Child Abuse & Neglect 65 (2017) 112–123

Contents lists available at ScienceDirect

Child Abuse & Neglect

Research article

Does defending come with a cost? Examining the

psychosocial correlates of defending behaviour among
bystanders of bullying in a Canadian sample
Laura J. Lambe a,∗ , Chloe C. Hudson a , Wendy M. Craig a , Debra J. Pepler b
a
Department of Psychology, Queen’s University, Kingston, ON, Canada
b
Department of Psychology, York University, Toronto, ON, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Bullying is a form of interpersonal trauma that impacts all parties involved, including the
Received 26 April 2016 youth who witness the bullying. Some bystanders choose to intervene and defend the child
Received in revised form being bullied. Defending may be positively associated with psychosocial difficulties because
13 December 2016
youth are becoming more involved in a traumatic event, or because youth may be actively
Accepted 13 January 2017
coping with the distress elicited from witnessing bullying; however, the link between
Available online 26 January 2017
defending and psychosocial difficulties has not yet been examined. The current study inves-
tigated the age-related differences and psychosocial difficulties associated with defending
Keywords:
behaviour in school bullying. Data were collected from 5071 Canadian youth from Grades
Bullying
Defending 4–12. Participants completed an online survey at school, which assessed demographic infor-
Bystanders mation, recent defending behaviour, location and frequency of witnessing bullying, and
Witnesses psychosocial difficulties (internalizing, anger, psychosomatic, academic, and relationship
Trauma difficulties). A subsample of 1443 pure bystanders (no current bullying involvement) was
Peer intervention used for regression analyses. Defending behaviour was more common among girls and
among younger students. For boys, defending behaviour was associated with more psy-
chosocial difficulties compared to boys who only witnessed the bullying. This relationship
was less consistent for girls. Defending behaviour was also associated with more psychoso-
cial difficulties at high levels of bullying exposure. These associations suggest that defending
may come at a cost for youth, or that youth are defending their peers to cope with negative
emotions associated with witnessing interpersonal trauma. More longitudinal research is
needed to clarify these associations.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction

Bullying is a pervasive relationship problem, with an estimated 27%-36% of Canadian youth being involved in school
bullying (Craig, Lambe, & McIver, 2016). Bullying can be viewed from a trauma perspective, as it is associated with physi-
ological and emotional dysregulation (Jones & Barlow, 1990); however, little is known about the traumatic nature of peer
intervention in school bullying. Peers are present for 85% of these bullying incidents (Hawkins, Pepler, & Craig, 2001), cre-
ating ample opportunities for youth to be exposed to this form of interpersonal trauma. Indeed, it is estimated that youth

∗ Corresponding author at: Department of Psychology, Queen’s University, Humphrey Hall, 62 Arch Street, Kingston, ON, K7L 3N6, Canada.
E-mail address: laura.lambe@queensu.ca (L.J. Lambe).

http://dx.doi.org/10.1016/j.chiabu.2017.01.012
0145-2134/© 2017 Elsevier Ltd. All rights reserved.
 L.J. Lambe et al. / Child Abuse & Neglect 65 (2017) 112–123 113

witness bullying once every seven minutes (Pepler & Craig, 1995). After witnessing bullying, 19% of youth intervene to
support their peers who are being victimized (Hawkins et al., 2001). Peer defending is associated with a number of positive
outcomes, such as stopping the bullying (Hawkins et al., 2001; Salmivalli, Lagerspetz, Björkqvist, Österman, & Kaukiainen,
1996; Salmivalli, Voeten, & Poskiparta, 2011) and higher levels of self-esteem and peer acceptance for youth who have
been defended (Sainio, Veenstra, Huitsing, & Salmivalli, 2010). Intervention programs have been developed that encourage
students to take a defender role to reduce victimization (e.g., Polanin, Espelage, & Pigott, 2012); however, the psychosocial
correlates of defending behaviour are unknown. Youth who defend others take an active role in a traumatic event, and as
such, may be at greater risk for psychosocial difficulties than youth who do not defend their peers. In contrast, youth may
find witnessing the bullying traumatic, which may prompt them to intervene as an active way to cope with this distress.
The current study is the first to explore the age-related differences and psychosocial difficulties associated with defending
behaviour in school bullying.
Trauma can be broadly defined as any form of psychological abuse perpetrated towards an individual that has the poten-
tial to cause harm (Bernstein et al., 2003). While traumatic events are traditionally thought of as being adult-perpetrated,
peer-perpetrated events can be associated with similar levels of psychological harm and have detrimental developmental
consequences (Aber, Brown, Jones, Berg & Torrente, 2011). Traumatic events can have serious physiological and psycho-
logical consequences − in the short-term, children often react to interpersonal violence with an immediate fear response
(Jones & Barlow, 1990). In the long-term, children exposed to trauma experience a dysregulation of both physiological and
emotional responses, and maladaptive social cognitions. Like other forms of interpersonal trauma, both direct victimization
and witnessing bullying are associated with symptoms of post-traumatic stress disorder (Carney, 2008; Idsoe et al., 2012;
Janson & Hazler, 2004). Youth who experience interpersonal trauma are at greater risk for internalizing and externalizing
difficulties, relationship problems, and school difficulties compared to those who do not experience trauma (Holt, Buckley,
& Whelan, 2008; Margolin, Vickerman, Oliver, & Gordis, 2010).
The factors associated with interpersonal violence are not just experienced by youth who are directly involved in the
traumatic event. Youth who witness parental and community violence experience more psychiatric illnesses than their
peers who do not witness these traumatic events (Zinzow et al., 2009). Exposure to peer violence is no exception; children
who witness bullying experience numerous psychosocial difficulties, including somatic complaints, depression, anxiety,
substance use, interpersonal sensitivity, helplessness, and suicidal ideation (Rivers, Poteat, Noret, & Ashurst, 2009; Rivers &
Noret, 2013). In sum, exposure to bullying is a traumatic event that is associated with maladaptive outcomes for children
and youth.
The potential for adverse outcomes following trauma exposure may be moderated by a number of factors (Evans, Davies,
& DiLillo, 2008; Jones & Barlow, 1990), including the decision to intervene, individual differences (e.g., age and gender), and
the magnitude of trauma exposure. The decision to defend in school bullying is based on a number of personal, interpersonal,
and contextual factors that can vary over time (Chen, Chang, & Cheng, 2016). The dynamic nature of these contextual factors
may explain why defending behaviour is only moderately stable over time (Sijtsema, Rambaran, Caravita, & Gini, 2014), and
unstable among boys (Salmivalli, Lappalainen, & Lagerspetz, 1998). Thus, the current study conceptualized defending as a
behaviour.
Theoretically, there are several mechanisms through which defending behaviour may be associated with psychosocial
difficulties for youth. First, it is possible that defending behaviour is associated with difficulties because youth are actively
involved in a traumatic situation. For example, youth who attempt to stop interparental conflict experience more inter-
nalizing and externalizing problems over time compared their peers who simply witness the violent behaviour (Jouriles,
Rosenfield, McDonald & Mueller, 2014). Like youth who defend in situations of interparental conflict, it is possible that youth
who defend against bullying are using aggressive behaviours (e.g., yelling or using physical aggression) to stop the bullying.
Observational research indicates that physical aggression is a common type of defending behaviour, particularly among boys
(Hawkins et al., 2001). If successful, these aggressive behaviours are negatively reinforced, leading to a pervasive pattern
of aggressive behaviour and externalizing difficulties (Davis, Hops, Alpert, & Sheeber, 1998; Jouriles et al., 2014). In con-
trast, youth who attempt to defend and are not successful in stopping the bullying may become more depressed or anxious
(O’Brien, Margolin, & John, 1995). Secondly, it is also possible that youth may be defending their peers to cope with neg-
ative emotions associated with witnessing interpersonal trauma. Defending behaviour is associated with problem-solving
coping strategies (i.e., approach coping), suggesting that it is a socially competent way to manage the distress elicited from
observing peer victimization (Pozzoli & Gini, 2010). In sum, there are many ways through which defending behaviour may
be associated with difficulties for youth, although this has yet to be evaluated.
The psychosocial difficulties associated with defending behaviour may also depend on individual characteristics, such as
age and gender. Girls and younger students consistently engage in more defending behaviours than boys and older students
(Fox, Jones, Stiff, & Sayers, 2014; Salmivalli et al., 1996; Salmivalli & Voeten, 2004), suggesting that girls and younger students
may experience more psychosocial difficulties due to their higher rates of defending behaviour. In contrast, the prevalence
of psychosocial difficulties generally increases with development regardless of defending behaviour (Steinberg & Morris,
2001), suggesting that younger students may have some protection from these psychosocial difficulties. In sum, both age-
related and gender differences may play a role in understanding the psychosocial correlates of defending behaviour among
youth.
The cumulative effects of bullying exposure can be measured in terms of frequency of exposure and whether the expo-
sure is pervasive (i.e., the number of places bullying is witnessed). Research on other forms of trauma suggests that the
114 L.J. Lambe et al. / Child Abuse & Neglect 65 (2017) 112–123

frequency of trauma exposure is positively associated with psychosocial difficulties (Cooley-Quille, Boyd, Frantz, & Walsh,
2001). Additionally, the number of settings in which youth experience victimization predicts emotional problems (Ho &
Cheung, 2010), suggesting that as bullying becomes more pervasive in youths’ lives, they experience greater psychosocial
difficulties. Together, the frequency and number of locations in which youth witness bullying determines their level of
bullying exposure, which may be associated with psychosocial difficulties. Bullying exposure may interact with defending
behaviour to predict psychosocial difficulties. As bullying has a greater presence in youths’ lives, bullying is more likely to
be perceived as normative behaviour (Pozzoli, Gini, & Vieno, 2012). Defending behaviour may be more risky when bullying
is perceived as normative, as youth who engage in defending behaviour are then acting contrary to the established social
norm. In other words, defending behaviour may be associated with more psychosocial difficulties at high levels of bullying
exposure.

2. Present study

In the current study, we investigated the age-related differences and psychosocial difficulties associated with recent
defending behaviour in school bullying. First, we aimed to replicate and update previous research describing the prevalence
of defending behaviour among Canadian youth. Consistent with past research, we hypothesized that the rate of defending
behaviour would decrease with grade and that girls would be more likely to report defending behaviour than boys. In addi-
tion, the current study is the first to investigate the psychosocial correlates of defending behaviour, as previous research
in this area often examines bystanders without exploring distinctions between defending and witnessing behaviours. We
hypothesized that youth who recently defended their peers would report more internalizing, anger, psychosomatic, aca-
demic, and relationship difficulties compared to youth who did not report defending their peers the last time they witnessed
bullying. We explored whether grade, gender, and level of bullying exposure moderated the relationship between defending
behaviour and psychosocial difficulties.

3. Method

3.1. Participants

Data were collected from a large sample of Canadian youth who reported on their bullying involvement in three domains:
bullying others, being bullied, and/or witnessing bullying. The full sample consisted of 5071 participants and was used to
examine the prevalence of recent defending behaviour. Of these participants, 19% reported bullying others, 41% experienced
victimization, and 60% witnessed bullying (participants could endorse a maximum of two out of three categories). The full
sample included both boys (55%) and girls (45%) in grades 4–6 (53%), 7–8 (24%), 9–10 (14%), and 11–12 (9%). Most participants
identified as European-Canadian (55%), followed by Native-Canadian (14%), Asian-Canadian (4%), South Asian-Canadian
(4%), African/West Indian-Canadian (2%), Middle-Eastern Canadian (2%), and South/Latin American-Canadian (1%), with 89%
speaking English as their first language. Eleven percent of participants indicated that they did not know their ethnicity.
To examine the psychosocial correlates of recent defending among bystanders, we created a subsample that were not
directly involved in bullying. This subsample consisted of the 1443 participants who had witnessed bullying in the past
four weeks, but who had not bullied others or been victimized during the past four weeks.1 z-test of proportions indicated
that the subsample did not significantly differ from the full sample in terms of gender or grade distribution. The subsample
consisted of a lower proportion of Middle-Eastern Canadians and Native-Canadians, and a higher proportion of South-Asian
Canadians relative to the full sample.

3.2. Measures

3.2.1. Demographics. Participants were asked to report on their gender (0 = male, 1 = female) and grade. Four grade groups
were formed: mid-elementary school (grades 4–6), late elementary school (grades 7–8), early high school (grades 9–10),
and late high school (grades 11–12).

3.2.2. Defending behaviour. Participants were asked to think of the last time they saw or heard someone being bullied and
whether they tried to stop it. Participants self-reported whether or not they intervened the last time they witnessed bullying
using a forced choice response (0 = No [n = 535], 1 = Yes [n = 908]). Participants also responded to fourteen items that assessed
what they did last time the witnessed bullying. Four items reflected witnessing behaviour (e.g., “I stood and watched it”),
eight items reflected defending behaviour (e.g., “I stood up to the youth who was doing it”), and two items reflected assisting
the bullying behaviour (e.g., “I joined in with the bullying”). These responses were used to validate our classification of
recent defending behaviour. Youth who reported that they recently defended were significantly more likely to report using

1
Bullying and victimization were each assessed using a single item (victimization: “how often have you been bullied in the past four weeks?”, bullying:
“how often have you taken part in bullying other children in the past four weeks?”) assessed on a 1 (never) to 4 (always) Likert-type scale. Participants were
excluded from the subsample of bystanders if they reported any involvement in bullying or victimization in the past four weeks (i.e., scores of 2–4).
 L.J. Lambe et al. / Child Abuse & Neglect 65 (2017) 112–123 115

a defending behaviour (M = 1.89, SD = 1.60) the last time they witnessed bullying relative to youth who reported that they
did not recently engage in defending behaviour (M = 0.72, SD = 1.05), t(1441) = 15.12, p < 0.001, d = 1.12.

3.2.3. Frequency of witnessing bullying. Participants reported how often they observed seven types of bullying behaviour
during the past four weeks (e.g., “How often have you seen other children being bullied verbally in the past four weeks?”).
Responses were recorded on a four-point Likert-type scale (1 = never, 2 = once or twice, 3 = three or four times, 4 = more than
four times). Scores were averaged to assess the total frequency of witnessing bullying (␣ = 0.72).

3.2.4. Location of witnessing bullying. Participants reported on the number of locations they witnessed bullying during the past
four weeks from 14 locations: classroom, school hallway, gymnasium, outside on school grounds, school bathroom, locker
rooms, cafeteria, library, bus, bus stop, walking to and from school, off school grounds, cell phone, and online. Responses
were dichotomized (0 = did not observe bullying in this location, 1 = observed bullying in this location) and totaled (␣ = 0.82) to
assess the total number of locations bullying was witnessed.

3.2.5. Bullying exposure. Bullying exposure was operationally defined as the total frequency and total locations that youth
witnessed bullying during the past four weeks. The frequency and location measures (r = 0.42, p < 0.001) were standardized
and summed to create a total bullying exposure score, such that higher scores reflected a higher degree of bullying exposure.

3.2.6. Psychosocial Difficulties. Participants were presented with psychosocial difficulties that have been empirically linked
to bullying (e.g., Reijntjes, Kamphuis, Prinzie, & Telch, 2010; Rivers et al., 2009) and asked to indicate how much witnessing
bullying affected them in each domain (e.g., “How did seeing other children being bullied affect you? I have been sad.”).
Participants indicated on a four-point Likert scale how much they agreed with each statement (1 = strongly disagree, 2 = slightly
disagree, 3 = slightly agree, 4 = strongly agree). Five types of psychosocial difficulties were examined: internalizing, anger,
psychosomatic, academic, and relationship difficulties. Internalizing difficulties consisted of two items (“I have been sad”,
“I have been scared”), which were averaged to compute a total score (␣ = 0.64). Anger (“I have been angry”), psychosomatic
(“I have felt sick [e.g., headaches, stomach aches, trouble sleeping]”), and academic (“I have had difficulties in my school
[e.g., school work, grades]”) difficulties each consisted of a single item. Lastly, relationship difficulties consisted of four items
(“I have had trouble getting along with my family”, “I have had trouble getting along with my friends”, “I have had trouble
getting along with adults at my school (e.g., teachers, principal)”, and “I have had trouble getting along with adults in my
community [e.g., coaches, recreation leaders]”), which were averaged to compute a total score (␣ = 0.73). Overall, higher
scores indicate more difficulties.

3.3. Procedure

This study was granted ethical clearance according to the recommended principles of Canadian ethics guidelines and
through Queen’s University General Research Ethics Board. Elementary and high schools were recruited across Canada (from
various school districts) to participate in an online questionnaire that was completed at school. Students, their parents, and
their schools consented to the students’ participation in the survey.

3.4. Data analysis

All analyses were conducted in SPSS version 20.0.

4. Results

4.1. Prevalence of witnessing bullying

The full sample (i.e., youth who bullied others, were victimized, or witnessed bullying) was used to assess how often
youth witness bullying and the location that this occurs. As shown in Table 1, the most common locations that students
witnessed bullying were outside on school grounds (62%), off of school grounds (53%), school hallway (51%), in the classroom
(47%), on the bus (43%) and online (41%). We then explored whether the frequency and location of witnessing bullying varied
by grade, gender, and defending behaviour.

4.1.1. Grade. One-way analysis of variance (ANOVA) indicated a significant main effect of grade on the frequency of witness-
ing bullying, F(3, 4770) = 10.46, p < 0.001. Post-hoc comparisons showed that students in grades 11–12 (M = 2.29, SD = 1.13)
witnessed bullying more frequently compared to students in grades 4–6 (M = 2.03, SD = 1.03), grades 7–8 (M = 1.98, SD = 1.02),
and grades 9–10 (M = 2.10, SD = 1.06), all ps < 0.02. Students from grades 4 through 10 did not significantly differ from each
other (all ps > 0.09). We conducted Chi-square tests of independence to determine whether the specific location students
witnessed bullying varied by grade (see Table 1). In general, participants were more likely to witness bullying in each location
as they got older.
116 L.J. Lambe et al. / Child Abuse & Neglect 65 (2017) 112–123

Table 1
Percentage of youth who witness bullying by location, gender, and grade.

Location Male Female Gr 4–6 Gr 7–8 Gr 9–10 Gr 11–12

Classroom 41% 53%* 44%bcd 49%ad 49%ad 56%abc

School hallway 48% 55%* 40%bcd 59%acd 66%ab 71%ab
Gymnasium 32% 38%* 34% 35% 34% 39%
School grounds (outside) 58% 66%* 66%bcd 59%ac 51%abd 60%ac
School bathrooms 22% 24% 24%bcd 21%ad 17%ad 31%abc
Locker rooms 24% 22% 13%bcd 31%a 30%ad 39%ac
Cafeteria 29% 32% 22%bcd 34%acd 36%abd 48%abc
Library 12% 15%* 14%bd 10%ad 11%d 21%abc
Bus 37% 50%* 44%bd 40%ad 39%d 57%abc
Bus stop 17% 19% 17%d 16%cd 18%bd 27%abc
Walking to/from school 21% 21% 22%b 18%a 19% 23%
Off school grounds 49% 57%* 50%bd 54%ad 52%d 65%abc
Cellphone 22% 35%* 18%bcd 31%acd 37%abd 54%abc
Online 38% 46%* 29%bcd 48%acd 58%abd 69%abc

Note: Significant differences between genders are denoted * p < 0.05. Footnotes denote a significant difference between the grade specified and a grades
4–6, b grades 7–8, c grades 9–10, and d grades 11–12. p < 0.05.

4.1.2. Gender. An independent samples t-test indicated that girls witnessed bullying more frequently than boys,
t(4772) = 2.58, p = 0.01. Additionally, Chi-square tests of independence indicated that, compared to boys, girls witnessed
bullying more in the classroom, in school hallways, in the gymnasium, outside on school grounds, in the library, on the bus,
off of school grounds, on cellphones, and online (see Table 1).

4.1.3. Defending behaviour. We conducted an independent samples t-test to determine whether frequency of witnessing
bullying varied by defending behaviour. Recent defending was not associated with the frequency that youth witnessed
bullying, t(3040) = 0.16, p = 0.87.

4.2. Prevalence of defending behaviour

The full sample (i.e., youth who bullied others, were victimized, or witnessed bullying) was used to assess the prevalence of
recent defending behaviour. Sixty-four percent of participants self-reported that they intervened the last time they witnessed
bullying. We then explored whether the likelihood of defending behaviour varied by grade, gender, and ethnicity.

4.2.1. Grade. A Chi-square test of independence indicated a significant interaction between grade and defending behaviour,
␹2 (3) = 138.47, p < 0.001. Follow-up analyses indicated that participants in grades 4–6 (72%) were significantly more likely
to report recently defending their peers than participants in grades 7–8 (62%), grades 9–10 (43%), or grades 11–12 (56%).
Participants in grades 7–8 and grades 11–12 were more likely to report recent defending than participants in grades 9–10.
There were no differences in recent defending between participants in grades 7–8 and grades 11–12.

4.2.2. Gender. Similarly, a Chi-square test of independence indicated a significant interaction between gender and defending
behaviour, ␹2 (1) = 29.17, p < 0.001. Girls (69%) were more likely to report recent defending than boys (59%).

4.2.3. Grade by gender. A Chi-square test of independence indicated a significant interaction for both boys, ␹2 (3) = 87.46,
p < 0.001, and girls, ␹2 (3) = 43.80, p < 0.001. Overall, older boys were less likely to engage in defending behaviour compared
to younger boys. Boys in grades 4–6 (68%) were significantly more likely to report defending behaviour than boys in grades
7–8 (60%), 9–10 (38%), and 11–12 (48%). Boys in grades 7–8 were significantly more likely to report defending behaviour
than boys in grades 9–12. Finally, boys in grades 9–10 were significantly less likely to report defending behaviour than boys
in grades 11–12. The pattern of results was similar for girls. Girls in grades 4–6 (75%) were significantly more likely to report
defending behaviour than girls in grades 7–8 (64%), 9–10 (52%), and 11–12 (62%). Girls in grades 7–8 were significantly more
likely to report defending behaviour than girls in grades 9–10 (all ps < 0.05). There were no differences in the frequencies of
defending behaviour for girls in grades 9–10 and 11–12.

4.2.4. Ethnicity. A Chi-square test of independence indicated a significant interaction between ethnicity and defending
behaviour, ␹2 (8) = 38.27, p < 0.001. South/Latin American-Canadian participants were most likely to report defending
behaviour (85%), followed by South Asian-Canadian (75%), African/West Indian-Canadian (67%), Middle Eastern-Canadian
(67%), European-Canadian (64%), Native Canadian (56%), and Asian Canadian (44%). Asian-Canadian participants were signif-
icantly less likely to report defending behaviour compared to participants of other ethnicities. Native Canadian participants
were less likely to report defending behaviour compared to participants who were European-Canadian, South Asian-
Canadian, and South/Latin American-Canadian. Participants who were European-Canadian were less likely to report recent
defending than participants who were South Asian-Canadian.
 L.J. Lambe et al. / Child Abuse & Neglect 65 (2017) 112–123 117

Table 2
Bivariate correlations and descriptive statistics for regression variables among bystanders.

1. 2. 3. 4. 5. 6. 7. 8. 9.

1. Gender –
2. Grade group −0.08** –
3. Defending 0.10** −0.18** –
4. Bullying Exposure 0.05 0.17** 0.09** –
5. Internalizing 0.22** −0.14** 0.21** 0.17** –
6. Anger 0.18** 0.09** 0.15** 0.21** 0.47** –
7. Psychosomatic 0.13** −0.04 0.18** 0.27** 0.41** 0.31** –
8. Academic 0.06* −0.06* 0.09** 0.18** 0.32** 0.19** 0.44** –
9. Relationship −0.001 0.06* 0.001 0.28** 0.19** 0.20** 0.39** 0.49** –
Mean – – – −0.28 2.56 2.88 1.93 1.69 1.47
SD – – – 0.87 0.86 1.03 1.07 0.95 0.60

Note: N = 1443. Gender was coded as male = 0, female = 1. Defending behaviour was coded as 0 = witnessing behaviour, 1 = defending behaviour. ** p < 0.01,
* p < 0.05. Means and standard deviations are presented only for continuous variables.

Table 3
Simultaneous multiple regression predictors of internalizing difficulties.

B t SE 95% CI B

Grade Group −0.06 −1.66 0.04 [−0.14, 0.01]

Gender 0.55*** 4.72 0.12 [0.18, 0.57]
Defending .0.45*** 4.20 .0.11 [0.24, 0.0.65]
Exposure 0.04 1.16 0.04 [−0.03, 0.11]
Grade Group x Gender −0.05 −1.05 0.04 [−0.13, 0.0.04]
Grade Group x Defending −0.04 −0.98 0.05 [−0.13, 0.04]
Grade Group x Exposure 0.01 0.39 0.01 [−0.02, 0.03]
Gender x Defending −0.22** −2.41 0.09 [−0.41, −0.04]
Gender x Exposure 0.04 1.44 0.03 [−0.01, 0.09]
Defending x Exposure 0.03 1.10 0.03 [−0.03, 0.09]

Note: N = 1443. ** p < 0.001, ** p < 0.01, * p < 0.05.

4.3. Psychosocial difficulties associated with defending behaviour

For all subsequent analyses, the subsample of pure bystanders (i.e., individuals who witnessed bullying, but had not
bullied others and were not victimized themselves) was used to examine the psychosocial difficulties in association with
recent defending behaviour. For each psychosocial difficulty, a simultaneous linear regression was conducted with gender,
grade group, defending behaviour, bullying exposure, and all two-way interactions entered as predictor variables. There
was no missing data and the dependent variables were all approximately normally distributed (skewness < 1.54). Bivariate
correlations and descriptive statistics are presented in Table 2. Unstandardized regression coefficients are reported for all
analyses.

4.3.1. Internalizing difficulties. The regression model accounted for a significant proportion of the variance within internaliz-
ing difficulties, R2 = 0.12, F(10, 1432) = 20.01, p < 0.001. As shown in Table 3, gender and defending behaviour were positively
associated with internalizing difficulties; however, these main effects were qualified by a significant gender by defending
behaviour interaction. For boys, defending was positively associated with internalizing difficulties, p < 0.001, d = 0.50 (see
Fig. 1). This difference was not observed among girls, p = 0.97.

4.3.2. Anger difficulties. The model explained a significant proportion of the variance within anger difficulties, R2 = 0.08, F(10,
1432) = 12.71, p < 0.001. As shown in Table 4, defending behaviour was positively associated with anger difficulties. Youth
who reported recent defending behaviour (M = 3.00, SD = 1.02) were more likely to report anger difficulties compared to
youth who did not recently defend (M = 2.67, SD = 1.02), p < 0.05, d = 0.32.

4.3.3. Psychosomatic difficulties. The model explained a significant proportion of the variance within psychosomatic dif-
ficulties, R2 = 0.12, F(10, 1432) = 19.25, p < 0.001. Both defending behaviour and bullying exposure emerged as significant
correlates; however, these associations were qualified by a significant bullying exposure by defending behaviour interac-
tion (see Table 5). Simple slopes analyses were used to further explore this effect. As shown in Fig. 2, defending behaviour
was positively associated with psychosomatic difficulties at a high level of bullying exposure (p < 0.001, d = 0.66) but not at
low levels of bullying exposure (p = 0.07).

4.3.4. Academic difficulties. The model explained a significant proportion of the variance within academic difficulties,
R2 = 0.06, F(10, 1432) = 9.13, p < 0.001. As shown in Table 6, both grade group and defending behaviour were significant
correlates of academic difficulties; however, these findings were once again qualified by significant gender by defending
118 L.J. Lambe et al. / Child Abuse & Neglect 65 (2017) 112–123

Fig. 1. Levels of internalizing, academic, and relationship difficulties reported by youth who reported defending or witnessing behaviours the last time
they witnessed bullying, separated by gender. * p < 0.05.

Table 4
Simultaneous multiple regression predictors of anger difficulties.

B t SE 95% CI B

Grade Group 0.07 1.53 0.05 [−0.02, 0.16]

Gender 0.22 1.53 0.14 [−0.06, 0.50]
Defending 0.30* 2.27 0.12 [0.04, 0.55]
Exposure 0.08 1.75 0.04 [−0.01, 0.16]
Grade Group x Gender 0.02 0.29 0.05 [−0.09, 0.12]
Grade Group x Defending 0.02 0.36 0.06 [−0.09, 0.13]
Grade Group x Exposure 0.01 0.91 0.02 [−0.02, 0.04]
Gender x Defending −0.06 −0.54 0.11 [−0.28, 0.16]
Gender x Exposure −0.01 −0.19 0.03 [−0.07, 0.06]
Defending x Exposure 0.02 0.47 0.04 [−0.05, 0.08]

Note: N = 1443. * p < 0.05.

Table 5
Simultaneous multiple regression predictors of psychosomatic difficulties.

B t SE 95% CI B

Grade Group −0.06 −1.24 0.05 [−0.15, 0.03]

Gender 0.24 1.65 0.15 [−0.05, 0.52]
Defending 0.41** 3.11 0.13 [0.15, 0.67]
Exposure 0.11* 2.45 0.04 [0.02, 0.19]
Grade Group x Gender 0.04 0.73 0.06 [−0.07, 0.15]
Grade Group x Defending −0.02 −0.41 0.06 [−0.13, 0.09]
Grade Group x Exposure −0.01 −0.59 0.02 [−0.04, 0.02]
Gender x Defending −0.15 −1.28 0.11 [−0.73, 0.08]
Gender x Exposure 0.03 0.73 0.03 [−0.04, 0.09]
Defending x Exposure 0.10** 2.86 0.04 [0.03, 0.17]

Note: N = 1443. ** p < 0.01, * p < 0.05.

Table 6
Simultaneous multiple regression predictors of academic difficulties.

B t SE 95% CI B

Grade Group −0.12** −2.69 0.04 [−0.20, −0.03]

Gender 0.07 0.54 0.13 [−0.19, 0.33]
Defending 0.24* 1.99 0.12 [0.003, 0.48]
Exposure 0.06 1.52 0.04 [−0.02, 0.14]
Grade Group x Gender 0.10* 2.08 0.05 [0.01, 0.20]
Grade Group x Defending −0.01 −0.21 0.05 [−0.11, 0.09]
Grade Group x Exposure −0.01 −0.49 0.01 [−0.03, 0.02]
Gender x Defending −0.30** −2.80 0.11 [−0.50, −0.09]
Gender x Exposure 0.04 1.31 0.03 [−0.02, 0.10]
Defending x Exposure 0.06 1.89 0.03 [−0.002, 0.12]

Note: N = 1443. ** p < 0.01, * p < 0.05.

 L.J. Lambe et al. / Child Abuse & Neglect 65 (2017) 112–123 119

Fig. 2. Levels of psychosomatic and relationship difficulties reported by youth who reported defending or witnessing behaviours the last time they witnessed
bullying, across levels of bullying exposure. * p < 0.05.

behaviour interaction. For boys, defending behaviour was positively associated with academic difficulties, p < 0.001, d = 0.32
(see Fig. 1). For girls, defending behaviour was negatively associated with academic difficulties, p < 0.05, although the mag-
nitude of this difference was very small, d = 0.03. There was also a significant interaction between grade group and gender.
In grades 4–6, boys reported more academic difficulties than girls, p < 0.05; in grades 9–10, girls reported more academic
difficulties than boys, p < 0.05. No gender difference was observed in grades 7–8 and 11–12, ps > 0.37.

4.3.5. Relationship difficulties. The model explained a significant proportion of the variance within relationship difficulties,
R2 = 0.10, F(10, 1432) = 15.49, p < 0.001. As shown in Table 7, bullying exposure was positively associated with relationship
difficulties. There was also a significant gender by defending behaviour interaction. Distinct from prior results, boys who
reported recent defending behaviour did not differ in their level of relationship difficulties from boys who did not recently
defend, p = 0.32. For girls, those who did not recently defend had higher levels of relationship difficulties relative to girls who
reported recent defending behaviour, p < 0.001, d = 0.20 (see Fig. 1). Lastly, a significant interaction between bullying exposure
and defending behaviour emerged. As shown in Fig. 2, defending behaviour was positively associated with relationship
difficulties only at high levels of bullying exposure, p < 0.01, d = 0.28.

5. Discussion

Unlike other forms of interpersonal trauma, bullying frequently occurs in the public, presenting the opportunity for
many youth to witness the traumatic event. Consistent with previous research (e.g., Rapp-Paglicci, Dulmus, Sowers, &
Theriot, 2004), the results of the current study suggest that youth frequently witness bullying in a variety of locations.
Girls tend to report witnessing bullying more frequently than boys; both boys and girls tend to report witnessing bullying
more frequently as they get older. We also found that the majority of youth reported that they recently engaged in defending
behaviour, with the likelihood of defending varying as a function of grade, gender, and ethnicity. Consistent with the current
literature, girls were more likely to report recently defending than boys, and younger students were more likely to report
recently defending than older students (Fox et al., 2014; Salmivalli et al., 1996). Finally, results suggested that even periph-

Table 7
Simultaneous multiple regression predictors of relationship difficulties.

B t SE 95% CI B

Grade Group −0.02 −0.88 0.03 [−0.08, 0.03]

Gender 0.05 0.57 0.08 [−0.12, 0.21]
Defending 0.001 0.01 0.08 [−0.15, 0.15]
Exposure 0.07** 2.84 0.03 [0.02, 0.12]
Grade Group x Gender 0.04 1.22 0.03 [−0.02, 0.10]
Grade Group x Defending 0.03 0.96 0.03 [−0.03, 0.09]
Grade Group x Exposure −0.01 −0.80 0.01 [−0.02, 0.01]
Gender x Defending −0.20** −3.01 0.07 [−0.33, −0.07]
Gender x Exposure 0.001 −0.01 0.02 [−0.04, 0.02]
Defending x Exposure 0.07** 3.22 0.02 [0.03, 0.10]

Note: N = 1443. ** p < 0.01.

120 L.J. Lambe et al. / Child Abuse & Neglect 65 (2017) 112–123

eral involvement in bullying is associated with psychosocial difficulties for youth. Higher levels of bullying exposure and
defending behaviour were associated with higher levels of psychosocial difficulties. These findings suggest that defending
behaviour may either: 1) come with a psychosocial cost for youth (Jones & Barlow, 1990), or 2) be a reaction to the psy-
chosocial difficulties associated with witnessing a traumatic event (Barhight, Hubbard, & Hyde, 2013). It is also possible that
there is a bi-directional relationship between psychosocial difficulties and defending behaviour. For example, youth may
feel distressed from witnessing bullying, defend, and then experience additional distress from becoming more involved in
the traumatic event.

5.1. Prevalence of recent defending

Contrary to previous research which reports that only a fifth of youth engage in defending behaviour (Hawkins et al.,
2001; Salmivalli et al., 1996), two thirds of youth in the current study reported that they engaged in defending behaviour
the last time they witnessed bullying. Rates of defending may have risen because defending behaviour has been encouraged
as an effective way to manage school bullying (e.g., Polanin et al., 2012). Youth may be internalizing the suggestions of
adults and engaging in defending behaviour to support their peers. Another possibility is that youth may be over-estimating
the frequency in which they defend because defending is an encouraged and “desirable” behaviour. Future research using
naturalistic designs or multi-respondent reports is needed to objectively measure current rates of defending.
As hypothesized and consistent with previous research (Fox et al., 2014; Salmivalli et al., 1996), girls and younger students
are more likely to report recent defending behaviour than boys and older students. Girls also report witnessing bullying more
often than boys, which suggests that they have more opportunities to act or are better able to recognize bullying situations.
In contrast, younger students were more likely to report recent defending behaviour despite being exposed to fewer bullying
interactions than older students. Older students may be less likely to defend because pro-bullying attitudes increase with
development (Salmivalli & Voeten, 2004). It is also possible that older students report witnessing more bullying interactions
because they are better able to recognize and label the event as bullying.

5.2. Psychosocial difficulties associated with defending behaviour

The current study is the first to explore whether defending behaviour is associated with psychosocial difficulties for
youth. The results of the current study partially support our hypotheses: the associations between defending behaviour
and psychosocial difficulties were more consistently observed for boys than for girls. Due to the cross-sectional nature of
the current study, it is unknown whether these psychosocial difficulties are the antecedents or consequences of defending
behaviour. While recent defending behaviour was associated with more anger, psychosomatic, and academic difficulties for
both boys and girls, defending behaviour was associated with internalizing difficulties for boys only.
Defending behaviour may be an attempt to cope with the emotional and physiological distress elicited from witnessing
bullying. Witnessing bullying is associated with distress among bystanders (Rivers et al., 2009), and thus it is possible that
psychosocial distress is the antecedent of defending behaviour. Experimental research demonstrates that youth respond
differently to witnessing bullying, with some youth exhibiting stronger emotional and physiological reactions than others
(Barhight et al., 2013; Caravita, Colombo, Stefanelli, & Zigliani, 2016). Furthermore, having an emotional reaction to wit-
nessing bullying is a significant predictor of defending behaviour (Barhight et al., 2013). An emotional reaction may be a
critical factor in the decision to intervene, especially for boys. This emotional reaction may explain why internalizing diffi-
culties were associated with defending behaviour among boys only, as boys may need more of an emotional reaction to elicit
defending behaviour compared to girls. In other words, the motivation to engage in defending behaviours may be different
for boys and girls. Girls may choose to defend due to social pressures, while boys may choose to defend because the bullying
elicited distress. This interpretation is consistent with previous research which indicates that social self-efficacy is most
strongly associated with defending among girls, whereas empathy is most strongly associated with defending among boys
(Cappadocia, Pepler, Cummings, & Craig, 2012).
Another potential interpretation of this finding is that boys may be more likely to engage in maladaptive defending
behaviours (e.g., physical aggression). Boys tend to use direct forms of aggression in social situations, such as pushing
or using profanity (Björkqvist, Lagerspetz, & Kaukiainen, 1992). Moreover, observational research indicates that physical
aggression is the most common type of defending behaviour among boys, whereas verbal assertion is the most common
type of defending behaviour among girls (Hawkins et al., 2001). Over time, aggressive defending behaviour may develop
into a more pervasive pattern of aggression and psychosocial difficulty among boys. Thus, it is also possible that defending
behaviour is the antecedent of increased psychosocial difficulties. Future research should examine the impact of specific
types of defending behaviours, as well as their motivations, to provide a more nuanced understanding of this relationship.
For girls, the same pattern of results arose with respect to anger, psychosomatic and academic difficulties, such that
defending behaviour was positively associated with psychosocial difficulties in these areas. Unexpectedly, girls who reported
recent defending behaviour reported fewer relationship difficulties compared to girls who had not recently defended. There
are several possible interpretations for this finding. Girls tend to be more opposed to bullying than boys (Salmivalli &
Voeten, 2004), suggesting it may be more socially normative for girls to exhibit defending behaviour. In a review of sex
differences of peer relationship processes, girls reported more interpersonal sensitivity, more empathy for others, and were
more relationship-focused than boys (Rose & Rudolph, 2006). In response to stress, girls are also more likely to seek social
 L.J. Lambe et al. / Child Abuse & Neglect 65 (2017) 112–123 121

support than boys, possibly conferring a protective advantage (Rose & Rudolph, 2006). In other words, defending behaviour
may be more socially normative for girls because of their relationship-focused socialization processes. This process may also
explain why we found that the potential social cost of peer intervention is actually associated with not defending for girls,
as this behaviour would be contrary to the established group norm for girls. It is also possible that girls are defending in
more adaptive ways than boys (e.g., seeking social support from an adult) as previous observational research indicates that
the majority (62%) of girls’ defending behaviours are nonaggressive (Hawkins et al., 2001). In contrast, girls who have fewer
relationship problems may have the social skills and status to successfully defend their peers. It has been suggested that
high social status is a requirement for defending (Pöyhönen, Juvonen, & Salmivalli, 2010), with recent research indicating
that popularity is positively associated with defending among girls only (Duffy, Penn, Nesdale, & Zimmer-Gembeck, 2016).
Limited research, however, has examined the associations between social status and defending behaviour over time. Indeed,
Meter and Card (2015) found that peer-perceived liking was not associated with concurrent defending behaviour, and
that defending behaviour was associated with a decrease in peer-perceived liking one year later. Overall, these varied
results emphasize the need for future research to consider how gender and social status influence the associations between
defending behaviour and relationship problems over time.
There also appears to be a dose-response relationship between defending behaviour and both psychosomatic and rela-
tionship difficulties: the higher the level of bullying exposure, the stronger the association was between recent defending
behaviour and these psychosocial difficulties. Exposure to bullying in a school context where it is both frequent and perva-
sive suggests that bullying may be perceived as a normal occurrence. By defending their peers, youth may be acting against
group norms, which may make defending behaviour more challenging. In other words, defending behaviour may be more
risky for youth when bullying is more frequent and pervasive, and therefore perceived to be more socially acceptable. Fur-
thermore, the risks associated with defending behaviour may be compounded with the risk of cumulative bullying exposure.
Overall, these findings suggest that youth may need more support to successfully defend in environments with a high level
of bullying exposure.

5.3. Limitations and future directions

The current study is limited by its cross-sectional design, which precludes an understanding of the directionality of the
effects. Longitudinal research is needed to examine the temporal associations between defending behaviour and psychosocial
difficulties among youth. Information regarding the specific type of defending behaviour used (e.g., aggression, seeking
social support) and its success in stopping the bullying may help to elucidate the link between defending behaviour and
psychosocial difficulties. For example, it is possible that certain defending behaviours are associated with psychosocial
difficulties, whereas other forms may be more adaptive. Future research also may wish to use peer nomination to measure
defending behaviour in order to rule out possible social desirability effects. The current findings pertain only to youth who
are involved in the bullying solely as bystanders, and are thus not generalizable to youth who both witness bullying and are
directly involved in bullying others and/or being victimized. In addition, our sample sizes for some ethnic groups were small.
Future research, using diverse samples, is needed to explore whether the psychosocial correlates of defending behaviour
vary by ethnicity. Such research may also wish to examine the role of early traumatic experiences, which may be associated
with both the likelihood of defending and with psychosocial difficulties
The association between defending behaviour and psychosocial difficulties may depend on certain individual or contex-
tual variables, such as social status and school climate. Youth with higher social status are more likely to defend (Pöyhönen
et al., 2010), suggesting that these youth may be less likely to experience psychosocial difficulties. It is also possible that
a school climate supportive of defending behaviour might buffer the associations between defending behaviour and psy-
chosocial difficulties. Defending behaviour is more common in schools with more normative pressure to intervene (Pozzoli
et al., 2012), suggesting that defending may be associated with fewer difficulties (or none at all) when youth feel supported
by their peers. Future research should consider such potential moderator variables to gain a more holistic understanding of
the psychosocial correlates of defending behaviour. Lastly, it should be noted that our regression models explained only a
small to moderate amount of the variance within psychosocial difficulties, suggesting that other factors may better explain
these difficulties.

6. Conclusion

Despite being frequently encouraged by educators (e.g., Polanin et al., 2012), defending behaviour is associated with
a number of psychosocial difficulties for youth. This association may be explained through increased involvement in a
traumatic event, through coping with the distress of witnessing peer victimization, or a combination of these theoretical
mechanisms. Despite the psychosocial difficulties associated with defending behaviour, the majority of youth report that
they intervene when they witness bullying. Although peers can be effectively included in bullying prevention efforts, policy
makers and educators must also consider the potential difficulties associated with defending behaviour. Research is required
to further explore the risks and benefits of defending behaviour in the complex social environments youth experience. This
research will help to ensure that bullying interventions can effectively reduce the adverse effects of bullying and promote
social justice for all parties involved.
122 L.J. Lambe et al. / Child Abuse & Neglect 65 (2017) 112–123

Funding

This work was supported by the Networks of Centres of Excellence [grant number 415604-12]. Laura Lambe and Chloe
Hudson were supported by Ontario Graduate Scholarships.

Acknowledgements

PREVNet gratefully acknowledges the contributions/participation of the following individuals in the creation of the PRE-
VNet School Safety Survey: Dr. Wendy Craig, Dr. Shelley Hymel, Dr. Wendy Josephson, Dr. John LeBlanc, Dr. Debra Pepler,
Dr. Darcy Santor, Dr. David Smith, Dr. Leslie Tutty, Dr. Tracy Vaillancourt and their colleagues J. Bradley Cousins, Charles
Cunningham, Lesley Cunningham, Clinton Davis, Eric Duku, Amanda Krygsman, Danielle Law, Patricia McDougall, Jessie
Miller, Ken Neale, Brent Olson, Jennifer Shapka, Kathy Short, Rebecca Stewart, Kelly Stiver, Jessica Trach, Vi Trinh, and Terry
Waterhouse.
PREVNet also wishes to acknowledge overlap of some items in the Ontario Ministry of Education’s School Climate Surveys,
“Bullying/Harassment” section that was based on a collaboration with Drs. Craig and Pepler and the Ontario Ministry of
Education.

References

Aber, L., Brown, J. L., Jones, S. M., Berg, J., & Torrente, C. (2011). School-based strategies to prevent violence, trauma, and psychopathology: The challenges
of going to scale. Development and Psychopathology, 23(02), 411–421. http://dx.doi.org/10.1017/S0954579411000149
Barhight, L. R., Hubbard, J. A., & Hyde, C. T. (2013). Children’s physiological and emotional reactions to witnessing bullying predict bystander intervention.
Child Development, 84(1), 375–390. http://dx.doi.org/10.1111/j.1467-8624.2012.01839.x
Bernstein, D. P., Stein, J. A., Newcomb, M. D., Walker, E., Pogge, D., Ahluvalia, T., et al. (2003). Development and validation of a brief screening version of
the Childhood Trauma Questionnaire. Child Abuse & Neglect, 27(2), 169–190. http://dx.doi.org/10.1016/s0145-2134(02)00541-0
Björkqvist, K., Österman, K., & Kaukiainen, A. (1992). The development of direct and indirect aggressive strategies in males and females. In K. Bjorkqvist, &
P. Niemela (Eds.), Of mice and women: Aspects of female aggression. London: Academic Press.
Cappadocia, M. C., Pepler, D., Cummings, J. G., & Craig, W. (2012). Individual motivations and characteristics associated with bystander intervention during
bullying episodes among children and youth. Canadian Journal of School Psychology, 27(3), 201–216. http://dx.doi.org/10.1177/0829573512450567
Caravita, S. C., Colombo, B., Stefanelli, S., & Zigliani, R. (2016). Emotional, psychophysiological and behavioral responses elicited by the exposition to
cyberbullying situations: Two experimental studies. Psicología Educativa, 22(1), 49–59. http://dx.doi.org/10.1016/j.pse.2016.02.003
Carney, J. (2008). Perceptions of bullying and associated trauma during adolescence. Professional School Counseling, 11(3), 179–188.
http://dx.doi.org/10.5330/PSC.n.2010–11.179
Chen, L.-M., Chang, L. Y. C., & Cheng, Y.-Y. (2016). Choosing to be a defender or an outsider in a school bullying incident: Determining factors and the
defending process. School Psychology International, 37(3), 289–302. http://dx.doi.org/10.1177/0143034316632282
Cooley-Quille, M., Boyd, R. C., Frantz, E., & Walsh, J. (2001). Emotional and behavioral impact of exposure to community violence in inner-city adolescents.
Journal of Clinical Child Psychology, 30(2), 199–206. http://dx.doi.org/10.1207/S15374424JCCP3002 7
Craig, W., Lambe, L., & McIver, T. (2016). Bullying and fighting. In J. Freeman, M. King, & W. Pickett (Eds.), Health-Behaviour in school-aged children (HBSC)
in Canada: Focus on relationships (pp. 145–160). Public Health Agency of Canada.
Davis, B. T., Hops, H., Alpert, A., & Sheeber, L. (1998). Child responses to parental conflict and their effect on adjustment: A study of triadic relations.
Journal of Family Psychology, 12(2), 163 [10.1037/0893-3200.12.2.163].
Duffy, A. L., Penn, S., Nesdale, D., & Zimmer-Gembeck, M. J. (2016). Popularity: Does it magnify associations between popularity prioritization and the
bullying and defending behavior of early adolescent boys and girls? Social Development, 1(-15) http://dx.doi.org/10.1111/sode.12206
Evans, S. E., Davies, C., & DiLillo, D. (2008). Exposure to domestic violence: A meta-analysis of child and adolescent outcomes. Aggression and Violent
Behavior, 13(2), 131–140. http://dx.doi.org/10.1016/j.avb.2008.02.005
Fox, C. L., Jones, S. E., Stiff, C. E., & Sayers, J. (2014). Does the gender of the bully/victim dyad and the type of bullying influence children’s responses to a
bullying incident? Aggressive Behaviour, 40, 359–368. http://dx.doi.org/10.1002/ab.21529
Hawkins, D. L., Pepler, D. J., & Craig, W. M. (2001). Naturalistic observations of peer interventions in bullying. Social Development, 10, 512–527.
http://dx.doi.org/10.1111/1467-9507.00178
Ho, M. Y., & Cheung, F. M. (2010). The differential effects of forms and settings of exposure to violence on adolescents’ adjustment. Journal of Interpersonal
Violence, 25(7), 1309–1337. http://dx.doi.org/10.1177/0886260509340548
Holt, S., Buckley, H., & Whelan, S. (2008). The impact of exposure to domestic violence on children and young people: A review of the literature. Child
Abuse & Neglect, 32(8), 797–810. http://dx.doi.org/10.1016/j.chiabu.2008.02.004
Idsoe, T., Dyregrov, A., & Idsoe, E. C. (2012). Bullying and PTSD symptoms. Journal of Abnormal Child Psychology, 40(6), 901–911.
http://dx.doi.org/10.1007/s10802-012-9620-0
Janson, G. R., & Hazler, R. J. (2004). Trauma reactions of bystanders and victims to repetitive abuse experiences. Violence and Victims, 19(2), 239–255.
http://dx.doi.org/10.1891/vivi.19.2.239.64102
Jones, J. C., & Barlow, D. H. (1990). The etiology of posttraumatic stress disorder. Clinical Psychology Review, 10(3), 299–328.
http://dx.doi.org/10.1016/0272-7358(90)90064-H
Jouriles, E. N., Rosenfield, D., McDonald, R., & Mueller, V. (2014). Child involvement in interparental conflict and child adjustment problems: A
longitudinal study of violent families. Journal of Abnormal Child Psychology, 42, 693–704. http://dx.doi.org/10.1007/s10802-013-9821-1
Margolin, G., Vickerman, K. A., Oliver, P. H., & Gordis, E. B. (2010). Violence exposure in multiple interpersonal domains: Cumulative and differential
effects. Journal of Adolescent Health, 47(2), 198–205. http://dx.doi.org/10.1016/j.jadohealth.2010.01.020
Meter, D. J., & Card, N. A. (2015). Effects of defending: The longitudinal relations among peer-percevied defending of victimized peers, victimization, and
liking. Social Development, 24(4), 734–747. http://dx.doi.org/10.1111/sode.12129
O’Brien, M., Margolin, G., & John, R. S. (1995). Relation among marital conflict, child coping, and child adjustment. Journal of Clinical Child Psychology, 24(3),
346–361. http://dx.doi.org/10.1207/s15374424jccp2403 12
Pöyhönen, V., Juvonen, J., & Salmivalli, C. (2010). What does it take to stand up for the victim of bullying? The interplay between personal and social
factors. Merrill-Palmer Quarterly, 56(2), 143–163. http://dx.doi.org/10.1353/mpq.0.0046
Pepler, D. J., & Craig, W. M. (1995). A peek behind the fence: Naturalistic observations of aggressive children with remote audiovisual recording.
Developmental Psychology, 31(4), 548. http://dx.doi.org/10.1037/0012-1649.31.4.548
Polanin, J. R., Espelage, D. L., & Pigott, T. D. (2012). A meta-analysis of school-based bullying prevention programs’ effects on bystander intervention
behavior. School Psychology Review, 41(1), 47–65.
 L.J. Lambe et al. / Child Abuse & Neglect 65 (2017) 112–123 123

Pozzoli, T., & Gini, G. (2010). Active defending and passive bystanding behavior in bullying: The role of personal characteristics and perceived peer
pressure. Journal of Abnormal Child Psychology, 38(6), 815–827. http://dx.doi.org/10.1007/s10802-010-9399-9
Pozzoli, T., Gini, G., & Vieno, A. (2012). The role of individual correlates and class norms in defending and passive bystanding behavior in bullying: A
multilevel analysis. Child Development, 83(6), 1917–1931. http://dx.doi.org/10.1111/j.1467-8624.2012.01831.x
Rapp-Paglicci, L., Dulmus, C. N., Sowers, K. M., & Theriot, M. T. (2004). Hotspots for bullying, Exploring the role of environment in school violence. Journal
of Evidence-Based Social Work: Advances in Practice, Programming, Research, and Policy, 1(2/3), 131–142. http://dx.doi.org/10.1300/J394v01n02 09
Reijntjes, A., Kamphuis, J. H., Prinzie, P., & Telch, M. J. (2010). Peer victimization and internalizing problems in children: A meta-analysis of longitudinal
studies. Child Abuse and Neglect, 34(4), 244–252. http://dx.doi.org/10.1016/j.chiabu.2009.07.009
Rivers, I., & Noret, N. (2013). Potential suicide ideation and its association with observing bullying at school. Journal of Adolescent Health, 53(1), S32–S36.
http://dx.doi.org/10.1016/j.jadohealth.2012.10.279
Rivers, I., Poteat, V. P., Noret, N., & Ashurst, N. (2009). Observing bullying at school: The mental health implications of witness status. School Psychology
Quarterly, 24(4), 211–223. http://dx.doi.org/10.1037/a0018164
Rose, A. J., & Rudolph, K. D. (2006). A review of sex differences in peer relationship processes: Potential trade-offs for the emotional and behavioral
development of girls and boys. Psychological Bulletin, 132(1), 98. http://dx.doi.org/10.1037/0033-2909.132.1.98
Sainio, M., Veenstra, R., Huitsing, G., & Salmivalli, C. (2010). Victims and their defenders: A dyadic approach. International Journal of Behavioral
Development, 1–8 [0165025410378068].
Salmivalli, C., & Voeten, M. (2004). Connections between attitudes, group norms, and behaviour in bullying situations. International Journal of Behavioral
Development, 28(3), 246–258. http://dx.doi.org/10.1080/01650250344000488
Salmivalli, C., Lagerspetz, K., Björkqvist, K., Österman, K., & Kaukiainen, A. (1996). Bullying as a group process: Participant roles and their relations to social
status within the group. Aggressive Behavior, 22(1), 1–15. http://dx.doi.org/10.1002/(SICI)1098-2337(1996)22:1<1:AID-AB1>3.0.CO;2-T
Salmivalli, C., Lappalainen, M., & Lagerspetz, K. M. (1998). Stability and change of behavior in connection with bullying in schools: A two-year follow-up.
Aggressive Behavior, 24(3), 205–218. http://dx.doi.org/10.1002/(SICI)1098-2337(1998)24:3<205:AID-AB5>3.0.CO;2-J
Salmivalli, C., Voeten, M., & Poskiparta, E. (2011). Bystanders matter: Associations between reinforcing, defending, and the frequency of bullying behavior
in classrooms. Journal of Clinical Child and Adolescent Psychology, 40(5), 668–676. http://dx.doi.org/10.1080/15374416.2011.597090
Sijtsema, J. J., Rambaran, J. A., Caravita, S. C. S., & Gini, G. (2014). Friendship selection and influence in bullying and defending: Effects of moral
disengagement. Developmental Psychology, 50(8), 2093–2104. http://dx.doi.org/10.1037/a0037145
Steinberg, L., & Morris, A. S. (2001). Adolescent development. Journal of Cognitive Education and Psychology, 2(1), 55–87.
http://dx.doi.org/10.1891/194589501787383444
Zinzow, H. M., Ruggiero, K. J., Resnick, H., Hanson, R., Smith, D., Saunders, B., et al. (2009). Prevalence and mental health correlates of witnessed parental
and community violence in a national sample of adolescents. Journal of Child Psychology and Psychiatry, 50(4), 441–450.
http://dx.doi.org/10.1111/j.1469-7610.2008.02004.x