Peers are the measure of all things: moderating and mediating the association between victimization and depressed affect among Brazilian adolescents

Introduction: Internalizing symptoms in general and depression in particular is a significant health concern among adolescents, with peer victimization as a consistent risk factor. The current study aimed to examine the association between peer victimization and depressed affect among Brazilian adolescents, as well as the potential buffering effect of close friends and the indirect effect of social support on this association.

Methods: Data were analyzed from the PeNSE National Survey of School Health, a nationally representative sample of 5th to 11th grade Brazilian students (final N = 154,122). Participants completed self-administered questionnaires assessing peer victimization, number of friends, social support, and internalizing symptoms.

Results: Peer victimization was found to be a significant predictor of depressed affect, and reporting more close friends was related to lower depressive symptoms. The study also found evidence for a curvilinear relationship between the number of close friends and depressed affect. More importantly, the number of close friends buffered the effect of peer victimization whereas peer support mediated the association to depressed affect.

Discussion: These findings highlight the importance of addressing both peer victimization and social support in promoting mental health among Brazilian adolescents in addition to making the school environment less challenging for all students.

1 Introduction

Protagoras (l.c. 485–415 BCE) is credited for the saying that “man(kind) is the measure of all things” in as much as our perceptions of the physical world serve as a reference point for what can be considered good or bad. Considering that, as adolescents' perceptions of their relations with peers plays such an important role in their mental health insinuates that, for adolescents at least, peers are the measure of all things. To which, depression is a prevalent mental health issue among adolescents, and peer victimization has been consistently identified as a risk factor. Peer victimization refers to repeated intentional harm inflicted upon an individual by their peers and can have negative impacts on various domains of functioning (da Cunha and Santo, 2023), not least of which, mental health (McDougall and Vaillancourt, 2015). Social support on the other hand, has been shown to be associated with better mental health outcomes and may serve as a protective factor against the effects of stressors, including peer victimization. Relatedly, positive relationships in general and friendships in particular can buffer the impact of peer victimization. This study examines the relationship between peer victimization and depression using a large representative sample of Brazilian adolescents, as well as the potential buffering effect of social support on this relationship. Using structural equation modeling, the current paper aims to test several established hypotheses regarding the various ways peers can qualify the link between peer victimization and depressive symptoms. The examination of this topic is important as it can inform the development of interventions aimed at promoting mental health among adolescents in Brazil specifically, and possibly the rest of the world.

1.1 Depression during adolescence

Adolescent depression is a significant public health concern, affecting up to 20% of youth worldwide (Merikangas et al., 2010; Santo et al., 2018). The onset of depression during adolescence can have negative impacts on multiple areas of life, including academic performance, social relationships, and overall wellbeing (Hammen et al., 2014). Depression is a chronic, recurrent condition that is characterized by a persistent irritable, sad, or bored mood and difficulty with familial relationships, school, and work that frequently first occurs in childhood or adolescence (Hankin et al., 1998). Research has identified various risk factors associated with adolescent depression, including genetic predisposition, life stressors, and cognitive vulnerabilities (Hankin, 2018; Essau, 2008). Importantly, aspects of the peer environment have been tied to depression from lack of social support, friends, and positive relationships in school (Klimes-Dougan et al., 2007) along with peer victimization (Juvonen and Graham, 2014).

1.2 Peer victimization

Peer victimization refers to experiences of intentional harm inflicted toward an individual by one or more peers (Velásquez et al., 2010). This behavior can take many forms, including physical aggression, verbal bullying, social exclusion, or cyberbullying (Velásquez et al., 2016; Martin-Storey et al., 2021a). Peer victimization is a widespread problem, affecting a significant proportion of adolescents and young people around the world (McDougall and Vaillancourt, 2015) and can have long-lasting negative effects on their mental and physical health, academic performance, and future relationships. Peer victimization can also have serious and lasting consequences for both the victim and the perpetrator, affecting a range of outcomes including mental health, academic performance, and social development (Adams et al., 2021). However, the impact of peer victimization is not monolithic in that several factors can modulate the severity or explain the processes at play (Bass et al., 2022). The current study is designed to elucidate the relations between peer victimization and depressed affect by including multiple peer relations variables as moderators and mediators elucidated below.

1.3 Friendships and peer support

There is mounting evidence to suggest that having supportive friendships can circumvent the negative impact of peer victimization on adolescent mental health outcomes (Hodges et al., 1999). Specifically, the friendship protection hypothesis states that friendships can serve as a protective factor for individuals facing negative experiences such as peer victimization. It its simplest form, children and adolescents with friends are less likely to be victimized, for the simple fact that they may have friends present. One study found lower levels of peer victimization among adolescents with higher number of friendships (Demaray and Malecki, 2003).

Moreover, the friendship protection hypothesis also proposes that the presence of supportive friendships can buffer the negative impact of peer victimization on mental health outcomes (Hawker and Boulton, 2000). According to this hypothesis, individuals with strong friendships are better able to cope with the stress of victimization and may be less likely to experience negative mental health outcomes as a result (Adams et al., 2011). For example, one study found that adolescents who reported high levels of peer victimization but also high levels of friendship quality had lower levels of depressive symptoms than those with low levels of friendship quality (Rose and Rudolph, 2006). Another found that having a best friend moderated the relationship between peer victimization and psychological distress, such that the negative impact of victimization was reduced for those with a best friend (Bernasco et al., 2022). Given these findings, we aim to use friends, specifically the number of close friends, as a moderator/buffer of the association between peer victimization and depressive symptoms.

Furthermore, research has shown that having a larger social support network, including friends, is associated with better mental health outcomes for adolescents (Stice et al., 2004). This suggests that having multiple supportive friendships, rather than relying on a single best friend, may also be beneficial. Social support is crucial for the healthy development and wellbeing of adolescents (Thomas et al., 2019). Adolescents who have strong social support networks are less likely to experience depression and other mental health problems and are more likely to have positive relationships and academic achievement (Furman and Buhrmester, 1985). Additionally, social support can provide a source of emotional and practical assistance in times of need, helping adolescents to cope with stress and adversity (Sheeber et al., 1997; da Cunha et al., 2021). These studies hint that the association between peer victimization and symptoms of depression might be indirectly explained by peer support.

Social support has been found to play a crucial role in qualifying the impact of peer victimization among adolescents (Burke et al., 2017). Research has shown that having a strong network of supportive relationships can provide a sense of security, comfort, and validation, which can indirectly counteract the negative effects of bullying and other forms of victimization (Wood et al., 2017). Research has shown that peer support can mediate the association between peer victimization and depression among adolescents. For instance, a study in the United Kingdom by Qualter et al. (2013) found that peer support mediated the relationship between peer victimization and depressive symptoms among 1,428 adolescents. Similarly, a study by Vanhalst et al. (2019) found that peer support mediated the association between victimization and depression in a sample of 482 Belgian adolescents. As such, in this paper we aim to test whether peer support serves as a mediator of the association between peer victimization and depression symptoms. Figure 1 illustrates a conceptual model of the moderating and mediating relations between the variables of interest in the current study. While this literature highlights the importance of peer social support on the association between peer victimization and symptoms of depression, many of the findings tend to come from WEIRD samples.

Figure 1

Figure 1. Conceptual model of the theorized associations of the moderators and mediators of the association between peer victimization and depressed affect.

1.4 From WEIRD to WILD

Psychology's evidence base still leans heavily on “Minority-World” contexts, with top journals dominated by European and other Western samples, a pattern that constrains generalizability and centers White, Euro-American experiences as the human “universal standard” (Henrich et al., 2010; Thalmayer et al., 2021). Recent decolonial arguments urge a re-centering on the Majority World, where approximately 85% of the world's adolescents live, and whose perspectives remain underrepresented in developmental science (Abubakar et al., 2024). In tandem, a WILD agenda (Worldwide, In situ, Local, and Diverse) calls for theories and methods that are grounded in local cultural ecologies rather than extrapolated from Western middle-class contexts (Santo et al., 2013; Bard et al., 2025).

Responding to this call, the present study offers a deep dive into a comprehensive Brazilian sample, leveraging nationally representative data to examine peer victimization, friendships, and support in relation to depressed affect among Brazilian adolescents.

1.5 Peer victimization, friendships, and depressed affect in the Brazilian context

Focusing on the Brazilian context, research shows that bullying and peer victimization are persistent problems, with large surveys, including the Pesquisa Nacional de Saúde do Escolar (PeNSE) consistently find that between 6% and 10% of adolescents report frequent victimization, with higher risks among girls, Indigenous and Afro-descendant students, and those who experience social exclusion (Silva et al., 2018; Veloso et al., 2020). Victimization is also associated with loneliness, insomnia, and lack of friendships, pointing to clear links between peer aggression and depressive symptoms, while illustrating how such victimization experiences are intertwined with broader social inequalities.

Beyond individual risk factors, studies emphasize the protective role of friendships and positive school climates. Analyses from the São Paulo PROSO-Brazil project show that supportive teacher-student and peer relationships reduce the likelihood of both victimization and perpetration, whereas disorganized or illegitimate school climates increase vulnerability (Nunes et al., 2025). Complementary evidence highlights that peer support can buffer the negative effects of bullying, helping adolescents cope with stress and reducing the risk of externalizing and internalizing problems (da Cunha et al., 2021; Roza et al., 2021). A consistent pattern across studies is that lack of friends exacerbates the association between victimization and poor health outcomes, underscoring the central role of peer networks in attenuating the impact of negative experiences.

Depressive symptoms represent one of the concerning repercussions of peer victimization in Brazil. Adolescents exposed to bullying are more likely to report sadness, hopelessness, low self-esteem, and social withdrawal (Forlim et al., 2014; Valle et al., 2015; Silva et al., 2018; Veloso et al., 2020). At the same time, research shows that supportive friendships and positive school environments can attenuate these risks (Roza et al., 2021; da Cunha et al., 2021; Nunes et al., 2025). Together, these findings situate Brazilian youth within the broader literature while pointing to specific contextual features, such as the variability in school climates, may shape the interplay between victimization, friendships, and depressed affect.

1.6 PeNSE data

The PeNSE (National Survey of School Health) is a nationally representative survey of school health in Brazil (Ministry of Health, n.d.). The survey provides data on a wide range of health topics, including physical activity, nutrition, tobacco use, alcohol consumption, sexual behaviors, and mental health, among others (Roza et al., 2021). The PeNSE is administered every 3 years and provides data at the national, state, and municipal levels (Ministry of Health, n.d.). The data collected through the PeNSE is used to monitor trends in health behaviors, to evaluate the impact of health interventions and programs, and to inform health policies aimed at improving the wellbeing of young people in Brazil (Ministry of Health, n.d.). The most recent wave, collected in 2019, is a valuable resource for understanding the factors that can impinge the mental health of adolescents.

1.7 Current study

An extensive literature suggests that experiences of victimization are associated with higher rates of depressed affect. Consequently, aspects of the peer context such as close friends and social support have been shown to temper and qualify this association. The current study focused on the association between experiences of victimization and depressed affect among a large sample of Brazilian youth using structural equation modeling. Consistent with previous research, it was expected that a strong association between peer victimization and depressed affect would be observed. However, those reporting more close friends were expected to be victimized less (supporting the friendship protection hypothesis). More importantly, the strength of the association between victimization and depression was expected to be buffered (i.e.,: moderated) by youths' close friendships. It's worth noting that aim to go beyond assuming solely linear relations between the variables in the current study and test potential curvilinear associations as well. Finally, the association between victimization and depressed affect was expected to be, at least partially, indirectly accounted (i.e.,: mediated) for by peer social support.

2 Method

2.1 Participants

The PeNSE - National Survey of School Health (2019) data was collected from a nationally representative sample of students in public and private schools in Brazil. The sample consisted of 165,838 students in the 5th through 11th grades. The students were selected using a multi-stage random sampling design, with schools and classes being selected in the first stage and students being selected in the second stage. The data was collected between May and June of 2019. All students enrolled in the 5th, 9th, and 11th grades in public and private schools in Brazil were eligible to participate in the PeNSE 2019 study. The sample was well-balanced in terms of gender and race/ethnicity, with approximately equal representation of male and female participants and a diverse representation of racial and ethnic groups.

2.2 Procedures

Data were collected using a self-administered questionnaire that was completed by the students in their classrooms. The questionnaire included a range of questions on health-related topics, including physical health, mental health, and health behaviors. The questionnaire also included questions on social support, peer victimization, and other relevant topics. The students were instructed to complete the questionnaire independently and to return it to the researchers after completing it. The data was double-entered and checked for errors and inconsistencies. The data was then cleaned and prepared for analysis. The data is stored securely and confidentially, in accordance with ethical and legal guidelines for the protection of human participants in research and available for download and secondary data analyses.

2.3 Measures

2.3.1 Demographics

Demographic information was collected for all participants (n = 165,838) including age, gender, race/ethnicity, socioeconomic status, and geographic location. Given the random sampling approach of the PeNSE dataset, every student in Brazil had the potential of being included. Regarding age, 15.50% of the sample was 12 and under, 49.70% between the ages of 13–15, 25.60% between the ages of 16–17 and 5.00% 18 and older (with 4.20% reported as missing) while 47.00% were male and 48.70% female (again, with 4.20% missing). Most of the students were white (36.40%) or of mixed ethnicity (41.30%) and the remaining were black (10.10%), Asian (3.30%) and indigenous (2.90%; with 6.00% reported as missing). In the current report, ethnicity was dummy coded with white students as “1” and the remaining students as “0.” A composite of socio-economic status was constructed by averaging items (“having a cellphone,” “a computer or laptop,” “internet access,” “access to a car” and “number of bathrooms at home”) in addition to mother education (α_{(6 items)} = 0.74). Finally, attending private school was dummy coded as “1” and the remaining students as “0.”

2.3.2 Depressed affect

The students were asked to report how often in the last 30 days they felt “sad?” (item B12004 in the survey), “irritated, nervous, or moody about anything?” (B12005), “that no one cares about you?” (B12006) and “that life is not worth living?” (B12008). Participants were asked to rate the frequency of each statement on a five-point Likert scale, ranging from “never” (1) to “always” (5). The items were standardized prior to analyses and were internally consistent with eachother (α = 0.80).

2.3.3 Peer victimization

To assess experiences of intentional harm inflicted toward an individual by one or more peers, participants were asked to report how often in the last 30 days classmates “bullied, or teased so much that you were hurt, annoyed, offended, or humiliated?” (B07007A), “refused to talk to you, pushed you aside for no reason, or caused other classmates to stop talking to you?” (B07011), “hit (punched, slapped, kicked, pushed) you or otherwise physically hurt you?” (B07012). They were asked to rate the frequency of each on a three-point Likert scale, ranging from “no days in the last month” (1), “once” (2) or “two or more times” (3). Meanwhile, one item “threatened, offended, or humiliated on social media or mobile apps?” (B011013) was binary (No, coded as 1 and Yes coded as 3). The items were standardized prior to analyses and were marginally internally consistent (α = 0.55).

2.3.4 Friendships and peer support

Students in the study indicated the number of close friends they had (B12003) from “none” (1), “1 friend” (2) or “2 friends” (3) and “three or more” (4). Most of the sample responded with “three or more” (78.20%). Whereas, peer social support was measured based off “how often have your schoolmates treated you well and/or been helpful to you? (B07006) from “Never” (1), “Rarely” (2) or “Sometimes” (3) and “Most of the time” (4) and “Always” (5). Most of the sample responded between “Sometimes” and “Always” (80.10%). These items too were standardized prior to analyses.

2.3.5 Weighting

Given the complex nature of large representative school data, analyses of PeNSE data included weights for each participant in the sample, calculated with the objective of accurately representing all students. The development of the sample weight took into consideration the weights for the schools, the classes and the students, specifically reflecting the selection probability of students within the selected classes who had provided valid responses. The weight for each student was created with post-stratification of the total number of students enrolled in the school census. The sample weights enabled to estimate the number of students enrolled who were present in the school (M = 93.97; ranging from 0 to 2,453.90). The weights had a zero value for the records of infrequently enrolled students. The analyses were conducted on the data with and without the use of the weighting variable with no differences in the results. Nevertheless, the results with the weighting variable are detailed below.

2.3.6 Analytic approach

First, zero-order correlations were examined for all of the study variables in addition to t-tests to explore differences based on public/private school, gender and the ethnicity dummy code. Hypothesis testing was conducted using structural equation modeling to allow the creation of latent factors and model the moderating and indirect effects (Muthén and Muthén, 2012). The items of depressed affect were combined into a latent construct that was used the dependent variable. First, we accounted for the effects of the covariates (using the private school dummy code, gender, age, the ethnicity dummy code, and SES) on depressed affect. Then we modeled the association with a latent factor of peer victimization. The main effect of number of friends was added to the model before testing moderating effect on peer victimization, then followed by the curvilinear association. Finally, the main effect of peer support was included in the model before testing for indirect effects. Given the size of the sample, there was a high likelihood of even trivial effects being statistically significant (alpha = 0.05). As such, the interpretation of significant effects was paired with estimates of effect size (as measured using proportional reductions in the variance of the outcomes) and significant improvements to the estimation of the models (based on decreasses in the Bayesian Information Criterion, BIC).

3 Results

Descriptive statistics of the relevant variables are included in Table 1 along with the zero order correlations. All of the associations were in the expected directions. The strongest correlation was between the private school dummy variable and SES (r = 0.55, p < 0.05), not surprisingly.

Table 1

Table 1. Zero-order correlations among the relevant variables (^*denotes p < 0.05), along with means and standard deviations.

Independent samples t-tests were used to test for mean differences in the relevant variables based off attending a private school, gender and the ethnicity dummy code. A number of significant differences emerged (also evidenced in the point biserial correlations in Table 1). Notably, private school adolescents reported more peer support (M = 3.90, S.D. = 0.99) than those in public school [M = 3.63, S.D. = 1.23; t_{(153, 807.85)} = −48.39, p < 0.05] and more friends (M = 3.73, S.D. = 0.64) than those in public school [M = 3.60, S.D. = 0.79; t_{(154, 314.78)} = −36.38, p < 0.05]. Meanwhile, boys reported slightly less peer support (M = 3.73, S.D. = 1.12) than girls [M = 3.79, S.D. = 1.10; t_{(156, 447.69)} = −10.11, p < 0.05] and slightly more friends (M = 3.69, S.D. = 0.72) than girls [M = 3.64, S.D. = 0.73; t_{(158, 094.16)} = 14.73, p < 0.05]. Finally, white adolescents reported more peer support (M = 3.83, S.D. = 1.07) than others [M = 3.71, S.D. = 1.17; t_{(136, 356.33)} = −20.94, p < 0.05] and slightly more friends (M = 3.69, S.D. = 0.69) as well [M = 3.65, S.D. = 0.74; t_{(134, 455.87)} = −10.33, p < 0.05].

Hypothesis testing was conducted using structural equation modeling (in M-Plus, ver. 8.0). In the first model, a latent factor of depressed affect was created. This model was a good fit to the data [${\chi }_{(2)}^2$ = 1,057.74, p < 0.05; CFI = 0.99; RMSEA = 0.06, 90% CI (0.06, 0.06); SRMR = 0.01]. The factor loadings were above 0.74 (save one at 0.56), providing acceptable reliability (ω = 0.81).

In the second model, the covariates (gender, age, SES and the private school and ethnicity dummy codes) were added as correlates of depressed affect. Similar to the correlations girls (b = 0.67, S.E. = 0.01, z = 65.84, p < 0.05), older participants (b = 0.12, S.E. = 0.01, z = 20.02, p < 0.05) and those of a high SES (b = 0.05, S.E. = 0.01, z = 6.30, p < 0.05) reported more depressed affect whereas private school participants reported fewer (b = −0.04, S.E. = 0.01, z = −3.95, p < 0.05). There was no effect of ethnicity. All told, the covariates accounted 14.31% of the variability in depressed affect, significant improving the model (ΔBIC = 65,889.40).

At this point, the latent factor of peer victimization was introduced to the model. Though it was marginally internally consistent (ω = 0.56), it was a good fit to the data [${\chi }_{(2)}^2$ = 143.40, p < 0.05; CFI = 1.00; RMSEA = 0.02, 90% CI (0.02, 0.02); SRMR = 0.01]. Peer victimization was then included as an additional predictor of internalizing symptoms and was significant and positive (b = 0.70, S.E. = 0.01, z = 56.70, p < 0.05). As expected, more peer victimization was related to more depressed affect (ΔR² = 34.41%). The potential curvilinear effect of peer victimization was tested but it was not significant and worsened the model. As such, it was removed from the remaining analyses.

Next, the number of close friends was included as an additional correlate of depressed affect. As hypothesized, reporting more close friends was related to lower depressive symptoms (b = −0.15, S.E. = 0.01, z = −15.77, p < 0.05; ΔBIC = 5,570.40). Moreover, there was also a curvilinear effect as well (b = −0.03, S.E. < 0.01, z = −6.75, p < 0.05; ΔBIC = 241.86). As illustrated in Figure 2, though generally negatively associated with depression, having three or more close friends was linked to markedly lower depressed affect (ΔR² = 1.30%). Specifically, the benefits of having friends as reflected in the raw values of depressed affect were modest among students who reported having one friend (5.30% of the sample), compared to none (3.30%). However, among those who reported having two friends (13.00%) or three or more (78.20%), the model predicted depressed affect factor scores that were significantly lower in magnitude.

Figure 2

Figure 2. The bar graphs reflect the raw values in the measure of depressed affect separated by the number of close friends reported. Errors bars representing the 95% confidence intervals (with zero depicting the latent mean). The solid line illustrates the predicted model latent mean score (controlling for the covariates, peer victimization, and the sampling weights) as a function of the number of friends.

To test the friendship protection hypothesis, the effect of number of close friends was added as an estimator of peer victimization. In line with the hypothesis, reporting more close friends was related to lower peer victimization as well (b = −0.08, S.E. = 0.01, z = −16.52, p < 0.05) but there was no curvilinear effect (see Figure 3). This accounted for a 3.43% reduction in variability of peer victimization, further improving the model (ΔBIC = 1,634.57).

Figure 3

Figure 3. The bar graphs reflect the raw values in the measure of peer victimization separated by the number of close friends reported. Errors bars representing the 95% confidence intervals (with zero depicting the latent mean). The solid line illustrates the predicted model latent mean score (controlling for the covariates and the sampling weights) as a function of the number of friends.

Now, the hypothesis that having friends would buffer the effect of peer victimization on depressed affect could be tested, above and beyond all the other variables thus far. To do so, the latent peer victimization by number of close friends interaction was created and added to the model. Here too, the hypothesis was supported (b = −0.03, S.E. = 0.01, z = −3.79, p < 0.05; ΔBIC = 56.36). As expected, reporting having more close friends was associated with less depressed affect even among those who were victimized (see Figure 4). This effect was less pronounced at lower levels of peer victimization but perceptibly stronger at higher levels.

Figure 4

Figure 4. Model estimates of the effect of peer victimization on depressed affect as a function of the number of friends reported. Zero depicts the latent mean. The bars reflect the relative proportion of youth at each level of peer victimization.

Lastly, the impact of peer support on depressed affect was tested. Specifically, peer support was regressed on victimization while peer support was added as a correlate of depressive symptoms. Peer victimization was associated with decreased reports of peer support (b = −0.64, S.E. = 0.02, z = −42.88, p < 0.05) accounting for 4.20% of the variability. Moreover, there was a negative linear (b = −0.03, S.E. = 0.01, z = −4.73, p < 0.05) and curvilinear (b = −0.05, S.E. < 0.01, z = −12.66, p < 0.05) association between peer support and depressed affect. As illustrated in Figure 5, though generally negative, compared to those who reported never feeling supported by their peers, those who rarely felt so had higher depressive symptoms. The inclusion of peer support explained an additional 1.06% to the modeling of depressed affect (45.40% all told; ΔBIC = 6,565.40).

Figure 5

Figure 5. The bar graphs reflect the raw values in the measure of depressed affect separated by the responses to peer support. Errors bars representing the 95% confidence intervals (with zero depicting the latent mean). The solid line illustrates the predicted model latent mean score (controlling for the covariates, peer victimization, number of friends, and the sampling weights) as a function of peer support.

Perhaps most interestingly were the indirect effects that emerged. First, part of the association between peer victimization and depressed affect was significantly indirectly through decreases in reports of support from peers (Δb = −0.02, S.E. < 0.01, z = −6.38, p < 0.05). Moreover, the potential indirect effect of peer support on the association between number of close friends and depressive symptoms was also tested, given that number of close friends was linked to more peer support (b = 0.17, S.E. = 0.01, z = 28.33, p < 0.05; ΔR² = 1.36%). Notably, part of the association between number of close friends and depressed affect was partially explained by peer support (Δb = −0.01, S.E. < 0.01, z = −6.01, p < 0.05). The final conceptual model is provided in Figure 6.

Figure 6

Figure 6. Final conceptual model of the associations to depressed affect. The indirect effects of peer support are included along with the original association in parentheses.

4 Discussion

The present study revealed a number of interesting findings. First, the correlation matrix showed that the measures of interest were related in the expected directions. There also were modest mean differences based on gender, ethnicity and attending private school. Ultimately though, these effects were less compelling than the results from the structural equation modeling approach. Nevertheless, they replicated previous findings from past versions of the PeNSE (Roza et al., 2021).

Latent factors for peer victimization and depressed affect were modeled along with the association between them. This provided an error-free estimate of the association between the two, above and beyond the effects of the covariates (gender, age, SES and the private school and ethnicity dummy codes). As evidenced in the literature (McDougall and Vaillancourt, 2015), self-reported peer victimization is strongly related to students' admission of depression symptoms. These findings in tandem with others (Bordin and Handegård, 2023) reinforce the troubling link between peer victimization and depressed affect among children and adolescents.

In these data, not only was reporting more close friends tied to fewer reports of peer victimization, but there was also a nuanced association to depressed affect as well. The curvilinear effect revealed that depression symptoms were markedly higher among those with no close friends or just one, as compared to two or more. These findings provide two forms of support for the friendship protection hypothesis (Hodges et al., 1999). To explain, not only are friended students less victimized as has been previously reported, they are also less likely to report feelings of depressed affect even if victimized, also consistent with past literature (Adams et al., 2011). These results reinforce the importance of friendships in the mental health in children and adolescents.

Perhaps the most revealing component of the current study's findings involves the effect of peer support. First off, students were more likely to report that they felt supported by their peers at school if they were friended. Conversely, if adolescents reported being victimized in school, they indicated feeling less supported by their peers. Although these findings were not surprising given the past literature (Burke et al., 2017), it's informative as to what might explain where individual differences in peer support stem from.

This is especially relevant when considering how peer support was related to depressed affect. In line with past research (Wood et al., 2017), not only was peer support tied to fewer symptoms of poor mental health overall, the effect was even stronger at higher levels of peer support, as a function of the curvilinear association. In other words, among adolescents who felt especially supported by their peers, they were all the less likely to report symptoms of depressed affect. To be clear, reducing levels of depressed affect in adolescents would require reducing incidents of peer victimization while also helping foster close friendships.

Finally, part of the association between both peer victimization and number of close friends on depressed affect was significantly indirectly through peer support. To explain, the nature of the relationship is such that understanding how victimization is related to depression symptoms partially through differences in peer support. These mediating effects support past literature (Adams et al., 2021) that peer support serves to explain part of these associations to depression. To our knowledge, this is the first study to examine these associations in combination and delineate the differential impact of number of friends and peer support together.

In Brazil, the impact of peer victimization may differ from other countries because the educational system has expanded rapidly but unevenly, leaving persistent inequalities in students' access and ability to remain in school. As Soares et al. (2021) note, ongoing absenteeism and irregular school trajectories highlight failures not only in access but in permanence, reflecting deeper structural inequities. Although national policies such as Lei n° 13.185/2015 have increased awareness of bullying and encouraged schools to adopt preventive measures, these institutional responses may create the appearance of improvement without fully resolving underlying problems. Furthermore, because school exclusion disproportionately affects the most vulnerable adolescents (Simões, 2016), national declines in measured victimization may reflect underreporting or the attrition of students most severely harmed rather than genuine reductions in bullying. Consequently, the observed effects of peer victimization in Brazil are shaped not only by bullying itself but also by broader issues of educational access, permanence, and equity.

4.1 Strengths, limitations, and implications

The current study has several notable strengths. First, the study utilized a large, representative sample of adolescent students, which enhances the generalizability of the findings to other similar populations. Second, the study controlled for potential confounding variables, including gender, age, SES, private school attendance, and ethnicity, which strengthens the study's ability to isolate the specific effects of peer victimization on depressed affect. Third, the study utilized a nuanced analytic approach that allowed for the exploration of more complex relationships among the variables of interest. Specifically, the use of latent variables ensured that the association between peer victimization and depressed affect was not confounded by measurement error. Additionally, the study explored curvilinear, moderating, and mediating relationships, which provided a more nuanced understanding of how peer victimization affects depressed affect and how other factors, such as social support, may mitigate these effects. Overall, the study's sample size, controls for potential confounds, and analytic approach supports the robustness and generalizability of the findings.

Perhaps the most notable strength of the current study is that it tested multiple variables that different theories have predicted would qualify the impact of peer victimization on the depressive symptoms of adolescents. These findings show that the varying theories (specifically, the friendship protection hypothesis, the buffering effect of friendship on peer victimization and the mediating impact of social support) were simultaneously supported. Although these results are not particularly surprising, it's worth highlighting how complimentary the different theorized effects are. It was only in combination were the effects in the current study able to explain as much as 45% of the variability in the latent factor of depression. Future studies would benefit from taking a similarly multi-faceted approach to the understanding of adolescent depressive symptoms.

The study also has several limitations that should be considered. Firstly, only depressed affect was measured and not other potential indicators of poor mental health (among them, anxiety). Secondly, the scale used did not provide clinical interpretations of the level of depression experienced by participants. Thirdly, different forms of peer victimization were not separated, which may have resulted in a lack of specificity in the results in addition to the latent factor of peer victimization having relatively low internal reliability. Fourthly, the measure of friendship did not take into account whether the relationships were reciprocated, which may have influenced the buffering effect of social support (Wood et al., 2017) nor relationship quality. Additionally, the measure of peer support only included a single item, which may not fully capture the complexity of this construct. Finally, it is important to note that the data were collected before the COVID-19 pandemic, and therefore the results may not fully reflect the current reality for adolescents (Commodari and La Rosa, 2020). The next wave of PeNSE data (collected in 2024 but not yet available) may quantify exactly how these effects have changed.

Based on the current study, schools should consider several factors when considering strategies for reducing peer victimization and depressive symptoms among their students. These findings suggest that fostering close friendships and increasing peer support can have a positive impact on the mental health of adolescents. Schools should prioritize creating a safe and supportive environment where students feel comfortable making friends and seeking peer support. This can be achieved through social events, extracurricular activities, and peer mentoring programs. In addition to fostering close friendships, schools could also consider prioritize increasing peer support (Thomas et al., 2022).

Perhaps most importantly, this study found that self-reported peer victimization is strongly related to students' admission of depression symptoms. Therefore, reducing incidents of peer victimization should be a top priority for schools. Some ways that schools can achieve this include implementing anti-bullying programs, increasing supervision during unstructured times (such as recess or lunch), and encouraging students to report incidents of bullying or victimization (Takizawa et al., 2014). Again, adolescents who reported feeling supported by their peers were less likely to report symptoms of depression. Therefore, schools can create opportunities for students to support each other, such as peer counseling programs or peer-led support groups.

4.2 Concluding remarks

In conclusion, depression is a common mental health issue among adolescents, and peer victimization is a crucial risk factor for its development. Peer victimization is the intentional harm inflicted on an individual by their peers, and it can negatively affect various domains of functioning, including mental health. However, positive relationships, specifically friendships and supportive peers, can buffer the negative effects of peer victimization on mental health outcomes. This study aimed to examine the relationship between peer victimization and depression in a large sample of Brazilian adolescents, the potential moderating role of close friendships and the mediating role of social support. The study's findings could help inform the development of school interventions aimed at promoting mental health among adolescents in Brazil and other parts of the world. Overall, social support and positive relationships, especially supportive friendships, are crucial for adolescents' wellbeing, and steps should be taken to foster them.

Positive relationships with one's peers are crucial to adolescent development because they provide opportunities for socialization, identity formation, emotional support, and positive influence on behavior (Martin-Storey et al., 2021b). It's not surprising therefore that adolescents consistently report that their relations with other students at school are of paramount importance to them. Given these findings bolstered by the remaining literature, among adolescents for good or ill, peers are the measures of all things.

Data availability statement

Publicly available datasets were analyzed in this study. This data can be found here: https://www.ibge.gov.br/estatisticas/sociais/educacao/9134-pesquisa-nacional-de-saude-do-escolar.html.

Ethics statement

Ethical approval was not required for the studies involving humans because the data from the study was part of a government sponsored research project on adolescent health in Brazil. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required from the participants or the participants' legal guardians/next of kin in accordance with the national legislation and institutional requirements because as this was a government project, they weren't required to use informed consent.

Author contributions

JS: Writing – original draft, Writing – review & editing. JC: Writing – original draft, Writing – review & editing.

Funding

The author(s) declare that no financial support was received for the research and/or publication of this article.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Abubakar, A., Brandelli Costa, A., Cui, L., Koller, S. H., Nwafor, C. E., and Raval, V. V. (2024). Towards a decolonial developmental science: adolescent development in the Majority World taking center stage. J. Res. Adolesc. 34, 246–256. doi: 10.1111/jora.12956

PubMed Abstract | Crossref Full Text | Google Scholar

Adams, R. E., Santo, J. B., and Bukowski, W. M. (2011). The presence of a best friend buffers the effects of negative experiences. Dev. Psychol. 47:1786. doi: 10.1037/a0025401

PubMed Abstract | Crossref Full Text | Google Scholar

Adams, R. E., Santo, J. B., and Bukowski, W. M. (2021). Indirect effects of HPA axis dysregulation in the association between peer victimization and depressed affect during early adolescence. Psychoneuroendocrinology 132:105356. doi: 10.1016/j.psyneuen.2021.105356

PubMed Abstract | Crossref Full Text | Google Scholar

Bard, K. A., Keller, H., and Leavens, D. A. (2025). Let's go WILD: increasing inclusivity in theories of developmental psychology. Behav. Brain Sci. 1–60. doi: 10.1017/S0140525X25000044

PubMed Abstract | Crossref Full Text | Google Scholar

Bass, E. C., Saldarriaga, L. M., Velasquez, A. M., Santo, J. B., and Bukowski, W. M. (2022). Unpacking the misfit effect: exploring the influence of gender and social norms on the association between aggression and peer victimization. Int. J. Behav. Dev. 46, 372–381. doi: 10.1177/0165025421992867

Crossref Full Text | Google Scholar

Bernasco, E. L., Van der Graaff, J., Meeus, W. H., and Branje, S. (2022). Peer victimization, internalizing problems, and the buffering role of friendship quality: disaggregating between-and within-person associations. J. Youth Adolesc. 51, 1653–1666. doi: 10.1007/s10964-022-01619-z

Crossref Full Text | Google Scholar

Bordin, I. A., and Handegård, B. H. (2023). Factors associated with peer victimization among Brazilian low-income adolescents. J. Aggress. Maltreatment Trauma 32, 1511–1530. doi: 10.1080/10926771.2023.2203078

Crossref Full Text | Google Scholar

Burke, T., Sticca, F., and Perren, S. (2017). Everything's gonna be alright! The longitudinal interplay among social support, peer victimization, and depressive symptoms. J. Youth Adolesc. 46, 1999–2014. doi: 10.1007/s10964-017-0653-0

PubMed Abstract | Crossref Full Text | Google Scholar

Commodari, E., and La Rosa, V. L. (2020). Adolescents in quarantine during COVID-19 pandemic in Italy: perceived health risk, beliefs, psychological experiences and expectations for the future. Front. Psychol. 11:559951. doi: 10.3389/fpsyg.2020.559951

PubMed Abstract | Crossref Full Text | Google Scholar

da Cunha, J. M., and Santo, J. B. (2023). School (socie)ties: individual and school level differences in the association between ethnic/racial victimization and academic functioning. Front. Educ. 8:1000328. doi: 10.3389/feduc.2023.1000328

Crossref Full Text | Google Scholar

da Cunha, J. M., Thomas, K. J., Sukhawathanakul, P., Santo, J. B., and Leadbeater, B. (2021). Socially responsible children: a link between school climate and aggression and victimization. Int. J. Behav. Dev. 45, 504–512. doi: 10.1177/01650254211020133

Crossref Full Text | Google Scholar

Demaray, M. K., and Malecki, C. K. (2003). Perceptions of the frequency and importance of social support by students classified as victims, bullies, and bully/victims in an urban middle school. Sch. Psychol. Rev. 32, 471–490. doi: 10.1080/02796015.2003.12086213

Crossref Full Text | Google Scholar

Essau, C. A. (2008). Comorbidity of depressive disorders among adolescents in community and clinical settings. Psychiatry Res. 158, 35–42. doi: 10.1016/j.psychres.2007.09.007

PubMed Abstract | Crossref Full Text | Google Scholar

Forlim, B. G., Stelko-Pereira, A. C., and Williams, L. C. D. A. (2014). Relação entre bullying e sintomas depressivos em estudantes do ensino fundamental. Estud. Psicol. 31, 367–375. doi: 10.1590/0103-166x2014000300005

Crossref Full Text | Google Scholar

Furman, W., and Buhrmester, D. (1985). Children's perceptions of the personal relationships in their social networks. Dev. Psychol. 21:1016. doi: 10.1037/0012-1649.21.6.1016

Crossref Full Text | Google Scholar

Hammen, C. L., Rudolph, K. D., and Abaied, J. L. (2014). Child and Adolescent Depression. New York, NY: The Guilford Press.

Google Scholar

Hankin, B. L. (2018). Future directions in vulnerability to depression among youth: Integrating risk factors and processes across multiple levels of analysis. Future Work Clin. Child Adolesc. Psychol. 41, 155–178. doi: 10.4324/9781315187914-12

PubMed Abstract | Crossref Full Text | Google Scholar

Hankin, B. L., Abramson, L. Y., Moffitt, T. E., Silva, P. A., McGee, R., and Angell, K. E. (1998). Development of depression from preadolescence to young adulthood: emerging gender differences in a 10-year longitudinal study. J. Abnorm. Psychol. 107, 128–140. doi: 10.1037/0021-843X.107.1.128

PubMed Abstract | Crossref Full Text | Google Scholar

Hawker, D. S. J., and Boulton, M. J. (2000). Twenty years' research on peer victimization and psychosocial maladjustment: a meta-analytic review of cross-sectional studies. J. Child Psychol. Psychiatry 41, 441–455. doi: 10.1111/1469-7610.00629

PubMed Abstract | Crossref Full Text | Google Scholar

Henrich, J., Heine, S. J., and Norenzayan, A. (2010). The weirdest people in the world? Behav. Brain Sci. 33, 61–83. doi: 10.1017/S0140525X0999152X

PubMed Abstract | Crossref Full Text | Google Scholar

Hodges, E. V., Boivin, M., Vitaro, F., and Bukowski, W. M. (1999). The power of friendship: protection against an escalating cycle of peer victimization. Dev. Psychol. 35:94. doi: 10.1037/0012-1649.35.1.94

PubMed Abstract | Crossref Full Text | Google Scholar

Juvonen, J., and Graham, S. (2014). Peer Harassment in School: The Plight of the Vulnerable and Victimized. New York, NY: Guilford Press.

Google Scholar

Klimes-Dougan, B., Brand, A. E., Zahn-Waxler, C., Usher, B., Hastings, P. D., Kendziora, K., et al. (2007). Parental emotion socialization in adolescence: differences in sex, age and problem status. Soc. Dev. 16, 326–342. doi: 10.1111/j.1467-9507.2007.00387.x

Crossref Full Text | Google Scholar

Martin-Storey, A., Recchia, H. E., and Santo, J. B. (2021a). Self-continuity moderates the association between sexual-minority status based discrimination and depressive symptoms. J. Homosex. 68, 2075–2096. doi: 10.1080/00918369.2020.1733350

PubMed Abstract | Crossref Full Text | Google Scholar

Martin-Storey, A., Santo, J., Recchia, H. E., Chilliak, S., Nardi, H. C., and Da Cunha, J. M. (2021b). Gender minoritized students and academic engagement in Brazilian adolescents: risk and protective factors. J. Sch. Psychol. 86, 120–132. doi: 10.1016/j.jsp.2021.03.001

PubMed Abstract | Crossref Full Text | Google Scholar

McDougall, P., and Vaillancourt, T. (2015). Long-term adult outcomes of peer victimization in childhood and adolescence: pathways to adjustment and maladjustment. Am. Psychol. 70:300. doi: 10.1037/a0039174

PubMed Abstract | Crossref Full Text | Google Scholar

Merikangas, K. R., He, J. P., Burstein, M., Swanson, S. A., Avenevoli, S., Cui, L., et al. (2010). Lifetime prevalence of mental disorders in U.S. adolescents: results from the National Comorbidity Survey Replication–Adolescent Supplement (NCS-A). J. Am. Acad. Child Adolesc. Psychiatry 49, 980–989. doi: 10.1016/j.jaac.2010.05.017

Crossref Full Text | Google Scholar

Ministry of Health (n.d.). PeNSE - National Survey of School Health. Retrieved from https://www.gov.br/saude/pt-br/composicao/svsa/inqueritos-de-saude/pense (Accessed December 10, 2025).

Google Scholar

Muthén, L. K., and Muthén, B. O. (2012). Statistical Analysis with Latent Variables. Mplus User's Guide 1998–2012. Los Angeles, CA: Muthén & Muthén.

Google Scholar

Nunes, M. P. D. C., Okada, L. M., Cunha, K. S., Delfino, H. B. P., Peres, M. F. T., and Azeredo, C. M. (2025). Association between school climate and occurrence of bullying among adolescents: SP-PROSO 2017. J. Interpers. Violence 2:8862605251355615. doi: 10.1177/08862605251355615

PubMed Abstract | Crossref Full Text | Google Scholar

PeNSE - National Survey of School Health (2019). Ministry of Health, Ministry of Education, Ministry of Science, Technology, and Innovation. Available online at: https://www.gov.br/saude/pt-br/composicao/svsa/inqueritos-de-saude/pense (Accessed December 10, 2025).

Google Scholar

Qualter, P., Brown, S. L., Munn, P., and Rotenberg, K. J. (2013). Childhood loneliness as a predictor of adolescent depressive symptoms: an 8-year longitudinal study. Eur. Child Adolesc. Psychiatry 22, 653–661. doi: 10.1007/s00787-009-0059-y

PubMed Abstract | Crossref Full Text | Google Scholar

Rose, A. J., and 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. Psychol. Bull. 132:98. doi: 10.1037/0033-2909.132.1.98

PubMed Abstract | Crossref Full Text | Google Scholar

Roza, T. H., Yano, V. A. N., Roza, S. A., Santo, J. B., and Cunha, J. M. D. (2021). Bullying victimization and friendship as influences on sleep difficulty among Brazilian adolescents. J. Genet. Psychol. 182, 348–360. doi: 10.1080/00221325.2021.1905597

PubMed Abstract | Crossref Full Text | Google Scholar

Santo, J. B., Bukowski, W. M., Stella-Lopez, L., Carmago, G., Mayman, S. B., and Adams, R. E. (2013). Factors underlying contextual variations in the structure of the self: differences related to SES, gender, culture, and “majority/nonmajority” status during early adolescence. J. Res. Adolesc. 23, 69–80. doi: 10.1111/j.1532-7795.2012.00793.x

Crossref Full Text | Google Scholar

Santo, J. B., Martin-Storey, A., Recchia, H., and Bukowski, W. M. (2018). Self-continuity moderates the association between peer victimization and depressed affect. J. Res. Adolesc. 28, 875–887. doi: 10.1111/jora.12372

PubMed Abstract | Crossref Full Text | Google Scholar

Sheeber, L., Hops, H., Alpert, A., Davis, B., and Andrews, J. (1997). Family support and conflict: prospective relations to adolescent depression. J. Abnorm. Child Psychol. 25, 333–344. doi: 10.1023/A:1025768504415

PubMed Abstract | Crossref Full Text | Google Scholar

Silva, J. L. D., Mello, F. C. M. D., Oliveira, W. A. D., Prado, R. R. D., Silva, M. A. I., and Malta, D. C. (2018). Bullying victimization among Brazilian students: results of the national survey of school health (PeNSE). Texto Contexto Enferm. 27:e0310017. doi: 10.1590/0104-07072018000310017s

Crossref Full Text | Google Scholar

Simões, A. A. (2016). As metas de universalização da educação básica no Plano Nacional de Educação: o desafio do acesso e a evasão dos jovens de famílias de baixa renda no Brasil. Brasília: Inep.

Google Scholar

Soares, J. F., Alves, M. T. G., and Fonseca, J. A. (2021). Trajetórias educacionais como evidência da qualidade da educação básica brasileira. Rev. Bras. Estud. População 38:e0167. doi: 10.20947/S0102-3098a0167

Crossref Full Text | Google Scholar

Stice, E., Ragan, J., and Randall, P. (2004). Prospective relations between social support and depression: differential direction of effects for parent and peer support? J. Abnorm. Psychol. 113, 155–159. doi: 10.1037/0021-843X.113.1.155

PubMed Abstract | Crossref Full Text | Google Scholar

Takizawa, R., Maughan, B., and Arseneault, L. (2014). Adult health outcomes of childhood bullying victimization: evidence from a five-decade longitudinal British birth cohort. Am. J. Psychiatry 171, 777–784. doi: 10.1176/appi.ajp.2014.13101401

PubMed Abstract | Crossref Full Text | Google Scholar

Thalmayer, A. G., Toscanelli, C., and Arnett, J. J. (2021). The neglected 95% revisited: is American psychology becoming less American? Am. Psychol. 76:116. doi: 10.1037/amp0000622

PubMed Abstract | Crossref Full Text | Google Scholar

Thomas, K. J., da Cunha, J., and Santo, J. B. (2022). Changes in character virtues are driven by classroom relationships: A longitudinal study of elementary school children. Sch. Ment. Health 14, 266–277. doi: 10.1007/s12310-022-09511-8

Crossref Full Text | Google Scholar

Thomas, K. J., Santo, J. B., and da Cunha, J. M. (2019). The predictive value of school climate and teacher techniques on students' just world beliefs: a comprehensive Brazilian sample. Soc. Psychol. Educ. 22, 1239–1257. doi: 10.1007/s11218-019-09524-3

Crossref Full Text | Google Scholar

Valle, J. E., Stelko-Pereira, A. C., Sá, L. G. C. D., and Williams, L. C. D. A. (2015). Bullying, vitimização por funcionários e depressão: Relações com o engajamento emocional escolar. Psicol. esc. educ. 19, 463–473. doi: 10.1590/2175-3539/2015/0193872

Crossref Full Text | Google Scholar

Vanhalst, J., Klimstra, T. A., Luyckx, K., Scholte, R. H. J., Engels, R. C. M. E., and Goossens, L. (2019). The interplay of loneliness and depressive symptoms across adolescence: exploring the role of personality traits. J. Youth Adolesc. 48, 937–950. doi: 10.1007/s10964-011-9726-7

PubMed Abstract | Crossref Full Text | Google Scholar

Velásquez, A. M., Drury, K. M., Saldarriaga, L. M., Santo, J. B., Stella-Lopez, L., and Bukowski, W. M. (2016). Peer rejection as a social regulation mechanism of group norms: the case of aggression across sex. J. Latino/Latin Am. Stud. 8, 47–58. doi: 10.18085/1549-9502-8.2.47

Crossref Full Text | Google Scholar

Velásquez, A. M., Santo, J. B., Saldarriaga, L. M., López, L. S., and Bukowski, W. M. (2010). Context-dependent victimization and aggression: differences between all-girl and mixed-sex schools. Merrill Palmer Q. 56, 283–302. doi: 10.1353/mpq.0.0054

Crossref Full Text | Google Scholar

Veloso, V. R., Costa, F. B. D. S., Marques, C. C. D. A., Andrade, J. X., Miranda, C. E. S., and Araújo, R. S. D. R. M. (2020). Suffering from bullying and associated factors in Brazilian students aged 13 to 17 years old: a population study. Rev. Bras. Epidemiol. 23:e200097. doi: 10.1590/1980-549720200097

PubMed Abstract | Crossref Full Text | Google Scholar

Wood, M. A., Bukowski, W. M., and Santo, J. B. (2017). Friendship security, but not friendship intimacy, moderates the stability of anxiety during preadolescence. J. Clin. Child Adolesc. Psychol. 46, 798–809. doi: 10.1080/15374416.2015.1094742

PubMed Abstract | Crossref Full Text | Google Scholar