doi.org/10.1177/0829573509357550
Gehlbach, H., Brinkworth, M. E., King, A. M., Hsu, L. M., McIntyre, J., & Konold, T., Cornell, D., Huang, F., Meyer, P., Lacey, A., Nekvasil, E., . . .
Rogers, T. (2016). Creating birds of similar feathers: Leveraging simi- Shukla, K. (2014). Multilevel multi-informant structure of the authori-
larity to improve teacher–student relationships and academic achieve- tative school climate survey. School Psychology Quarterly, 29, 238 –
ment. Journal of Educational Psychology, 108, 342–352. http://dx.doi 255. http://dx.doi.org/10.1037/spq0000062
.org/10.1037/edu0000042 Koth, C. W., Bradshaw, C. P., & Leaf, P. J. (2009). Teacher Observation
Graham, S., Bellmore, A. D., & Mize, J. (2006). Peer victimization, of Classroom Adaptation-Checklist: Development and factor Structure.
aggression, and their co-occurrence in middle school: Pathways to Measurement & Evaluation in Counseling & Development, 42, 15–30.
adjustment problems. Journal of Abnormal Child Psychology, 34, 363– http://dx.doi.org/10.1177/0748175609333560
378. http://dx.doi.org/10.1007/s10802-006-9030-2 Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001). The PHQ-9:
Gregory, A., Cornell, D., Fan, X., Sheras, P., Shih, T.-H., & Huang, F. Validity of a brief depression severity measure. Journal of General
(2010). Authoritative school discipline: High school practices associated Internal Medicine, 16, 606 – 613. http://dx.doi.org/10.1046/j.1525-1497
with lower bullying and victimization. Journal of Educational Psychol- .2001.016009606.x
ogy, 102, 483– 496. Kumpulainen, K., Räsänen, E., & Henttonen, I. (1999). Children involved
Gregory, A., & Korth, J. (2016). Teacher–student relationships and behav- in bullying: Psychological disturbance and the persistence of the in-
ioral engagement in the classroom. In K. Wentzel and G. Ramani (Eds.), volvement. Child Abuse & Neglect: The International Journal, 23,
Handbook of social influences in school contexts: Social-emotional, 1253–1262. http://dx.doi.org/10.1016/S0145-2134(99)00098-8
motivation, and cognitive outcomes (pp. 178 –191). Boca Raton, Florida: Kumpulainen, K., Räsänen, E., & Puura, K. (2001). Psychiatric disorders
CRC Press, Taylor & Francis Group. and the use of mental health services among children involved in
Gregory, A., & Ripski, M. B. (2008). Adolescent trust in teachers: Impli- bullying. Aggressive Behavior, 27, 102–110. http://dx.doi.org/10.1002/
cations for behavior in the high school classroom. School Psychology ab.3
Review, 37, 337–353. Ladd, G. W., & Kochenderfer-Ladd, B. (2002). Identifying victims of peer
Hawker, D. S. J., & Boulton, M. J. (2000). Twenty years’ research on peer aggression from early to middle childhood: Analysis of cross-informant
victimization and psychosocial maladjustment: A meta-analytic review data for concordance, estimation of relational adjustment, prevalence of
 BULLYING AND TEACHER–STUDENT RELATIONSHIPS 233

victimization, and characteristics of identified victims. Psychological Oberle, E., Schonert-Reichl, K. A., Guhn, M., Zumbo, B. D., & Hertzman,
Assessment, 14, 74 –96. http://dx.doi.org/10.1037/1040-3590.14.1.74 C. (2014). The role of supportive adults in promoting positive develop-
Langille, D., Rasic, D., Kisely, S., Flowerdew, G., & Cobbett, S. (2012). ment in middle childhood: A population-based study. Canadian Journal
Protective associations of school connectedness with risk of depression of School Psychology, 29, 296 –316. http://dx.doi.org/10.1177/0829
in Nova Scotia adolescents. Canadian Journal of Psychiatry, 57, 759 – 573514540116
764. http://dx.doi.org/10.1177/070674371205701208 Obsuth, I., Murray, A. L., Malti, T., Sulger, P., Ribeaud, D., Eisner, M., . . .
LaRusso, M., Romer, D., & Selman, R. (2008). Teachers as builders of Ribeaud, D. (2016). A non-bipartite propensity score analysis of the
respectful school climates: Implications for adolescent drug use norms effects of teacher–student relationships on adolescent problem and
and depressive symptoms in high school. Journal of Youth and Adoles- prosocial behavior. Journal of Youth and Adolescence, 46, 1661–1687.
cence, 38, 386 –398. http://dx.doi.org/10.1007/s10964-007-9212-4 O’Connor, E. E., Collins, B. A., & Supplee, L. (2012). Behavior problems
Leff, S., Freedman, M., Macevoy, J., & Power, T. (2011). Considerations in late childhood: The roles of early maternal attachment and teacher-
when measuring outcomes to assess for the effectiveness of bullying-and child relationship trajectories. Attachment & Human Development, 14,
aggression-prevention programs in the schools. In D. Espelage & S. 265–288. http://dx.doi.org/10.1080/14616734.2012.672280
Swearer (Eds.), Bullying in American schools: A social-ecological per- Olweus, D. (1993). Bullying at school: What we know and what we can do.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

spective on prevention and intervention (Vol. 2, pp. 205–223). New Oxford, UK: Blackwell Publishing.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

York, NY: Routledge. Østergaard, S. D., Foldager, L., Allgulander, C., Dahl, A. A., Huuhtanen,
Leff, S. S., Patterson, C. J., Kupersmidt, J. B., & Power, T. J. (1999). M.-T., Rasmussen, I., & Munk-Jørgensen, P. (2010). Psychiatric case-
Factors influencing teacher identification of peer bullies and victims. ness is a marker of major depressive episode in general practice. Scan-
School Psychology Review, 28, 505. dinavian Journal of Primary Health Care, 28, 211–215. http://dx.doi
Lessne, D., & Cidade, D. (2015). Student reports of bullying and cyber- .org/10.3109/02813432.2010.501235
bullying: Results from the 2013 School Crime Supplement to the Na- Pavri, S. (2004). General and special education teachers’ preparation needs
tional Crime Victimization Survey. Washington, DC: U.S. Department of in providing social support: A needs assessment. Teacher Education and
Education, National Center for Education Statistics. Retrieved from Special Education, 27, 433– 443. http://dx.doi.org/10.1177/08884
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid⫽2015056 0640402700410
Pellegrini, A. D. (2002). Bullying, victimization, and sexual harassment
Li-Grining, C. P., Votruba-Drzal, E., Maldonado-Carreño, C., & Haas, K.
during the transition to middle school. Educational Psychologist, 37,
(2010). Children’s early approaches to learning and academic trajecto-
151–163. http://dx.doi.org/10.1207/S15326985EP3703_2
ries through fifth grade. Developmental Psychology, 46, 1062–1077.
Pellegrini, A. D., & Bartini, M. (2000). An empirical comparison of
http://dx.doi.org/10.1037/a0020066
methods of sampling aggression and victimization in school settings.
Little, R. J. A. (1988). A test of missing completely at random for
Journal of Educational Psychology, 92, 360 –366. http://dx.doi.org/10
multivariate data with missing values. Journal of the American Statis-
.1037/0022-0663.92.2.360
tical Association, 83, 1198 –1202. http://dx.doi.org/10.1080/01621459
Pianta, R. C., & Stuhlman, M. W. (2004). Teacher-child relationships and
.1988.10478722
children’s success in the first years of school. School Psychology Re-
Lynch, M., & Cicchetti, D. (1997). Children’s relationships with adults and
view, 33, 444 – 458.
peers: An examination of elementary and junior high school students.
Pouwels, J. L., Lansu, T. A. M., & Cillessen, A. H. N. (2016). Peer
Journal of School Psychology, 35, 81–99. http://dx.doi.org/10.1016/
victimization in adolescence: Concordance between measures and asso-
S0022-4405(96)00031-3
ciations with global and daily internalizing problems. Journal of Ado-
Mehta, S. B., Cornell, D., Fan, X., & Gregory, A. (2013). Bullying climate
lescence, 53, 195–206. http://dx.doi.org/10.1016/j.adolescence.2016.10
and school engagement in ninth-grade students. The Journal of School
.004
Health, 83, 45–52. http://dx.doi.org/10.1111/j.1746-1561.2012.00746.x R Core Team. (2017). R: A language and environment for statistical
Mikami, A. Y., Gregory, A., Allen, J. P., Pianta, R. C., & Lun, J. (2011). computing. Vienna, Austria: R Foundation for Statistical Computing.
Effects of a teacher professional development intervention on peer Retrieved from http://www.R-project.org/
relationships in secondary classrooms. School Psychology Review, 40, Reavis, R. D., Keane, S. P., & Calkins, S. D. (2010). Trajectories of peer
367–385. victimization: The role of multiple relationships. Merrill-Palmer Quar-
Murnane, R. J., & Willett, J. B. (2011). Methods matter: Improving causal terly, 56, 303–332. http://dx.doi.org/10.1353/mpq.0.0055
inference in educational and social science research. New York, NY: Reddy, R., Rhodes, J. E., & Mulhall, P. (2003). The influence of teacher
Oxford University Press. support on student adjustment in the middle school years: A latent
Murray, C., & Greenberg, M. T. (2000). Children’s relationship with growth curve study. Development and Psychopathology, 15, 119 –138.
teachers and bonds with school an investigation of patterns and corre- http://dx.doi.org/10.1017/S0954579403000075
lates in middle childhood. Journal of School Psychology, 38, 423– 445. Rigby, K. (2000). Effects of peer victimization in schools and perceived
http://dx.doi.org/10.1016/S0022-4405(00)00034-0 social support on adolescent well-being. Journal of Adolescence, 23,
Murray-Harvey, R., & Slee, P. T. (2010). School and home relationships 57– 68. http://dx.doi.org/10.1006/jado.1999.0289
and their impact on school bullying. School Psychology International, Rigby, K. (2001). Health consequences of bullying and its prevention. In J.
31, 271–295. http://dx.doi.org/10.1177/0143034310366206 Juvonen & S. Graham (Eds.), Peer harassment in school: The plight of
Nansel, T. R., Overpeck, M., Pilla, R. S., Ruan, W. J., Simons-Morton, B., the vulnerable and victimized (pp. 310 –331). New York, NY: Guilford
& Scheidt, P. (2001). Bullying behaviors among U.S. youth: Prevalence Press.
and association with psychosocial adjustment. Journal of the American Rodkin, P. C., & Hodges, E. V. (2003). Bullies and victims in the peer
Medical Association, 285, 2094 –2100. http://dx.doi.org/10.1001/jama ecology: Four questions for psychologists and school professionals.
.285.16.2094 School Psychology Review, 32, 384 – 400.
Nurmi, J.-E., & Kiuru, N. (2015). Students’ evocative impact on teacher Roeser, R. W., Midgley, C., & Urdan, T. C. (1996). Perceptions of the
instruction and teacher– child relationships: Theoretical background and school psychological environment and early adolescents’ psychological
an overview of previous research. International Journal of Behavioral and behavioral functioning in school: The mediating role of goals and
Development, 39, 445– 457. http://dx.doi.org/10.1177/016502541 belonging. Journal of Educational Psychology, 88, 408 – 422. http://dx
5592514 .doi.org/10.1037/0022-0663.88.3.408
 234 HUANG, LEWIS, COHEN, PREWETT, AND HERMAN

Rønning, J. A., Sourander, A., Kumpulainen, K., Tamminen, T., Niemelä, Troop-Gordon, W., & Kopp, J. (2011). Teacher-child relationship quality
S., Moilanen, I., . . . Almqvist, F. (2009). Cross-informant agreement and children’s peer victimization and aggressive behavior in late child-
about bullying and victimization among eight-year-olds: Whose infor- hood. Social Development, 20, 536 –561. http://dx.doi.org/10.1111/j
mation best predicts psychiatric caseness 10 –15 years later? Social .1467-9507.2011.00604.x
Psychiatry and Psychiatric Epidemiology, 44, 15–22. http://dx.doi.org/ Ttofi, M. M., & Farrington, D. P. (2011). Effectiveness of school-based
10.1007/s00127-008-0395-0 programs to reduce bullying: A systematic and meta-analytic review.
Roorda, D. L., Koomen, H. M. Y., Spilt, J. L., & Oort, F. J. (2011). The Journal of Experimental Criminology, 7, 27–56. http://dx.doi.org/10
influence of affective teacher–student relationships on students’ school .1007/s11292-010-9109-1
engagement and achievement: A meta-analytic approach. Review of Veenstra, R., Lindenberg, S., Oldehinkel, A. J., De Winter, A. F., Verhulst,
Educational Research, 81, 493–529. F. C., & Ormel, J. (2005). Bullying and victimization in elementary
Rothon, C., Head, J., Klineberg, E., & Stansfeld, S. (2011). Can social schools: A comparison of bullies, victims, bully/victims, and uninvolved
support protect bullied adolescents from adverse outcomes? A pro- preadolescents. Developmental Psychology, 41, 672– 682. http://dx.doi
spective study on the effects of bullying on the educational achieve- .org/10.1037/0012-1649.41.4.672
ment and mental health of adolescents at secondary schools in East Vreeman, R. C., & Carroll, A. E. (2007). A systematic review of school-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

London. Journal of Adolescence, 34, 579 –588. http://dx.doi.org/10 based interventions to prevent bullying. Archives of Pediatrics & Ado-
.1016/j.adolescence.2010.02.007 lescent Medicine, 161, 78 – 88. http://dx.doi.org/10.1001/archpedi.161
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Rueger, S. Y., Chen, P., Jenkins, L. N., & Choe, H. J. (2014). Effects .1.78
of perceived support from mothers, fathers, and teachers on depres- Waasdorp, T. E., Bradshaw, C. P., & Leaf, P. J. (2012). The impact
sive symptoms during the transition to middle school. Journal of ofschoolwide positive behavioral interventions and supports on bullyin-
Youth and Adolescence, 43, 655– 670. http://dx.doi.org/10.1007/ gand peer rejection: A randomized controlled effectiveness trial. Ar-
s10964-013-0039-x chivesof Pediatrics & Adolescent Medicine, 166, 149 –156. http://dx.doi
Rueger, S. Y., Malecki, C. K., & Demaray, M. K. (2008). Gender differ- .org/10.1001/archpediatrics.2011.755
ences in the relationship between perceived social support and student Wang, C., Swearer, S. M., Lembeck, P., Collins, A., & Berry, B. (2015).
adjustment during early adolescence. School Psychology Quarterly, 23, Teachers matter: An examination of student-teacher relationships,
496 –514. http://dx.doi.org/10.1037/1045-3830.23.4.496 attitudes toward bullying, and bullying behavior. Journal of Applied
Rutter, M. (1985). Resilience in the face of adversity. Protective factors School Psychology, 31, 219 –238. http://dx.doi.org/10.1080/153
and resistance to psychiatric disorder. The British Journal of Psychiatry, 77903.2015.1056923
147, 598 – 611. http://dx.doi.org/10.1192/bjp.147.6.598 Wang, M. T., & Eccles, J. S. (2013). School context, achievement moti-
Shin, Y., & Kim, H. Y. (2008). Peer victimization in Korean preschool vation, and academic engagement: A longitudinal study of school en-
children: The effects of child characteristics, parenting behaviors and gagement using a multidimensional perspective. Learning and Instruc-
teacher-child relationships. School Psychology International, 29, 590 – tion, 28, 12–23. http://dx.doi.org/10.1016/j.learninstruc.2013.04.002
605. http://dx.doi.org/10.1177/0143034308099203 Wentzel, K. R., Battle, A., Russell, S. L., & Looney, L. B. (2010). Social
Skinner, E., Furrer, C., Marchand, G., & Kindermann, T. (2008). Engage- supports from teachers and peers as predictors of academic and social
ment and disaffection in the classroom: Part of a larger motivational motivation. Contemporary Educational Psychology, 35, 193–202. http://
dynamic? Journal of Educational Psychology, 100, 765–781. http://dx dx.doi.org/10.1016/j.cedpsych.2010.03.002
.doi.org/10.1037/a0012840 Wienke Totura, C. M., Green, A. E., Karver, M. S., & Gesten, E. L.
Slee, P. T. (1995). Peer victimization and its relationship to depression (2009). Multiple informants in the assessment of psychological,
among Australian primary school students. Personality and Individ- behavioral, and academic correlates of bullying and victimization in
ual Differences, 18, 57– 62. http://dx.doi.org/10.1016/0191-8869(94) middle school. Journal of Adolescence, 32, 193–211. http://dx.doi
00114-8 .org/10.1016/j.adolescence.2008.04.005
Solberg, M. E., & Olweus, D. (2003). Prevalence estimation of school Wigfield, A., Lutz, S., & Laurel Wagner, A. (2005). Early adolescents’
bullying with the Olweus Bully/Victim Questionnaire. Aggressive Be- development across the middle school years: Implications for school
havior, 29, 239 –268. http://dx.doi.org/10.1002/ab.10047 counselors. Professional School Counseling, 9, 112–119. http://dx.doi
Stipek, D., & Miles, S. (2008). Effects of aggression on achievement: Does .org/10.5330/prsc.9.2.2484n0j255vpm302
conflict with the teacher make it worse? Child Development, 79, 1721– Willson, V. L., & Hughes, J. N. (2009). Who is retained in first grade? A
1735. http://dx.doi.org/10.1111/j.1467-8624.2008.01221.x psychosocial perspective. The Elementary School Journal, 109, 251–
Strøm, I. F., Thoresen, S., Wentzel-Larsen, T., & Dyb, G. (2013). Vio- 266. http://dx.doi.org/10.1086/592306
lence, bullying and academic achievement: A study of 15-year-old Yen, C.-F., Ko, C.-H., Yen, J.-Y., Tang, T.-C., Chang, Y.-P., & Cheng,
adolescents and their school environment. Child Abuse & Neglect, 37, C.-P. (2010). Internalizing and externalizing problems in adolescent
243–251. http://dx.doi.org/10.1016/j.chiabu.2012.10.010 aggression perpetrators, victims, and perpetrator-victims. Comprehen-
Swearer, S. M., Song, S. Y., Cary, P. T., Eagle, J. W., & Mickelson, W. T. sive Psychiatry, 51, 42– 48. http://dx.doi.org/10.1016/j.comppsych.2009
(2001). Psychosocial correlates in bullying and victimization: The rela- .03.002
tionship between depression, anxiety, and bully/victim status. Journal of
Emotional Abuse, 2, 95–121. http://dx.doi.org/10.1300/J135v02n02_07
Tierney, T., & Dowd, R. (2000). Use of social skills groups to support girls Received September 1, 2017
with emotional difficulties in secondary schools. Support for Learning, Revision received October 30, 2017
15, 82– 85. http://dx.doi.org/10.1111/1467-9604.00151 Accepted December 22, 2017 䡲