Supportive relationships have been hypothesized to protect against psychological distress in general and to reduce the impacts of adversity on distress. Bullying victimization by peers is a salient adverse experience for many adolescents. Being bullied increases the risk of psychological distress which in turn increases the risk of further bullying victimization. There is minimal previous research on whether and how supportive relationships protect adolescents from bullying victimization, psychological distress and the recursive relationships between victimization and distress. This study investigated the direct and moderating effects of supportive relationships with parents, peers, and teachers on later psychological distress and bullying victimization of adolescents. This longitudinal study involved 1425 Australian adolescents aged 12–18 years (mostly male, 74.3%), using questionnaires on bullying victimization, psychological distress, and supportive relationships at two-time points, 6 months apart. All types of supportive relationships reduced the risk of later psychological distress predicted from earlier psychological distress, suggesting a direct compensatory effect. Support from classmates and parents also protected adolescents against ongoing bullying victimization and mitigated the impact of bullying victimization on later psychological distress, consistent with a stress-buffering hypothesis. The pattern of results confirms the important role that supportive relationships play in protecting adolescents from psychological distress and bullying victimization. Persistent bullying victimization and later psychological distress are less likely to occur for victimized adolescents supported by parents and classmates. Strategies and programs that improve parental support may enhance the effectiveness of programs to prevent bullying and improve outcomes for victims.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Bullying is a common form of adversity, experienced by around 15% of adolescents each year (Jadambaa et al., 2019). The serious consequences for victims include internalizing problems (depression and anxiety), that are both consequences of being bullied, and risk factors for further victimization (Christina et al., 2021); this results in worsening victimization as well as mental health problems over time. A strong body of evidence shows that bullying victimization (being bullied) during adolescence precedes the development of internalizing problems (Moore et al., 2017), including after taking into account pre-existing emotional problems, demographic, family, and environmental factors (Jadambaa et al., 2020). Despite these serious consequences, school bullying prevention programs only reduce bullying by around 15% (Fraguas et al., 2021; Gaffney et al., 2019), with effect sizes reducing with age, resulting in negligible impact with older adolescents (Yeager et al., 2015). The current study investigates the protective role of supportive relationships in the associated problems of bullying victimization and psychological distress in adolescence. We use the terms “internalizing problems” and “psychological distress” interchangeably to describe emotional distress or symptoms of anxiety and depression, consistent with previous use of this terminology (Slade et al., 2011).
The idea that supportive relationships might mitigate the harm of bullying victimization is consistent with well-established theory about resilience following adversity. Rutter (1985, 1987) theorized that positive relationships could protect against the negative psychological impacts of adversity by strengthening an individual’s resilience. Accordingly, a meta-analysis found a moderate positive association between social support (in general) and depression in childhood and adolescence (r = 0.26; Rueger et al., 2016). A large cross-national study of adolescents found that supportive peer and parent relationships were associated with reduced depression and anxiety, psychological distress, and suicidal ideation (Biswas et al., 2020). Support from significant others may therefore be an important modifiable factor for interventions for adolescents experiencing co-occurring problems of bullying victimization and psychological distress.
Cohen and Wills (1985) proposed two different ways through which social support could reduce the negative impacts of adversity. The “direct effects model” hypothesizes that positive relationships directly reduce future stress, so compensate for the impact of adversity; so, for instance, it may be that positive interactions may reduce an adolescent’s stress by providing pleasant interactions. In contrast, the “stress-buffering model” hypothesizes that positive relationships interact with the stressor to moderate the detrimental impact of the stressor on future distress; for instance, this might happen if interaction with a supportive person changed the way the adolescent felt about a previous negative incident with peers.
Given the recursive relationship between bullying victimization and psychological distress (Christina et al., 2021), Healy and Sanders (2018) argued that supportive relationships may also mitigate future bullying victimization through either (a) a direct effect on victimization (e.g., if friends discourage further bullying), or (b) buffering the impact of distress on ongoing victimization (e.g., if friends help the victim to be less emotionally reactive). Figure 1 illustrates mechanisms by which supportive relationships may impact victimisation and distress. A small body of literature has explored these pathways. However, little is known about how support from different relationships affect bullying victimization and distress. This study investigates whether, and how, supportive relationships with parents, teachers and peers protect adolescents against bullying victimization and psychological distress. The literature relevant to each of these relationships is summarized below.
Fig. 1
Direct and buffering effects of supportive relationships on bullying victimization and internalizing problems
×
Supportive Parenting
Previous research suggests that supportive parenting may have both a direct effect on psychological distress, as well as buffer the impacts of bullying victimization on psychological distress. In support of direct effects on distress, a longitudinal study found that mother-adolescent relationship quality negatively predicted depression for boys and girls and that the quality of the father-adolescent relationship also predicted depression for boys (Branje et al., 2010). Another study found that relationship quality with both parents predicted adolescents’ depression 1 year later; and that a lack of paternal support predicted increased depression, regardless of adolescents genetic vulnerability (Brouillard et al., 2018). In support of a buffering effect, a meta-analysis found that strong parental support buffered children and adolescents from internalizing problems after bullying victimization (Ttofi et al., 2014). Another longitudinal study found that maternal and warmth and a positive home environment protected children from psychological distress after being bullied (Bowes et al., 2010). Having a supportive family was also found to protect adolescents from increased risk of depression in young adulthood, resulting from school bullying (Isaacs et al., 2008). This all suggests that supportive parenting affects distress both directly and through buffering the adverse impacts of bullying victimization.
Other evidence suggests that supportive parenting may also protect against bullying victimization. A meta-analysis of cross-sectional and prospective longitudinal studies found that supportive parenting predicted reduced bullying victimization and negative parenting predicted increased peer victimization (Lereya et al., 2013). A longitudinal study found that family social support was protective against later cyberbullying victimization (Fanti et al., 2012). What remains unclear is whether supportive parenting affects bullying victimization directly or provides a buffering effect through reducing psychological distress.
Supportive Relationships with Teachers
There have been at least three prospective longitudinal studies, investigating how supportive relationships with teachers affect bullying victimization and psychological distress. All have involved middle school students. One study found that supportive teacher relationships had a main effect on depressive symptoms and moderated ongoing depression for provocative victims, but not passive victims, of bullying (Huang et al., 2018). Another study found that supportive teacher relationships predicted higher self-esteem for students who were bullied, and moderated the impact of bullying victimization on self-esteem (Van Aalst et al., 2021). Another recent study found that supportive teacher relationships did not predict later peer victimization (Ten Bokkel et al., 2021). Although previous research is minimal, this suggests that teacher support may have a direct and buffering effect on psychological distress but no impact on ongoing bullying victimization.
Supportive Peer Relationships
A small body of research examines whether supportive peer relationships protect adolescents from psychological distress and bullying victimization. A structured review, examining whether friendships protect adolescents from adverse health (and mental health) consequences of peer victimization, reported mixed results for both quantity and quality of friendships (Schacter et al., 2021). Although not specific to bullying, one study found that classmate support moderated the relationship between interpersonal stressors (i.e., conflict with parents and peers) and depression (Auerbach et al. 2014). Studies of elementary school children, although also limited, suggest that supportive peer relationships may protect children from both psychological distress and bullying victimization by peers (Denio et al., 2020; Goldbaum et al., 2003) including by buffering children from internalizing problems following bullying victimization (Hodges et al., 1999; Laursen et al., 2007). In summary, although research with adolescents is limited and inconclusive, combined research about children and adolescents suggests that peer support affects both bullying victimization and psychological distress, and may moderate the relationship between bullying victimization and psychological distress over time.
Current Study
This study investigated whether and how supportive relationships with parents, teachers, and peers protect adolescents against bullying victimization and psychological distress over time. The aim was to test whether supportive relationships had direct effects as well as moderating effects on psychological distress and bullying victimization, and the relationship between psychological distress and bullying victimization over time. We hypothesized that:
1.
Supportive relationships would reduce adolescents’ trajectory of psychological distress, predicted from previous psychological distress.
2.
Supportive relationships would reduce adolescents’ trajectory of ongoing bullying victimization, predicted from previous bullying victimization.
3.
Supportive relationships would reduce the impact of bullying victimization on later psychological distress.
4.
Supportive relationships would reduce the impact of psychological distress on later bullying victimization.
Method
Participants
Eighty secondary schools from south-east Queensland Australia were contacted and invited to take part in the study. Of these, 10 schools agreed to participate. All students in Years 7–12 were invited to participate. Data were collected between 2014 and 2016 from a non-random and non-representative convenience sample of 1425 adolescents, as previously described by Thomas et al. (2019), in a validation study of a measure of bullying. Adolescents from private schools were over-represented in the sample (seven schools). Remaining schools were state government schools. Approximately 35% of students attend private schools in the state of Queensland and nationally in Australia (ABS, 2017). When provided, the most common reasons for declining the invitation to take part in the study were having too little time in the school schedule or that the school was already participating in another research project. Six of the seven private schools had a religious affiliation and, of these, four were Catholic. Four were same-gender schools (two boys only and two girls only). Participation in the study was open to adolescents in Years 7–12 (secondary school). The age/year group selected to participate was decided by the school and most schools chose to involve lower year levels only. There was a strong participation rate in one boys-only school. Participating schools sent 3794 invitations to their students’ households. 36.6% of parents responded and provided active consent, of which 87.6% of adolescents also provided consent (response rate = 32%). Our response rate is in line with previous Australian research on adolescent bullying whose ethical clearance protocol also required active parental consent (Shaw et al., 2015). Participating adolescents were mostly male (74.3%), aged 12–18 years from school Years 7–12 with a mean age of 13.99 (SD = 1.28). Most were born in Australia (85.0%) and spoke English at home (90.2%). A majority (74.7%) of adolescents lived with their married biological parents; 66.5% lived in urban centers, and 2.1% identified as Indigenous Australians. The sample had good socio-economic diversity with 15.2% being “most advantaged”, 31.3% “advantaged”, 15.8% “neither advantaged nor disadvantaged”, 29.4% were “disadvantaged”, and 8.3% from the “most disadvantaged” in Australia (Thomas et al., 2019).
Measures
Data were collected through two assessments 6 months apart, using the measures reported below.
Bullying victimization by peers
Bullying victimization was measured through the Total Victimization Scale of the Bullying and Cyberbullying Scale for Adolescents which presents items following a definition of bullying and cyberbullying consistent with research consensus (Thomas et al., 2019). The Total Victimization scale includes 13 items about experiences of victimization including physical (e.g., punched, hit, kicked, pushed or shoved me, on purpose), verbal (e.g., called me mean or hurtful names), relational (e.g., spread lies or rumors about me, to hurt me or make others not like me) and cyber (e.g., sent or posted, mean or hurtful pictures/videos about me). Respondents rated how often they experienced each item in the last three months, on a scale from 0 (never) to 4 (four or more times). This scale demonstrated very good internal consistency at both time points (α = .90, α = .90). Total bullying victimization scores were calculated by computing the mean.
Adolescent psychological distress
Psychological distress was measured by the 10-item Kessler Psychological Distress Scale (Kessler et al., 2003). Respondents rate how often they have experienced each symptom of distress (e.g., Tired for no reason) in the last 4 weeks, on a Likert scale ranging from 1 (none of the time) to 5 (all of the time). Internal consistency was very good at both timepoints (α = .90, α = .90). The mean score provides an overall measure of psychological distress.
Supportive relationships
Adolescents reported supportive relationships with parents, peers, and teachers through the Child and Adolescent Social Support Scale (Malecki & Kilpatrick, 2002) This 40-item scale measures perceived social support from four sources: parents, teachers, classmates, and friends, each of which is measured by a subscale of 10 items. Respondents rate each item on a 6-point scale ranging from 1 (never) to 6 (always). Example items include: “My parent(s)… help me make decisions”, “My teacher(s)…praises me when I’ve tried”, “My classmates… say nice things to me” and “My close friend(s)… make me feel better”. Internal consistency was very good at both timepoints respectively for social support from parents (α = .94, α = .92), teachers (α = .95, α = .94), classmates (α = .95, α = .96), and close friends (α = .96, α = .95). Mean subscale scores were calculated. Total social support was the mean score across all items.
Procedure
Ethical clearance was obtained from The University of Queensland (UQ 201400759), as well as approval from relevant education authorities and school principals. Parents of students from participating schools were invited to provide consent for their adolescent child to complete a survey on bullying and mental wellbeing. A researcher (Author 3) visited the schools to deliver the questionnaires to students. Students with parental consent met with the researcher in a classroom setting and were offered an information sheet and informed consent sheet. Consenting participants completed the pencil-paper survey during a 40-min period in school time. Data were later entered manually into a Statistical Packages for Social Sciences file for analysis. The researcher re-visited the schools 6 months later and administered the assessments again to the students involved. For further information about the procedure and sample, please refer to Thomas et al. (2019).
Statistical Analyses
We conducted hierarchical multiple regressions (HMR) from the Statistical Package for Social Sciences (SPSS) to test main effects and moderating effects of supportive relationship measures at 0 months on the outcome variables at 6 months (Time 2) after controlling for the corresponding outcome variables at 0 months (Time 1), as described by Tabachnick et al. (2013). We first centered each variable before calculating product terms to avoid problems of multicollinearity associated with utilizing product terms in regression (Aiken et al., 1991). To test the main effect of supportive relationships for Time 2 bullying victimization as the outcome variable, Time 1 bullying victimization was entered first, followed by Time 1 supportive relationships. To test the moderating role of supportive relationships, we entered variables in the following order: Time 1 psychological distress at Step 1, Time 1 supportive relationships and Time 1 bullying victimization at Step 2, and the product term of bullying victimization and supportive relationships at Time 1 at Step 3. As well as analyses for total support, we replicated these analyses for scale scores of different types of supportive relationships (parents, teachers, classmates, and friends). Finally, we replicated the above analyses with Time 2 psychological distress as the outcome variable, by entering Time 1 psychological distress at Step 1, Time 1 bullying victimization and supportive relationships at Step 2, and the product term at Step 3. Again, this was repeated for different types of relationship support. All results reported are for one-tailed analyses.
Results
Tests of Basic Assumptions
We checked whether our data met basic assumptions for linear regression. The data-set met assumptions of linearity and multivariate normality (i.e., there was no homoscedasticity). The risk of multi-collinearity was successfully managed by centering variables prior to including them in interaction terms. Some individual variables had non-normal distributions. Most notably, the measures of bullying victimisation were highly skewed and kurtosed. Non-normality is very common in social sciences with most samples investigated showing non-normal distributions (Yuan & Gomer, 2021). Furthermore, large samples tend to produce high scores for skewness and kurtosis in response to small deviations in normality (Field, 2009). Although non-normality can produce errors in analyses with small sample sizes, regression analyses have been found to be robust to non-normality in samples of >100 (McGuinness et al., 1997) and robust to extreme non-normality for sample sizes of >500 (Lumley et al., 2002). Therefore, our large sample of 1425 should protect against any deviations to normality in distributions of individual variables.
Of more concern, a missing data analysis revealed that 18.43% of values were missing, partially driven by a large proportion of students who had completed either the initial or second survey but not both (n = 48; 8%). Little’s test revealed data were not missing completely at random (χ2 [38,505] = 48,266.34, p < .001). This is not unusual for psychological data (Enders, 2017). Pattern analysis revealed a pattern of monotonicity associated with dropout. Through ANOVAs, using a dummy variable for missingness, we ascertained that missingness was associated with the measured variables of student age at Time 1 (F [1, 1203] = 93.14, p < .001), and Time 2 (F [1, 1423] = 1707.56, p < .001) with older students more likely to have missed a whole assessment. Typically, age was missing because the whole assessment at that time-point was missing; consistent with this, the supervising researcher reported that older students were sometimes unavailable for the second assessment. Missingness was also related to psychological distress (F [1, 1184] = 8.20, p = .004), social support from parents at (F [1, 1156] = 6.71, p = .01), social support from teachers (F [1, 1423] = 1707.56, p < .001) and total social support (F [1, 1157] = 4.14, p = .042) at Time 1. Together, this pattern of analyses suggests that missingness was related to measured variables i.e., Missing at Random.
Multiple imputation is the most appropriate and sophisticated way to manage data that is missing at random including that caused by systematic dropout (Van Ginkel et al., 2020). This involves the creation of several alternative plausible data sets, each based on different estimates of missing values (Enders, 2017); analyses of data that includes a range of options for missing values reduces the risk of bias associated with either listwise/pairwise deletion or simple imputation (Van Ginkel et al., 2020). We conducted imputations prior to regression analyses through SPSS. Where pooled results were available through SPSS, these have been reported. Otherwise, to best represent the diversity of results across all imputed datasets, we have reported both the median and the range of each relevant statistic across the five sets of analyses. Multiple imputations are appropriate with non-normal data (Dong & Peng, 2013). Estimates of achieved power through G*power showed that, for all analyses, power approached 1.00 (Faul et al., 2009).
Preliminary Analyses
Table 1 shows means and standard deviations of original data (without imputations) and pooled correlation coefficients calculated through SPSS across the five imputed data sets. As expected, there were strong correlations between most variables. In particular, there were consistently strong associations between all variables at Time 1. All types of social support were strongly correlated with each other (p < .001) and had strong negative correlations with psychological distress (p < .001) and bullying victimization at Time 1 (p < .001). All Time 1 variables were highly correlated with psychological distress at Time 2 (p < .001). However, Time 2 bullying victimization was not associated with Time 1 variables apart from a significant positive association with bullying victimization (p < .001) and a significant negative association with support from classmates (p < .05). The association between bullying victimization and psychological distress at Time 1 was r = .40 whereas the correlation between these variables at Time 2 was only r = .07. Bullying victimization significantly reduced between Time 1 and Time 2, (t [45] = 4.75, p = < .001) whereas psychological distress remained stable over time, (t [44] = −0.02, p = .982).
Table 1
Means, standard deviations and inter-correlations between all variables
Variables
Mean(SD)
1.
2.
3
4.
5.
6.
7.
8.
9.
1. Total support T1
4.70 (0.86)
–
2. Parental support T1
4.74 (1.08)
.80***
–
3. Friend support T1
5.04 (0.99)
.79***
.50***
–
4. Classmate support T1
4.49 (1.12)
.81***
.46***
–
5. Teacher support T1
4.49 (1.11)
.79 ***
.56***
.44***
.48***
–
6. Psychological distress T1
2.01 (0.80)
−.36***
−.29***
−.17***
−.35***
−.30***
–
7. Bullying victimization T1
0.40 (0.59)
−.24***
−.14***
−.13***
−.29***
−.21***
.40***
–
8. Psychological distress T2
2.01 (0.81)
−.26***
−.22***
−.14***
−.28***
−.25***
.45***
.33***
–
9. Bullying victimization T2
0.29 (0.49)
−.05
−.03
−.02
−.08*
−.01
.06
.17***
.07*
–
All means and correlations are pooled
T1 Time 1 (0 months), T2 Time 2 (6 months)
*p < .05 **p < .01 ***p < .001
Effects of Supportive Relationships on Psychological distress
Table 2 displays HMRs testing the effect of supportive relationships on later psychological distress. After controlling for psychological distress at Time 1, all types of social support at Time 1 (i.e., from parents, teacher, classmates and friends and total support) predicted lower levels of psychological distress at Time 2 (p < .001 for F-change). Inclusion of the interaction term at Step 3 made a significant improvement to the model for classmate support only (with parental support trending towards significance). Figure 2 (in the Supplementary Materials) shows the interaction after splitting participants into three equally sized groups for low, medium and high levels of classmate support. The relationship between Time 1 and Time 2 psychological distress was weakest for participants who reported high levels of classmate support.
Table 2
Hierarchical multiple regression investigating whether support moderates the relationship between psychological distress at Times 1 and 2
Step
Predictor variables
B (pooled) [95% CI]
SEB (pooled)
β median [range]
R2 adj median [range]
∆ R2 median [range]
F Change median [range]
Sign F-change p median [range]
Effects of total support
1.
Psychological distress (T1)
0.47***
[0.40; 0.53]
.03
.45
[.43; .47]
.20
[.18; .22]
.20
[.18; .22]
355.14
[321.82; 408.70]
<.001***
[<.001; <.001]
2.
Total support (T1)
−0.14*
[−0.24; −0.04]
.04
−.15
[−.11; −.20]
.22
[.20; .25]
.02
[.01; .04]
41.20
[21.04; 67.33]
<.001***
[<.001; <.001]
3.
Interaction term (T1 Psychological distress X Total support)
−0.04
[−0.11; 0.04]
.03
−.04
[−.01; −.08]
.23
[.20; .25]
.00
[.00; .01]
2.53
[0.10; 9.07]
.112
[.796; .003]
Effects of parental support
1.
Psychological distress (T1)
0.47***
[0.40; 0.53]
.03
.45
[.43; .47]
.20
[.19; .22]
.20
[.18; 0.22]
355.14
[321.82; 408.70]
<.001***
[<.001; <.001]
2.
Parent support (T1)
−0.11*
[−0.16; −0.06]
.02
−.14
[−.12; −.16]
.22
[.20; .25]
.02
[.01; .02]
32.46
[24.25; 45.85]
<.001***
[<.001; <.001]
3.
Interaction term (T1 Psychological distress X Parent support)
−0.03
[−0.08; 0.02]
.02
−.04
[−.02; −.07]
.22
[.20; .25]
.00
[.00; .01]
3.02
[0.54; 9.00]
.083
[.463; .003]
Effects of teacher support
1.
Psychological distress (T1)
0.47***
[0.40; 0.53]
.03
.45
[.43; .47]
.20
[.19; .22]
.20
[.18; 0.22]
355.14
[321.82; 408.70]
<.001***
[<.001; <.001]
2.
Teacher support (T1)
−0.09*
[−0.17; −0.01]
.03
−.13
[−.08; −.15]
.22
[.19; 0.24]
.00
[.00; .01]
28.19
[10.10; 38.92]
<.001***
[.002; <.001]
3.
Interaction term (T1 Psychological distress X Teacher support)
−0.02
[−0.08; 0.04]
.03
−.02
[−.01; −.07]
.22
[.19; 0.24]
.00
[.00; .01]
0.64
[0.08; 9.48]
.425
[.784; .002]
Effects of classmate support
1.
Psychological distress (T1)
0.47***
[0.40; 0.53]
.03
.45
[.43; .47]
.20
[.19; .22]
.20
[.18; .22]
355.14
[321.82; 408.70]
<.001***
[<.001; <.001]
2.
Class support (T1)
−0.09*
[−0.18; −0.01]
.03
−.14
[−.08; −.16]
.22
[.19; .25]
.02
[.01; .02]
34.55
[9.47; 42.04]
<.001***
[.002; <.001]
3.
Interaction term (T1 Psychological distress X Class support)
−0.05
[−0.11; 0.02]
.03
−.06
[−.05; −.12]
.22
[.20; .25]
.00
[.00; .01]
6.58
[4.34; 22.28]
.010*
[.038; <.001]
Effects of friend support
1.
Psychological distress (T1)
0.47***
[0.40; 0.53]
.03
.45
[.43; .47]
.20
[.19; .22]
.20
[.18; .22]
355.14
[321.82; 408.70]
<.001***
[<.001; <.001]
2.
Friend support (T1)
−0.07
[−0.15; 0.01]
.03
−.08
[−.03; −.10]
.21
[.19; .23]
.01
[.00; .01]
12.90
[1.21; 18.35]
<.001***
[.289; <.001]
3.
Interaction term (T1 Psychological distress X Friend support)
−0.02
[−0.10; 0.06]
.03
−.02
[−.09; .02]
.21
[.19; .24]
.00
[.00; .01]
0.61
[0.44; 12.97]
.436
[.505; <.001]
Significance levels for B are from the final model with all variables included
T1 Time 1; T2 Time 2
*p < .05; **p < .01; ***p < .001
Fig. 2
Interaction of class support with ongoing psychological distress
×
Effects of Supportive Relationships on Bullying Victimization
Table 3 (Supplementary Materials) shows HMRs testing the effect of supportive relationships on later bullying victimization. After controlling for bullying victimization at Time 1, no sources of social support (i.e., from parents, teachers, classmates, and friends or total support) reliably predicted bullying victimization at Time 2 across the five sets of imputed data. However, the interaction of both classroom support and parental support with bullying victimization at Time 1 significantly improved prediction of Time 2 bullying victimization. As shown in Fig. 3, the relationship between Time 1 and Time 2 bullying victimization was strongest for participants who reported the lowest levels of classmate support; the same was true for parental support.
Table 3
Hierarchical multiple regression investigating whether support buffers ongoing bullying victimization
Step
Predictor variables
B (pooled) [95% CI]
SEB (pooled)
β median [range]
R2 adj median [range]
∆ R2 median [range]
F Change median [range]
Sign F-change p median [range]
Effects of total support
1.
Bullying victimization (T1)
0.08**
[0.02; 0.13]
.03
.10
[.07; .11]
.01
[.01; .01]
.01
[.01; .01]
11.11
[7.71; 17.49]
.001***
[.006; <.001]
2.
Total support (T1)
−0.02
[−0.07; 0.03]
.02
−.03
[−.00; −.07]
.01
[.01; .02]
.00
[.00; .01]
1.48
[0.00; 6.71]
.224
[.947; .01]
3.
Interaction term (T1 Bullying victimization X Total support)
0.00
[−0.08; 0.08]
.04
−.01
[.06; −.02]
.01
[.01; .02]
.00
[.00; .00]
0.27
[0.03; 5.32]
.606
[.857; .021]
Effects of parental support
1.
Bullying victimization (T1)
0.08**
[0.03; 0.15]
.03
.09
[.07; .11]
.01
[.01; .01]
.01
[.01; .01]
11.11
[7.71; 17.49]
.001***
[.006; <.001]
2.
Parent support (T1)
−0.00
[−0.04; 0.02]
.02
−.01
[.00; −.03]
.01
[.01; .01]
.00
[.00; .01]
0.14
[0.00; 1.50]
.706
[.990; .221]
3.
Interaction term (T1 Bullying victimization X Parent support)
0.04
[−0.03; 0.11]
.03
.06
[.04; .13]
.01
[.01; .02]
.00
[.00; .02]
4.30
[2.17; 24.57]
.038*
[.141; < .001]
Effects of teacher support
1.
Bullying victimization (T1)
0.08*
[0.02; 0.13]
.03
.09
[.07; .11]
.01
[.01; .01]
.01
[.01; .01]
11.11
[7.71; 17.49]
.001***
[.006; <.001]
2.
Teacher support (T1)
−0.01
[−0.05; 0.04]
.02
−.01
[.03; −.06]
.01
[.01; .01]
.00
[.00; .00]
1.05
[0.03; 4.40]
.305
[.869; .036]
3.
Interaction term (T1 Bullying victimization X Teacher support)
0.01
[−0.02; 0.05]
.02
.02
[.01; .05]
.01
[.01; .01]
.00
[.00; .00]
0.55
[0.05; 4.00]
.457
[.827; .046]
Effects of classmate support
1.
Bullying victimization (T1)
0.08
[−0.00; 0.11]
.03
.09
[.07; .11]
.01
[.01; .01]
.01
[.01; .01]
11.11
[7.71; 17.49]
.001***
[.006; <.001]
2.
Classmate support (T1)
−0.02
[−0.07, 0.02]
.02
−.05
[−.02; −.10]
.01
[.01; .02]
.00
[.00; .01]
2.81
[0.52; 14.18]
.094
[.470; <.001]
3.
Interaction term (T1 Bullying victimization X Class support)
−0.03*
[−0.07; −0.00]
.02
−.07
[−.05; −.09]
.01
[.01; .02]
.00
[.00; .01]
5.55
[3.79; 9.77]
.019*
[.052; .002]
Effects of friend support
1.
Bullying victimization (T1)
0.08**
[0.02; 0.13]
.03
.09
[.07; .11]
.01
[.01; .01]
.01
[.01; .01]
11.11
[7.71; 17.49]
.001***
[.006; <.001]
2.
Friend support (T1)
−0.01
[−0.05; 0.02]
.02
−.02
[.00; −.05]
.01
[.01; .01]
.00
[.00; .00]
0.84
[0.09; 3.67]
.359
[.925; .056]
3.
Interaction term (T1 Bullying victimization X Friend support)
0.00
[−0.02; 0.03]
.01
.00
[−.00; .03]
.01
[.00; .01]
.00
[.00; .00]
0.020
[.003; 1.51]
.887
[.956; .219]
Significance levels for B are for the final model with all variables included
T1 Time 1, T2 Time 2
*p < .05; **p < .01; ***p < .001
Fig. 3
Interaction of class support with ongoing bullying victimization
×
Supportive Relationships as Moderators of the Impact of Bullying Victimization on Psychological Distress
Table 4 shows HMRs testing whether supportive relationships moderated the impact of bullying victimization on later distress. The significant negative regression coefficients at Step 2, show that, after controlling for psychological distress at Step 1 and taking into account the impact of bullying victimization, relationship support from parents as well as total support predicted lower levels of later psychological distress; in addition, support from classmates was marginally significant (p = .053). For parental support (p < .001), total support (p < .001), and support from classmates (p = .038) only, inclusion of the interaction term at Step 3 further improved the model, consistent with a stress-buffering effect. For regressions testing support from teachers or friends, neither the direct effect at Step 2 nor the stress-buffering effect of the interaction term at Step 3, improved prediction of later psychological distress.
Table 4
Hierarchical multiple regressions investigating whether support buffers the impact of bullying victimization at Time 1 on psychological distress at Time 2
Step
Predictor variables
B (pooled) [95% CI]
SEB (pooled)
β median [range]
R2 adj median [range]
∆ R2 median [range]
F Change median [range]
Sign F-change p median [range]
Effects of total support
1.
Psychological distress (T1)
0.47***
[0.40; 0.53]
.07
.45
[.43; .47]
.20
[.19; .22]
.20
[.18; .22]
355.14
[321.82; 408.70]
<.001***
[<.001; <.001]
2.
Bullying victimization (T1)
0.16*
[0.03; 0.29]
.06
.10
[.08; .16]
.23
[.22; .26]
.03
[.02; .05]
42.51
[18.90; 42.51]
<.001***
[<.001; <.001]
Total support (T1)
−0.13*
[−0.23; −0.03]
.04
−.15
[−.10; −.19]
3.
Interaction term (bullying victimization X Total support)
−0.12
[−0.29; 0.06]
.07
−.12
[−.04; −.13]
.24
[.22; .27]
.01
[.00; .05]
26.59
[2.18; 30.14]
<.001***
[.140; <.001]
Effects of parental support
1.
Psychological distress (T1)
0.47***
[0.40; 0.53]
.07
.45
[.43; .47]
.20
[.19; .22]
.20
[.18; .22]
355.14
[321.82; 408.70]
<.001***
[<.001; <.001]
2.
Bullying victimization (T1)
0.17**
[0.06; 0.29]
.05
.12
[.08; .16]
.23
[.22; .26]
.03
[.03; .04]
27.09
[22.74; 37.19]
<.001***
[<.001; <.001]
Parent support (T1)
−0.11***
[−0.15; −0.06]
.02
−.13
[−.12; −.16]
3.
Interaction term (bullying victimization X Parent support)
−0.07
[−0.17; 0.04]
.04
−.10
[−.08; −.16]
.23
[.22; .27]
.01
[.00; .01]
12.26
[0.53; 20.74]
<.001***
[.468; <.001]
Effects of teacher support
1.
Psychological distress (T1)
0.47***
[0.40; 0.53]
.07
.45
[.43; .47]
.20
[0.19; .22]
.20
[.18; .22]
355.14
[321.82; 408.70]
<.001***
[<.001; <.001]
2.
Bullying victimization (T1)
0.16*
[0.04; 0.28]
.03
.11
[.09; .16]
.22
[.21; .23]
.03
[.02; .03]
24.71
[14.70; 29.42]
<.001***
[<.001; <.001]
Teacher support (T1)
−0.08
[−0.16; 0.00]
.04
−.12
[−.06; −.14]
3.
Interaction term (bullying victimization X Teacher support)
−0.04
[−0.20; 0.12]
.06
−.02
[−.11; .01]
.23
[.21; .26]
.00
[.00; .01]
0.95
[0.06; 21.85]
.329
[.957; <.001]
Effects of classmate support
1.
Psychological distress (T1)
0.47***
[0.40; 0.53]
.07
.45
[.43; .47]
.20
[.19; .22]
.20
[.18; .22]
355.14
[321.82; 408.70]
<.001***
[<.001; <.001]
2.
Bullying victimization (T1)
0.16*
[0.03; 0.28]
.06
.10
[.08; .16]
.23
[.21; .26]
.03
[.00; .03]
26.15
[13.01; 31.19]
<0.001***
[<.001; <.001]
Classmate support (T1)
−0.08
[−0.17; 0.00]
.04
−.13
[−.06; −.14]
3.
Interaction term (bullying victimization X Classmate support)
−0.04
[−0.13; 0.06]
.04
−.05
[−.00; −.09]
.23
[.21; .26]
.00
[.00; .01]
4.33
[0.00; 12.93]
.038*
[.958; .001]
Effects of friend support
1.
Psychological distress (T1)
0.47***
[0.40; 0.53]
.07
.45
[.43; .47]
.20
[.19; .22]
.20
[.18; .22]
355.14
[321.82; 408.70]
<.001***
[<.001; <.001]
2.
Bullying victimization (T1)
0.17**
[0.05; 0.29]
.05
.11
[.10; 0.16]
.21
[.21; .24]
.02
[.01; .03]
19.94 [10.40; 24.96]
<.001***
[<.001; <.001]
Friend support (T1)
−0.06
[−0.14; 0.02]
.04
−.08
[−.02; −.10]
3.
Interaction term (bullying victimization X Friend support)
−0.00 [−0.06; 0.06]
.03
.00
[.03; −.06]
.21
[.21; .25]
.00
[.00; .00]
0.274
[0.001; 7.12]
.601
[.970; .008]
Significance levels for B are for the final model, with all variables included
T1 Time 1, T2 Time 2
*p < .05; **p < .01; ***p < .001
We graphed the interactions for total support, parenting support and classmate support. Figure 4 shows that bullying victimization more strongly predicted psychological distress at low levels of parenting support. That is, medium or high levels of parenting support buffered the impact of bullying victimization on later psychological distress. The graphs for total support and classmate support were visually similar.
Fig. 4
Effect of bullying victimisation at time 1 on internalising problems at time 2 depending on level of parenting support
×
Supportive Relationships as Moderators of the Impact of Psychological Distress on Bullying Victimization
We conducted HMRs to test whether supportive relationships moderated the effect of psychological distress on later bullying victimization. After controlling for bullying victimization at Time 1, neither the inclusion of psychological distress, nor any types of support at Step 2, nor the interaction terms at Step 3 improved the model. These HMRs are included in Table 5 (Supplementary Materials 5).
Table 5
Hierarchical multiple regression investigating if support buffers impact of psychological distress at Time 1 on bullying victimization at Time 2
Step
Predictor variables
B (pooled) [95% CI]
SEB (pooled)
β median [range]
R2 adj median [range]
∆ R2 median [range]
F Change median [range]
Sign F-change p median [range]
Effects of total support
1.
Bullying victimization (T1)
0.08**
[0.03; 0.13]
.03
.09
[.08; .10]
.01
[.01; .01]
.01
[.01; .01]
11.11
[7.71; 17.49]
.001*
[.006; <.001]
2.
Psychological distress (T1)
−0.01
[−0.05; 0.05]
.03
−.01
[−.05; .02]
.01
[.01; .02]
.00
[.00; .01]
2.15
[0.03; 3.36]
.117
[.970; .035]
Total support (T1)
−0.03
[−0.07; 0.02]
.02
−.02
[−.07; −.02]
3.
Interaction term (Psychological distress X Total support)
0.01
[−0.05; 0.07]
.03
.01
[−.01; .08]
.01
[.01; .01]
.00
[.00; .01]
0.15
[0.02; 7.43]
.698
[.891; .007]
Effects of parent support
1.
Bullying victimization (T1)
0.08**
[.03; .13]
.03
.09
[.08; .11]
.01
[.01; .01]
.01
[.01; .01]
11.11
[7.71; 17.49]
.001*
[.006; <.001]
2.
Psychological distress (T1)
0.01
[−0.06; 0.06]
.03
.02
[−.04; .03]
.01
[.01; .01]
.00
[.00; .00]
0.75
[0.01; 1.06]
.474
[.993; .348]
Parent support (T1)
−0.01
[−0.04; 0.02]
.02
−.01
[−.04; −.00]
3.
Interaction term (Psychological distress X Parent support)
0.02
[−0.03; 0.06]
.02
.03
[−.00; .08]
.01
[.01; .01]
.00
[.00; .01]
1.51
[0.001; 7.93]
.220
[.945; .005]
Effects of teacher support
1.
Bullying victimization (T1)
0.08**
[0.03; 0.13]
.03
.09
[.09; .10]
.01
[.01; .01]
.01
[.01; .01]
11.11
[7.71; 17.49]
.001*
[.006; <.001]
2.
Psychological distress (T1)
−0.01
[−0.05; 0.04]
.02
−.00
[−.04; .01]
.01
[.01; .01]
.00
[.00; .01]
0.68
[0.05; 2.21]
.509
[.947; .110]
Teacher support (T1)
−0.01
[−0.05; 0.04]
.02
−.01
[−.06; .03]
3.
Interaction term (Psychological distress X Teacher support)
0.00
[−0.02; 0.03]
.01
.00
[−.01; .02]
.01
[.01; .01]
.00
[.00; .00]
0.23
[0.01; 4.50]
.636
[.916; .034]
Effects of class support
1.
Bullying victimization (T1)
0.08**
[0.02; 0.12]
.03
.08
[.07; .10]
.01
[.01; .01]
.01
[.01; .01]
11.11
[7.71; 17.49]
.001*
[.006; <.001]
2.
Psychological distress (T1)
−0.01
[−0.07; 0.04]
.03
−.02
[−.06; .01]
.01
[.01; .02]
.00
[.00; .01]
1.45
[0.63; 7.14]
.236
[.534; .001]
Class support (T1)
−0.03
[−0.07; 0.02]
.02
−.04
[−.10; −.02]
3.
Interaction term (Psychological distress X Class support)
−0.01
[−0.04; 0.03]
.02
−.02
[−.06; .03]
.01
[.00; .02]
.00
[.00; .00]
1.17
[0.06; 4.46]
.279
[.806; .035]
Effects of friend support
1.
Bullying victimization (T1)
0.08**
[0.03; 0.13]
.03
.09
[.08; .10]
.01
[.01; .01]
.01
[.01; .01]
11.11
[7.71; 17.49]
.001*
[.006; <.001]
2.
Psychological distress (T1)
−0.01
[−0.05; 0.04]
.02
.00
[−.04; .01]
.01
[.00; .01]
.00
[.00; .00]
0.52
[0.22; 1.83]
.594
[.803; .160]
Friend support (T1)
−0.01
[−0.05; 0.02]
.02
−.03
[−.05; −.01]
3.
Interaction term (Psychological distress X Friend support)
0.01
[−0.04; 0.05]
.02
.02
[−.03; .05]
.01
[.01; .01]
.00
[.00; .00]
1.28
[0.34; 3.11]
.259
[.559; .078]
Values for B, β and SEB are all for the final model with all variables included
T1 Time 1, T2 Time 2
*p < .05; **p < .01; ***p < .001
Discussion
This study investigated whether and how adolescents’ supportive relationships with significant others impact the inter-related problems of bullying victimization and psychological distress over time. It has been theorized that supportive relationships protect against psychological distress, ongoing bullying victimization, and the recursive relationship between these (Rutter, 1985, 1987; Healy & Sanders, 2018). We investigated supportive relationships with parents, classmates, teachers, and friends as well as total support. Previous research across these different types of relationships has been limited. We hypothesized that supportive relationships would mitigate the ongoing course of distress, the ongoing course of bullying victimization, and the recursive relationship between bullying victimization and distress. These hypotheses were supported differentially for different kinds of relationships; furthermore, there were difference in how different relationships impacted outcomes (i.e., whether they directly impacted outcomes or had a buffering effect).
Consistent with our first hypothesis, all types of relationships had a direct effect in reducing the risk of later psychological distress, predicted from earlier psychological distress. This suggests that supportive relationships have a buoyant effect on distress in general. In addition, positive relationships with classmates significantly interacted with earlier distress to reduce later distress, suggesting that supportive classmates may influence the internal processing of distress.
Our second hypothesis was that supportive relationships would reduce adolescents’ trajectory of bullying victimization. This hypothesis was supported only for relationships with parents and classmates. Higher levels of support from parents and classmates interacted with previous bullying victimization to reduce later bullying victimization. This suggests that supportive relationships with parents and classmates disrupted the trajectory of bullying victimization, perhaps by impacting the social interaction between the victim and the perpetrator/s of bullying. Possible reasons for this are discussed below. No type of supportive relationship had a direct effect on bullying victimization at Time 2.
Hypotheses 3 predicted that supportive relationships would reduce distress predicted from earlier bullying victimization. Support from parents (and total support) only had a significant direct effect on later distress. Supportive relationships with both parents and classmates also interacted with bullying victimization to reduce later distress, suggesting these relationships may have helped victims reframe bullying incidents so that they were less upsetting.
Our fourth hypothesis, that supportive relationships would reduce bullying victimization predicted from distress, was not supported for any relationship. Possible reasons for this are discussed below.
Together, these results confirm that supportive relationships protect adolescents from psychological distress and bullying victimization. In particular, supportive relationships with parents and classmates impacted distress and bullying victimization in multiple ways. The differential impacts of different types of relationship are discussed below, with reference to previous research.
Differential Impacts of Different Types of Supportive Relationships
Support of parents
In this study, supportive parenting influenced the outcomes of bullying victimization and psychological distress by several distinct ways. Consistent with both the direct effects and stress-buffering models, supportive parent relationships both directly reduced later psychological distress and buffered adolescents from the impact of bullying victimization on later distress. This is consistent with previous research suggesting that supportive parenting has both a direct effect (e.g., Brouillard et al., 2018) and a stress-buffering effect (Isaacs et al., 2008).
Supportive parenting also buffered adolescents from the risk of later bullying victimization predicted from earlier bullying victimization. That is, higher levels of parental support effectively disrupted ongoing bullying victimization. This is consistent with previous studies that have focused mainly on children. A meta-analysis of cross-sectional and longitudinal studies found that supportive parenting protected children and adolescents against bullying victimization (Lereya et al., 2013). Supportive parenting also mitigated the course of ongoing overt victimization of children who participated in a randomized controlled trial (RCT) of a family intervention (Healy & Sanders, 2014). Another study found that children who experienced lower quality parent-child relationships were more likely to experience persistent victimization following a whole-school bullying intervention (Kaufman et al., 2018). Together, this suggests that supportive actions of parents may be effective in reducing bullying victimization. Examples of strategies taught to parents in the above mentioned RCT (Healy & Sanders, 2014) included coaching children how to respond to negative peer behavior (so as not to encourage bullying) and communicating with the school to address any ongoing concerns.
Support of classmates
Like supportive parenting, support from classmates also influenced the outcomes of bullying victimization and psychological distress through multiple distinct ways. Consistent with both the direct effect and stress-buffering models, classmate support both directly reduced later psychological distress and mitigated the contribution of bullying victimization to later distress levels. This is consistent with previous research focussing on general interpersonal stressors rather than bullying (Auerbach et al., 2014).
Support from classmates was the only type of relationship support to interact with previous psychological distress to predict future psychological distress, suggesting classmates may have influenced internal mechanisms that perpetuate distress. Previous research shows that the appraisal of adverse events impacts later distress (Huh et al., 2017). Classmates are perhaps more likely than others to witness events that cause psychological distress (like bullying), so may be in the best position to impact how adolescents appraise such events, which could impact their ongoing distress. Supportive classmates may also foster a sense of belonging that affects how distressing events are appraised and recalled.
Like support from parents, support from classmates interacted with previous bullying victimization to impact later bullying victimization. Previous research informing this question has been very limited. A previous study with primary school children found that low friendship quality, in general, protected against ongoing bullying victimization (Goldbaum et al., 2003). However, neither this study, nor previous research in general, has distinguished between classmates and friends.
Support of close friends
Unlike relationships with classmates, relationships with close friends did not protect against future adverse outcomes in multiple ways. Like all relationships, supportive friendships had a direct effect on later psychological distress. However, unlike relationships with classmates, close friendships did not protect against ongoing bullying victimization, nor buffer adolescents from distress following victimization.
As noted above, previous research has not distinguished between support from friends and classmates. A structured review reported mixed results for whether the quantity and quality of friendships buffers adolescents from distress following bullying victimization and suggested these differences may be due to different methodologies (Schacter et al., 2021). However, the current study used the same methodology for evaluating the impact of close friendships and classmates, with quite different results. This suggests these differences may reflect more than measurement methodology.
Perhaps, when responding to questions about friends, adolescents considered peers outside their class or school. Unlike classmates, outside friends would be less likely to be present in environments in which distressing events, like bullying, occurred; thus, they may have had less influence on the behavior of either the victim or the student/s perpetrating bullying. Outside friends may also have been less influential in how the victim interprets events like bullying, as they may not even be aware of these events to enter into discussions. This interpretation is consistent with a study that found reduced parent and classmate support made a stronger contribution to adolescent depression, compared with general peer support (Auerbach et al., 2014).
Support of teachers
Teacher support played a very limited role in influencing later psychological distress and bullying victimization, specifically, it had a direct effect on distress only. Previous research, although very limited, is broadly consistent with these findings. One study found that supportive teacher relationships had a main effect on depressive symptoms but did not moderate ongoing depression for passive victims of bullying (Huang et al., 2018). Another study found that supportive relationships with teachers did not predict later bullying victimization (Ten Bokkel et al., 2021). Perhaps support from teachers has mixed impacts on adolescents’ peer relationships and bullying victimization, depending on how it is enacted. For instance, overt teacher support may sometimes lead to resentment from peers which impacts peer relationships.
Impact of Supportive Relationships on Bullying Victimization
In this study, supportive relationship had limited impacts on later bullying victimization. Although support by parents and classmates interacted with previous bullying victimization to influence later bullying victimization, none of the relationship variables had a direct effect on bullying victimization. Similarly, no relationship variables had a direct effect or buffering effect on the prediction of bullying victimization from earlier distress. Table 1 shows that bullying victimization at Time 2 had relatively low levels of association with all other variables, including bullying victimization at Time 1 and psychological distress at Time 2.
Unlike psychological distress, there was a significant reduction in bullying victimization between the two waves of data collection. The observed maintenance of psychological distress over time is consistent with previous research that shows that this can perpetuate for months and years after bullying ceases (Moore et al., 2017). Perhaps bullying reduced between timepoints because involvement in this study prompted greater awareness of school staff or more requests for help from students who were bullied. A similar reduction was reported in the control condition of an RCT of a family intervention in primary schools, and was hypothesized as attributable to actions by school staff (Healy & Sanders, 2014). Another possible reason for the reductions in bullying victimization is that the 3-month timescale was easier for adolescents to quantify the second time they did the questionnaire, as answering the first questionnaires in itself would have provided a point of reference.
Strengths and Limitations
Strengths of this study included the longitudinal design, a large sample, use of well-validated measures, distinction between different types of relationship support, and management of missing data through multiple imputations. However, several limitations should be kept in mind when interpreting results. We acknowledge the limitations of a convenience sample which resulted in an over-representation of male adolescents, due to involvement of two male-only schools. The study was also conducted in Australia, a high-income country where concerns about bullying victimization in schools are widely acknowledged. Other factors inherent to the design may have influenced the representativeness of the sample. Use of active parental consent (required for ethical clearance), has been found to result in under-representation and dropout of students with lower levels of functioning and behavior problems, including perpetration of bullying (Shaw et al., 2015; Wolke et al., 2009). This can lead to marginal underestimation of the strength of prediction in regression models (Wolke et al., 2009). Future studies could replicate the current study with different samples including those with more equal gender balance, in other countries and with different socio-economic levels, to assess the generalizability of findings.
Other Implications for Further Research
This was one of the first studies to examine the impact of different types of relationship support on adolescent outcomes of bullying victimization and psychological distress. Differing results for different kinds of relationships suggests that the effectiveness of relational support depends on the type of relationship. Previous research has not distinguished between friends and classmates and has reported mixed results. In the current study, classmates were more influential than friends. Future research could investigate our hypothesized reasons for this difference (including presence of classmates during adverse social events). This finding may also help elucidate the reason for mixed results in previous research investigating the impact of friendships.
Practical Implications for Supporting Adolescents who are Bullied
The central roles of classmates and parents in this study is consistent with intervention research. A meta-analysis comparing different elements of school bullying prevention programs found that programs that included parental involvement and informal peer support were associated with reduced bullying victimization over time (Gaffney et al., 2019). A targeted family program, that included strategies for parents to provide emotional support, and coach their child in improving their peer relationships, reduced overt victimization and depression in children bullied by peers (Healy & Sanders, 2014). Children who continued to experience bullying victimization after a whole-school bullying prevention program was implemented were distinguished by particularly high levels of peer rejection and less supportive parenting (Kaufman et al., 2018). Similarly, a study of non-responders to a family intervention targeting children bullied by peers found that children with clinically elevated symptoms of depression at the 9-month follow-up were distinguished by lower levels of peer and supportive parenting at the 3-month post-assessment (Healy & Sanders, 2017).
These findings have practical implications for improving programs to prevent bullying and reduce the psychological distress of victims. School bullying prevention programs are widely implemented in many countries but achieve only small reductions in bullying victimization and perpetration (Fraguas et al., 2021; Gaffney et al., 2019), meaning that many adolescents who are victimized do not benefit from the school-based programs alone. Parental components of school-based bullying prevention programs have been understudied, and vary between programs from information sheets to in-person workshops (Gaffney et al., 2021). Further development of parenting components may strengthen the impact of school-based bullying prevention programs. In addition, a new body of research evaluates targeted programs to increase parenting and peer support of children bullied by peers (Healy et al., 2022). This approach may be applicable to adolescents.
Conclusions
In summary, this study provides further evidence of the key role of supportive relationships, especially those with classmates and parents, in protecting adolescents from the inter-related problems of psychological distress (internalizing problems) and bullying victimization. Although most school bullying prevention programs already have a strong focus on improving support from peers, less attention has been given to parental involvement. Outcomes for adolescents who are bullied may be improved through developing and utilising interventions that enhance parental as well as classmate support.
Data collection for this study was supported by a philanthropic PhD scholarship awarded to the third author (HJT) by the Bryan Foundation (2013-2016). Ethical clearance was obtained from The University of Queensland (UQ 201400759), as well as approval from relevant education authorities and school principals. Data is not available for other purposes as we did not obtain consent for this. The authors declare no conflict of interest.
Compliance with ethical standards
Conflict of interest
The authors declare no competing interests.
Ethics and informed consent
Ethical clearance was obtained from The University of Queensland (UQ 201400759), as well as approval from relevant education authorities and school principals. Data are not available for other purposes as we did not obtain consent for this.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Met BSL Psychologie Totaal blijf je als professional steeds op de hoogte van de nieuwste ontwikkelingen binnen jouw vak. Met het online abonnement heb je toegang tot een groot aantal boeken, protocollen, vaktijdschriften en e-learnings op het gebied van psychologie en psychiatrie. Zo kun je op je gemak en wanneer het jou het beste uitkomt verdiepen in jouw vakgebied.
