Can school climate or peer preference moderate the adjustment difficulties associated with peer victimization among Chinese middle school students?

Abstract

Most research on peer victimization has focused on Western samples, but in recent years peer victimization in China has become more prevalent. As a result, limited information is available on how peer victimization, school climate, peer preference, and adjustment difficulties impact Chinese middle school students. This cross-sectional study explored these relations. With a sample of 734 7th and 8th-grade students (54.1% male, 45.8% female) from two middle schools in Beijing, China, findings revealed that peer victimization and peer preference predicted adjustment difficulties; however, school climate did not. Although peer preference was not a significant moderator of the association between peer victimization and adjustment difficulties, school climate significantly moderated the relation between relational peer victimization and adjustment difficulties. Limitations, future directions, and implications are further discussed.

Keywords

peer victimization peer preference adjustment school climate Chinese middle school students

Peer victimization, or being a victim of bullying, is an increasing problem among adolescents and has been linked to negative long-term mental health consequences. Research in Western countries has shown a strong relation between peer victimization and internalizing symptoms and externalizing problems, such as depression, anxiety, sleep difficulties, and school dropout (Kumpulainen & Räsänen, 2000). Most of the current research has been conducted in Western countries but, in recent years, peer victimization in China has become a growing concern that requires more attention. For example, earlier studies in China have shown relatively low rates of bullying (e.g., 2%, Eslea et al., 2004), while recent studies have indicated prevalence rates as high as 66% (see Chan & Wong, 2015 for a review). Variability in these findings may be attributed to a number of methodological issues, including differences in study criteria for defining bullying victimization, samples (age group), or sampling procedures (school wide data collection vs. convenience sampling); alternatively, it may be attributable to changes in Chinese society (i.e., changes in family structure such as increased divorce rates and rapid economic development, globalization, and China’s One-Child policy; Cheng et al., 2010; Huang et al., 2013). Nevertheless, it is important to identify protective factors for bullying and adjustment difficulties among Chinese students.

School climate is a factor that has been found to relate to peer victimization and psychosocial adjustment. According to the National School Climate Council (2007), school climate reflects the norms, goals, values, interpersonal relationships, and organizational structures at a school. It also includes the quality of relationships between students and teachers, perceptions of safety, engagement, respect for diversity, and fairness of rules (Bear et al., 2011). Besides school climate, peer preference – how popular (liked) or unpopular (disliked) one is perceived as by their classmates – has also been linked to peer victimization and psychosocial adjustment, as bullying is a peer phenomenon and often occurs in front of observing peers (Coie & Dodge, 1983). Can school climate and/or peer preference serve as a buffer for middle school students who experience bullying? This study sought to answer this question by examining the relationship between peer victimization (i.e., physical, verbal, relational, cyber) and adjustment difficulties as well as the moderating roles of school climate and peer preference. Most prior studies have combined different types of victimization, which fails to capture the unique effects and contributions of each type. Our study fills a gap in the literature by uncovering the relations among different types of victimization, school climate, peer preference, and adjustment difficulties among middle school students in China.

Theoretical framework

This study was guided by Bronfenbrenner’s Person-Process-Context-Time (PPCT, Bronfenbrenner & Morris, 2006) model. It is the interaction of these four components that influences an individual’s developmental outcomes. The first element, person, refers to a child’s individual-level factors, such as their biological characteristics (e.g., gender, age). The second component, proximal process, refers to the reciprocal interaction between an individual and their immediate external environment (e.g., peer victimization). The third component, context, involves four interrelated and interactive system levels: micro-, meso-, exo-, and macrosystems. Finally, the last component of the model, time, involves all the preceding factors (person, context, and process) as they develop and interact over time (Bronfenbrenner & Morris, 2006). PPCT theory informs this study because person (e.g., gender, grade level), proximal process (e.g., peer victimization), and contextual factors (school climate and peers’ status/popularity) may influence each other to predict youth psychosocial adjustment. As part of the macrosystem, values and norms present in Chinese society may influence the child’s experiences.

School climate as a potential moderator

A positive school climate is often associated with less peer victimization and fewer internalizing symptoms among students (e.g., Leadbeater et al., 2015; Olenik-Shemesh et al., 2012; Waasdorp et al., 2011), while more negative perceptions of school climate have been linked to higher rates of delinquency for adolescents in China (Bao et al., 2015). Positive perceptions of school climate may act as a buffer for adolescents by providing supportive environments despite encounters with peer victimization (Wang et al., 2018); however, findings have been mixed, and we do not know if the relations found in American samples apply to Chinese middle school students. For example, Wang et al. (2018) found that positive school climate buffered the relation between traditional victimization and suicidal thoughts and behaviors for Asian American middle school students. Conversely, it was found that more positive perceptions of school climate strengthened the association between cybervictimization and internalizing symptoms (Holfeld & Baitz, 2020). Specifically, when students reported increased levels of safety and caring from teachers at school, they were more susceptible to high levels of internalizing symptoms associated with high cybervictimization (Holfeld & Baitz, 2020). This unexpected finding may be attributed to the “healthy context paradox,” which posits that students who experience victimization in schools where rates of peer victimization are generally lower may attribute victimization to oneself (e.g., “there is something wrong with me that cause this treatment”) and feel worse, since victimization is unexpected in those environments (Salmivalli, 2018, p. 70). Likewise, in schools where students have more negative perceptions of school climate (i.e., poor relationships between students and teachers, unclear or unfair rules), peer victimization may be more common and not consistently addressed; therefore, its impact on students’ internalizing symptoms may be less severe because these students are accustomed to victimization in general. Ultimately, more research is needed to examine these relations, especially among Chinese adolescents, considering the unique context of Chinese schools.

Unique context of Chinese schools

Chinese school context is different from Western schools; however, most studies have focused on Western samples, and less is known about the relations between school climate, peer victimization, and adjustment difficulties in China. For example, Chinese culture (macrosystem) places more emphasis on respect for authority (teachers), education, social harmony, and behavioral regulation as compared with US culture. For Chinese students, learning is a way to contribute to society, and teachers are held in high regard (Li, 2005). In addition to upholding the cultural value of education, Chinese schools are also structured such that character and moral education are included in the curriculum, and students are taught civilized behavior, patriotism, personal integrity, and morality (Zhao, 2005). Furthermore, looping – a classroom arrangement where teachers remain with the same group of students for multiple years – in Chinese schools allows for more opportunities for teacher-student and student-student bonding, social harmony, and improved student achievement (Yang et al., 2013). Collectively, the cultural value of education, the character and model education curriculum, and respect for teachers in Chinese schools could contribute to a more positive school climate and less bullying and victimization than in US middle schools. In addition, because of the focus on social harmony in Chinese schools, school climate and peer preference may be more likely to serve as a buffer between peer victimization and adjustment difficulties for Chinese students.

Peer preference

Sociometric peer preference, or popularity, is measured based on liking (acceptance) and disliking (rejection) by peers (Coie & Dodge, 1983). Adolescents with different peer preferences are likely to have different peer interactions. Those with a low peer preference (e.g., being unpopular and rejected) may be at a higher risk for experiencing peer victimization and adjustment difficulties because they may receive less support from friends (Long et al., 2020). On the other hand, popular peers are less likely to be victimized and more likely to have social support, peer attention, and reinforcement from peers. They are also more likely to have higher self-esteem and lower levels depression and anxiety (Litwack et al., 2012). Thus, peer preference could buffer against adjustment difficulties related to peer victimization. For example, Long et al. (2020) found that, among Chinese adolescents (7^th and 8^th graders), relationally victimized students with low peer preference (or popularity) were more likely to experience popularity status insecurity and were at greater risk for experiencing depression and anxiety; however, this was not the case for victims with high peer preference (i.e., who were more popular). Moreover, the opposite pattern was identified among a group of 7^th graders in the US. Specifically, Swirsky and Xie (2021) found that, although high peer preference was negatively related to peer victimization and loneliness, it actually intensified the relation between victimization and loneliness. The authors speculated that when highly preferred youth experience peer victimization, which contradicts the “typical social script” in schools, they may attribute it to their own negative personal characteristics, which in turn can contribute to negative feelings. Due to discrepant findings regarding the relations among peer victimization, school climate, peer preference, and psychological adjustment in the US and China, further research is needed to clarify these relations and better support victimized students.

Current study

This study sought to answer the following questions: 1) Do different types of peer victimization predict adjustment difficulties? 2) Does school climate predict adjustment difficulties? 3) Does school climate moderate the relation between different types of peer victimization and adjustment difficulties? 4) Does peer preference predict adjustment difficulties? and 5) Does peer preference moderate the relation between different types of victimization and adjustment difficulties? Accordingly, we hypothesized that (1) higher levels of victimization would lead to more adjustment difficulties; (2) more positive perceptions of school climate would lead to fewer adjustment difficulties; (3) students who were more preferred by their peers would have fewer adjustment difficulties; (4) school climate would buffer the relation between peer victimization and adjustment difficulties; and (5) peer preference would buffer the relation between peer victimization and adjustment difficulties.

Methods

Participants

Data were collected from 734 7^th (55.2%, n = 405) and 8th-grade (44.8%, n = 329) students (54.1% male, 45.8% female) from two typical middle schools in Beijing, China. One of these two schools was in an urban area of Beijing and the other was in the suburb. Based on student self-report, family monthly income differed (1% reported making 3000 yuan or less, 9.7% making between 3000-5000 yuan, 20.6% making 5000-8000 yuan, 25.1% making 8000-12,000 yuan, 21.7% making 12,000-20,000 yuan, and 18% making 20,000 yuan and above). Fathers’ highest level of education varied (1.7% = elementary or below, 10.4% = middle school, 25% = high school, 45% = college graduates, and 13.4% = advanced degrees). Mothers’ highest level of education also varied (1.8% = elementary or below, 8.9% = middle school, 28% = high school, 45.9% = college graduates, and 11.2% = advanced degrees). Most fathers (93.7%) and mothers (85.5%) were employed. Parents’ highest level of education in our sample was representative of adults in Beijing because the average levels of education among adults were 12.64 years in Beijing (Yin et al., 2021).

Procedure

The study was approved by the Institutional Review Board at Beijing Normal University. The principals of the two middle schools in Beijing, China agreed to participate in the project. Parents were then notified of the study and given the option to withdraw their children from participation. No parents withdrew their children. The students from the two middle schools completed all measures in Chinese using paper and pencil. Data collection was conducted during students’ regular classes. Trained graduate research assistants in school psychology collected data in the classroom. They read the instructions and answered student questions during data collection. The measures were collected immediately upon completion, and data were de-identified to maintain confidentiality.

Measures

Demographics

Student participants were asked to report their age, gender, grade level, family monthly income, parent highest level of education, and parent employment status (employed or not) via multiple-choice questions.

Adjustment difficulties

Adjustment difficulties were measured using the Social and Emotional Maladjustment Scale from the Swearer Bullying Survey (Swearer, 2001; Werth et al., 2015), which assessed how students perceived their social and emotional maladjustment related to bullying victimization at school. Our team translated the measure into Chinese using the back-translation technique (Rose, 1985) to enhance cultural relevancy in the Chinese version. Specifically, the research team in China translated the measure into Chinese. A bilingual school psychologist in the U.S. back-translated the Chinese version into English. Then the team compared both versions to resolve any discrepancies in translation. Firstly, the definition of bullying was provided: “Bullying happens when someone hurts or scares another person on purpose and the person being bullied has a hard time defending himself or herself. Usually, bullying happens over and over.” Following were some examples of bullying: (a) punching, shoving and other acts that hurt people physically; (b) spreading bad rumors about people; (c) keeping certain people out of a group; teasing people in a mean way; (d) getting certain people to “gang up” on others. Then the students rated how bullying impacted them on six items (e.g., “Made me feel bad or sad”; “Made me feel sick”; “Didn’t come to school”) using a Likert scale (1 = never a problem; 5 = always a problem). Werth and colleagues (2015) conducted a principal component analysis (PCA) to assess the factor structure of the scale. They found that the six-item scale explained 55% of the variance, χ² (15) = 559.22, p <.001, with factor loadings ranging from .57 to .79, and the measure also had good internal consistency (Cronbach's α = .83). For our sample, Cronbach’s α = 0.93.

Peer victimization

Students’ perceived peer victimization was measured using the Delaware Bullying Victimization Scale – Student Chinese Version (DBVS-S; Bear et al., 2016). The scale consisted of 12 items measuring physical (e.g., “I was deliberately pushed by others”), verbal (e.g., “A classmate said mean things to me”), and relational victimization (e.g., “A classmate told others to not be friends with me”). Additionally, four items were added to assess cyber victimization (e.g., “Sent me harsh or hurtful messages using email, mobile phone, text messages, WeChat, QQ, or similar electronic means”). Students indicated the frequency of their perceived peer victimization using a six-point Likert scale (1 = never; 6 = every day). Prior studies have shown that the DBVS-S Chinese Version has high internal consistency (α = 0.70 to 0.82) and strong evidence of validity (Xie et al., 2016). The physical, verbal, and relational victimization subscales also have high internal consistency (α = 0.76 to 0.90; Wang et al., 2020). For this sample, Cronbach’s α for the victimization subscales was as follows: physical (0.82), verbal (0.86), relational (0.87), cyber (0.95), and total victimization (0.89). Only the four victimization subscales (not the total score) were used in the regression analysis.

Peer preference

Peer preference was measured using the peer nomination technique as applied in prior research with Chinse participants (e.g., Long et al., 2020). Students were invited to name three classmates they wanted to play with the most (“popular”), and three classmates they wanted to play with the least (“unpopular”). Nominations for both were tallied and then standardized within classrooms. To calculate peer preference, the standardized “unpopular” item was subtracted from the standardized “popular” item. The subtracted scores were standardized again within classrooms to obtain the peer preference score. A higher peer preference score indicated more peer likability, whereas a lower score indicated less peer likability.

School climate

School climate was measured using three subscales from the Delaware School Climate Survey – Student Chinese Version (Bear et al., 2011). Each subscale has four items which total 12 items for the full measure. We used subscales measuring Teacher-Student Relationships (e.g., “I like my teachers”), Respect for Diversity (e.g., “Students respect others who are different”), and Clarity of Expectations (e.g., “Students know what the rules are”). Students responded to items on a 4-point Likert scale (1 = strongly disagree; 4 = strongly agree). The Chinese version has shown high reliability (0.80) and strong evidence of validity (Xie et al., 2016). Like peer victimization, the internal consistency was high (α = 0.73 to 0.84; Wang et al., 2018). For this sample, Cronbach’s α for each swas as follows: teacher-student relationships (0.84), respect for diversity (0.75), clarity of expectations (0.82), and school climate total (0.88). We only used the school climate total score in the regression analyses.

Data analyses

We first calculated the descriptive statistics and Pearson correlations for key variables. Hierarchical linear regression analyses were conducted using SPSS 24.0 and the PROCESS 3.5.3 package, model 1 (to test moderation). We ran separate models for school climate and peer preference in order to test their unique contributions to adjustment. In addition, to avoid multicollinearity, we mean-centered the independent variables and moderators before calculating the interaction effects. For both regression analyses, grade (0 for 7^th grade and 1 for 8^th grade) and gender (coded as 0 for girl and 1 for boy) were used as control variables. Hierarchical linear regression was used to test the moderation effect. We entered the control variables (gender and grade) and school climate (or peer preference) into the regression model first. We then added the four peer victimization variables. Last, we added the interaction terms to the regression. When a moderation effect was significant, we used simple slope analysis to further examine the effect.

Results

The results showed that 30% of students reported experiencing verbal victimization, 21.2% reported relational victimization, 21.4% reported physical victimization, and 14.2% reported cyber victimization. See Table 1 for means, standard deviations (SD), and correlations for key variables.

Table 1.
Descriptive statistics of variables of interest.

1 2 3 4 5 6 7

1. School Climate

2. Physical Victimization −.19

3. Verbal Victimization −.18 .72

4. Relational Victimization −.16 .74 .75

5. Cyber Victimization −.20 .63 .57 .69

6. Peer preference −.03 −.20 −.25 −.28 −.19

7. Adjustment Difficulties −.01 .42 .44 .51 .36 −.17

Mean 4.21 1.14 1.27 1.14 1.12 .000 1.61

SD .73 . 44 .57 .45 .51 1.63 . 93

Question 1 & 2: Do peer victimization and school climate predict adjustment difficulties?

We used hierarchical linear regression to examine our hypotheses. We first entered the control variables and school climate only into the regression. The model revealed that students’ overall perception of school climate did not significantly predict adjustment difficulties (β = -.01, t(685) = −0.37, p = .71). We then added the four victimization subscales in the model to predict adjustment. Results showed significant main effects for gender/being a boy (β = -.10, t(679) = −2.89, p < .01), school climate (β = .08, t(679) = 2.49, p = .01), verbal victimization (β = .13, t(679) = 2.42, p = .02), and relational victimization (β = .37, t(679) = 6.33, p < .001) but not for physical (β = .07, t(679) = 1.21, p = .23) or cyber victimization (β = .03, t(679) = 0.59, p = .56). This positive relation between school climate and adjustment difficulties may be the result of the suppressor effect, meaning that the relationship between school climate and adjustment difficulties became more positive due to the influence of additional victimization variables (because of the negative correlation between school climate and peer victimization variables).

Question 3: Is school climate a moderator?

After adding the interaction terms to the model, only the interaction between school climate and relational victimization was significant (β = .14, t(675) = 2.38, p = .02; see Table 2).

Table 2.
School climate moderation relations between victimization and adjustment difficulties.

Main effect model Main effect model Interaction model

b(SE) β β b(SE) β

Constant −1.71 (.06) 1.75(.05)* 1.75(.05)*

Gender (1 = boy, 0 = girl) −0.09 (.07) −.05 −0.18 (.06) −.10 −0.19 (.06) −.10

Grade level (0 = 7^th and 1 = 8^th) −0.11 (.07) −.06 −0.06(.06) −.03 −0.06(.06) −.03

School climate −0.02 (.05) −.01 .11(.04) .08** .10(.04)* .08*

Physical Victimization .08(.07) .07 .08(.07) .07

Verbal Victimization .13(.05) * .13* .12(.05) * .12*

Relational Victimization .44(.07)* .37* .41(.07)* .35*

Cyber Victimization .04(.07) .03 .09(.07) .06

School ClimatePhysical Victimization .04(.08) .03

School Climate Verbal Victimization −.09(.07) −.07

School ClimateRelational Victimization .22(.09) .14*

School ClimateCyber Victimization −.05(.09) −.03

F(df) 1.24 (3,683) 39.67 (7, 679) 26.25 (11, 675)*

R² .01 .29 .30

Δ R² .28 .01

Note: p < .05, p < .01, p < .001.

Simple slope analysis showed that when student perceptions of school climate were more positive (1 SD above the mean), the relationship between relational victimization and adjustment difficulties was significant (b = .72, p < .001). When student perceptions of school climate were less positive (1 SD below the mean), the association between relational victimization and adjustment difficulties was also significant (b = .51, p < .001); however, results suggested that a more positive school climate did not buffer but rather intensified the relation between relational victimization and adjustment difficulties (see Figure 1).

Figure 1.
Interaction between school climate and relational victimization on adjustment difficulties.

Question 4: Does peer preference predict adjustment difficulties?

We used hierarchical linear regression to examine the hypotheses. We first entered the control variables and peer preference into the regression. Results revealed that peer preference significantly and negatively predicted adjustment difficulties (β = -.17, t(695) = −4.61, p < .001). We then added four types of peer victimization into the model. Results showed that peer preference was no longer a significant predictor of adjustment difficulties (β = -.04, t(691) = −1.20, p = .23). We believe this is the result of the suppressor effect, meaning the relation between peer preference and adjustment difficulties was reduced due to the influence of the four peer victimization variables (because peer preference was negatively correlated with all four types of victimization). Only gender/being a boy (β = -.09, t(691) = - 2.77, p < .01), verbal (β = .12, t(691) = 2.34, p = .02) and relational victimization (β = .37, t(691) = 6.23, p < .001) predicted adjustment difficulties (see Table 3).

Table 3.
Peer preference moderation relations between victimization and adjustment difficulties.

Main effect model Main effect model Interaction model

b(SE) β b(SE) β b(SE) β

Constant 1.73 (.06) .75(.09)* 0.74(.09)*

Gender (0 = girl, 1 = boy) −0.12 (.07) ⁺ −.07 ⁺ −0.17 (.06) −.09 −0.18 (.06) −.10

Grade level (0 = 7^th and 1 = 8^th) −.11(.07) −.06 −0.09(.06) −.05 −0.09(.06) −.05

Peer preference −.10 (.02)* −.17* −.02(.02) −.04 −.02(.02) −.04

Physical Victimization .07(.07) .06 .10(.07) .09

Verbal Victimization .12(.05) * .12* .12(.05) * .11*

Relational Victimization .44(.07)* .37* .43(.08)* .36*

Cyber Victimization .02(.07) .01 −.002(.07) −.001

Peer preferencePhysical Victimization .06(.04) .09

Peer preferenceVerbal Victimization −.03(.03) −.06

Peer preferenceRelational Victimization −.002(.03) −.003

Peer preferenceCyber Victimization −.02(.04) −.03

F(df) 8.34 (3,695) 38.87 (7, 691) 24.93 (11, 687)***

R² .04 .28 .29

Δ R² .24 .01

Note: p < .05, p < .01, p < .001.

Question 5: Is peer preference a moderator?

After adding the interaction terms to the model, the model did not account for any significant increased variance in adjustment difficulties, and none of the interaction terms were significant (see Table 3). Thus, peer preference did not buffer the effects of physical (β = .09, t(687) = 1.55, p = .12), verbal (β = -.06, t(687) = −1.08, p = .28), relational (β = -.003, t(687) = -.06, p = .96), or cyber (β = -.03, t(687) = −0.62, p = .54) victimization on adjustment difficulties.

Discussion

This study contributes to the literature by closely examining the relations between different types of peer victimization and adjustment difficulties in Chinese middle school students. It also explored the potential moderating effects of school climate and peer preference on the relation between these different types of victimization and psychosocial adjustment.

Peer victimization predicts adjustment difficulties

Consistent with prior research in both Western countries (e.g., Kaltiala-Heino et al., 2000) and China (Cheng et al., 2010), we found that peer victimization (when entered in the model with only the covariates) did predict adjustment difficulties, such that a higher rate of victimization was associated with more adjustment difficulties. Specifically, relational and verbal victimization (instead of physical or cyber victimization) were associated with more detrimental outcomes for Chinese middle school students when all four types of victimization were considered. Additionally, middle school students in our sample more frequently reported verbal victimization (30%) than physical (21.4%) and relational victimization (21.2%). They were least likely to experience cyber victimization (14.2%). Cyberbullying may be less prevalent than other types of bullying in China because access to social media is less common among middle school students, as compared with youth in the US. Moreover, physical victimization may be less prevalent in Chinese schools given that social harmony is strongly encouraged and teachers are more likely to intervene during incidents of physical aggression. On the other hand, verbal and relational victimization may be more detrimental because they target students’ social relationships, something highly valued in collectivistic society.

Peer preference predicts less adjustment difficulties

As hypothesized, peer preference predicted adjustment difficulties (when entered in the model with only the covariates), such that a higher preference score was associated with fewer adjustment difficulties. However, peer preference was not a significant predictor of adjustment when victimization variables were accounted for in the model. We believe this is the result of the suppressor effect, meaning the relation between peer preference and adjustment difficulties is reduced due to the influence of the four peer victimization variables. It is also important to mention that peer preference was negatively correlated with all four types of victimization. As a result, it is still important to promote positive peer preference for students.

This finding is partially congruent with research conducted in Western countries, where youth who are perceived as unpopular by peers are at a greater risk of being victimized and experience adjustment difficulties, whereas youth who are popular may experience fewer adjustment difficulties because of peer support (Gorman et al., 2011). Much less research has been done on the relationship between peer preference and adjustment difficulties in Chinese adolescents. This study contributes to the literature by closely examining how popularity can relate to adjustment difficulties in this population.

School climate as a moderator

Different from our hypothesis, school climate was not related to adjustment difficulties before controlling for peer victimization. After adding peer victimization into the regression, school climate predicted more adjustment difficulties. The positive relation between school climate and adjustment difficulties may be due to the suppressor effect and should be interpreted with caution.

We also found that school climate strengthened the relation between relational victimization and adjustment difficulties. The moderation finding is consistent with those of Yang et al. (2018) and Wang et al. (2021) who found that school-level school climate exacerbated the negative relationship between victimization, academic engagement, and mental health. In line with the “healthy context paradox” phenomenon (Salmivalli, 2018, p. 70), students may blame themselves for the victimization they experience and report more mental health difficulties when they are situated in classrooms where there is less victimization (i.e., healthy environments; Gini et al., 2019). In other words, it is possible that when students who are victimized view their school climate as more positive, they may feel increased loneliness and self-blame because their experience is different from that of their peers (i.e., “Why me?”; Wang et al., 2021). To our knowledge, this is the first study to examine this phenomenon in Chinese middle school students. The results suggested that, for Chinese middle school students who viewed their school climate as more positive, the relation between relational victimization and adjustment difficulties may be exacerbated.

As noted previously, it is possible that school climate was only a moderator of the association between relational victimization and adjustment because Chinese schools emphasize maintaining social harmony. Moreover, relational aggression is generally more covert than other forms of aggression (e.g., physical) and thus can be harder for teachers to notice and intervene in. Relational aggression may be more harmful to students because it targets their relationships with others, which may be highly valued by students in a collectivistic society. As a result, when students perceived a more positive school climate (and thus did not expect to experience peer difficulties), they felt worse when experiencing relational victimization as compared to students who perceived a more negative school climate (and thus expected to experience peer victimization).

Peer preference as a moderator

Although previous studies have examined how peer preference can be a moderator between victimization and adjustment (Long et al., 2020; Swirsky & Xie, 2021), such studies only focused on one or two types of victimization. Our study extended this literature by examining whether peer preference moderated the associations of physical, verbal, relational, and cyber victimization, respectively, with adjustment difficulties in Chinese middle school students. Contrary to our hypothesis, peer preference was not a significant moderator. Students with high peer preference may tend not to express negative emotions in order to maintain their positive image among peers, which may be harmful to their adjustment after victimization. Similarly, if adolescents with high peer preference do not seek support from peers (in order to maintain their positive image) during victimization, peer preference would not necessarily buffer the effects of victimization on adjustment. Other factors, such as peer support (based on student’s perception) may moderate the relationship between victimization and adjustment difficulties instead (Casper & Card, 2017).

Limitations and future directions

Though this study has contributed to the literature in unique ways, it also has its limitations. First, although we collected data from peers regarding students’ popularity/ peer preference, only self-report data were collected about peer victimization and adjustment difficulties. This may have resulted in students underestimating their experiences of victimization and adjustment difficulties due to social desirability bias (Ivarsson et al., 2005). Future studies should collect data from multiple informants (e.g., from the parent, teacher, and self-report) to obtain more accurate measures of victimization and adjustment difficulties. Second, data were only collected from two middle schools in Beijing, China. Therefore, these findings may not be generalizable to schools in other settings (e.g., rural areas). Future studies should consider collecting data from a variety of schools. In addition, we did not control for potential nesting of the data (i.e., students nested within schools). Future researchers should collect data from more schools and analyze data using hierarchical linear modeling to examine the impact of school-level school climate on adjustment difficulties. Lastly, this study utilized a cross-sectional design; thus, the relations observed are only correlational in nature and cannot be used to make causal inferences. Future studies should collect longitudinal data to further examine these relations.

Summary and implications

According to the findings of this study, a substantial percentage of Chinese middle school students experienced different types of peer victimization. Verbal and relational victimization predicted more adjustment difficulties. Perceiving a more positive school climate intensified the association between relational peer victimization and adjustment difficulties, perhaps due to the “healthy context paradox” phenomenon.

Results from our study have several important implications. First, school personnel (including educational and school psychologists) should identify evidence-based programs to decrease victimization and support students who experience adjustment difficulties as a result of such victimization. While positive school climate and school-wide bullying prevention are important (Olweus, 2001), our results suggest that positive school climate alone cannot mitigate the effects of peer victimization on mental health difficulties for all students (Salmivalli, 2018). It is equally important to identify individual students who struggle with bullying and provide additional support, especially when the school climate is generally perceived positively and there are relatively few students experiencing bullying. Our results suggest that students who experience relational victimization and who perceive a positive school climate may be more likely to experience adjustment difficulties. Educators, school-based mental health providers, and parents should attend carefully to supporting these students. For example, it may be beneficial to provide individual and/or group counseling services to victimized students to help them understand that the victimization was not their fault and encourage them to attribute bullying to external factors instead of personal characteristics (Swirsky & Xie, 2021). It is also important to teach students social skills and strategies to cope with bullying in order to reduce adjustment difficulties. Moreover, schools can involve parents in bullying prevention. For example, parents can reinforce the message that bullying is not the students’ fault and help students identify their own strengths.

We also found that relational and verbal victimization were associated with more detrimental outcomes for Chinese middle school students than physical or cyber victimization. Considering that Chinese and East Asian culture value social harmony and relationships, it is important for teachers and parents to take relational victimization seriously. Educators should be prepared to identify relational victimization and to intervene immediately (similar to how they might for physical victimization). Because peer preference is negatively related to victimization, schools should provide students with opportunities for positive peer interaction (i.e., pairing students with peer mentors) to prevent such victimization. It is also important for schools to foster group harmony and congenial relationships among peers to reduce peer victimization.

	1	2	3	4	5	6	7
1. School Climate
2. Physical Victimization	−.19**
3. Verbal Victimization	−.18**	.72**
4. Relational Victimization	−.16**	.74**	.75**
5. Cyber Victimization	−.20**	.63**	.57**	.69**
6. Peer preference	−.03	−.20**	−.25**	−.28**	−.19**
7. Adjustment Difficulties	−.01	.42**	.44**	.51**	.36**	−.17**
Mean	4.21	1.14	1.27	1.14	1.12	.000	1.61
SD	.73	. 44	.57	.45	.51	1.63	. 93

	Main effect model	Main effect model	Interaction model
Constant	−1.71 (.06)		1.75(.05)***		1.75(.05)***
Gender (1 = boy, 0 = girl)	−0.09 (.07)	−.05	−0.18 (.06)**	−.10	−0.19 (.06)**	−.10
Grade level (0 = 7^th and 1 = 8^th)	−0.11 (.07)	−.06	−0.06(.06)	−.03	−0.06(.06)	−.03
School climate	−0.02 (.05)	−.01	.11(.04)**	.08**	.10(.04)*	.08*
Physical Victimization			.08(.07)	.07	.08(.07)	.07
Verbal Victimization			.13(.05) *	.13*	.12(.05) *	.12*
Relational Victimization			.44(.07)***	.37***	.41(.07)***	.35***
Cyber Victimization			.04(.07)	.03	.09(.07)	.06
School Climate*Physical Victimization			__	__	.04(.08)	.03
School Climate* Verbal Victimization			__	__	−.09(.07)	−.07
School Climate*Relational Victimization			__	__	.22(.09)*	.14*
School Climate*Cyber Victimization			__	__	−.05(.09)	−.03
F(df)	1.24 (3,683)	39.67 (7, 679)***	26.25 (11, 675)***
R²	.01	.29	.30
Δ R²		.28	.01

	Main effect model	Main effect model	Interaction model
Constant	1.73 (.06)		.75(.09)*		0.74(.09)*
Gender (0 = girl, 1 = boy)	−0.12 (.07) ⁺	−.07 ⁺	−0.17 (.06)**	−.09**	−0.18 (.06)**	−.10**
Grade level (0 = 7^th and 1 = 8^th)	−.11(.07)	−.06	−0.09(.06)	−.05	−0.09(.06)	−.05
Peer preference	−.10 (.02)***	−.17***	−.02(.02)	−.04	−.02(.02)	−.04
Physical Victimization			.07(.07)	.06	.10(.07)	.09
Verbal Victimization			.12(.05) *	.12*	.12(.05) *	.11*
Relational Victimization			.44(.07)***	.37***	.43(.08)***	.36***
Cyber Victimization			.02(.07)	.01	−.002(.07)	−.001
Peer preference*Physical Victimization			__	__	.06(.04)	.09
Peer preference*Verbal Victimization			__	__	−.03(.03)	−.06
Peer preference*Relational Victimization			__	__	−.002(.03)	−.003
Peer preference*Cyber Victimization			__	__	−.02(.04)	−.03
F(df)	8.34 (3,695***)	38.87 (7, 691)***	24.93 (11, 687)***
R²	.04	.28	.29
Δ R²			.24	.01

Footnotes

Ethical approval

A follow-up study on bullying/victimization and related factors among primary and secondary school students BNU202109100023

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

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

ORCID iDs

Cixin Wang

Ami Patel

Beilei Li

Author biographies

Cixin Wang, PhD, is an Associate Professor of School Psychology in the Department of Counseling, Higher Education, and Special Education at the University of Maryland, College Park. Her research focuses on bullying prevention, mental health promotion, and parenting, especially for youth from culturally and linguistically diverse backgrounds.

Jocelyn Yao, MA, is a graduate student in the Department of Counseling, Higher Education, and Special Education at the University of Maryland, College Park. Her research interests include bullying prevention among children and adolescents and school climate.

Ami Patel, MEd, is a graduate student in the Department of Counseling, Higher Education, and Special Education at the University of Maryland, College Park. Her research interests include early intervention, mental health promotion in South Asians, and culturally responsive interventions.

Beilei Li (corresponding author), PhD, is an Assistant Professor in Faculty of Education, Beijing Normal University. Dr. Li’s research focuses on school-based mental health, and her research interests include the relations between school climate, classroom environment, and student development.

References

Bao

Zhang

Wang

(2015). School climate and delinquency among Chinese adolescents: Analyses of effortful control as a moderator and deviant peer affiliation as a mediator. Journal of Abnormal Child Psychology, 43(1), 81–93. https://doi.org/10.1007/s10802-014-9903-8

Bear

G. G.

Gaskins

Blank

Chen

(2011). Delaware school climate survey—student: Its factor structure, concurrent validity, and reliability. Journal of School Psychology, 49(2), 157–174. https://doi.org/10.1016/j.jsp.2011.01.001

Bear

G. G.

Yang

Harris

Mantz

Hearn

Boyer

(2016). Technical manual for 2016 Delaware school survey: Scales of school climate; bullying victimization; student engagement; positive, punitive, and social emotional learning techniques; and the Delaware social and emotional competencies scale. Newark, DE: Center for Disabilities Studies, 1–217.

Bronfenbrenner

Morris

P. A.

(2006). The bioecological model of human development. In Lerner

R. M.

Damon

(Eds.), Handbook of child psychology: Theoretical models of human development (pp. 793–828). John Wiley & Sons Inc.

Casper

D. M.

Card

N. A.

(2017). Overt and relational victimization: A meta–analytic review of their overlap and associations with social–psychological adjustment. Child Development, 88(2), 466–483. https://doi.org/10.1111/cdev.12621

Chan

H. C. O.

Wong

D. S.

(2015). Traditional school bullying and cyberbullying in Chinese societies: Prevalence and a review of the whole-school intervention approach. Aggression and Violent Behavior, 23(1), 98–108. https://doi.org/10.1016/j.avb.2015.05.010

Cheng

Newman

I. M.

Mbulo

Chai

Chen

Shell

D. F.

(2010). Being bullied and psychosocial adjustment among middle school students in China. Journal of School Health, 80(4), 193–199. https://doi.org/10.1111/j.1746-1561.2009.00486.x

Coie

J. D.

Dodge

K. A.

(1983). Continuities and changes in children's social status: A five-year longitudinal study. Merrill-Palmer Quarterly (1982–), 29(3), 261–282. https://www-jstor-org.web.bisu.edu.cn/stable/23086262

Eslea

Menesini

Morita

O'Moore

Mora-Merchán

J. A.

Pereira

(2004). Friendship and loneliness among bullies and victims: Data from seven countries. Aggressive Behavior, 30(1), 71–83. https://doi.org/10.1002/ab.20006

10.

Gini

Holt

Pozzoli

Marino

(2019). Victimization and somatic problems: The role of class victimization levels. Journal of School Health, 90(1), 39–46. https://doi.org/10.1111/josh.12844

11.

Gorman

A. H.

Schwartz

Nakamoto

Mayeux

(2011). Unpopularity and disliking among peers: Partially distinct dimensions of adolescents’ social experiences. Journal of Applied Developmental Psychology, 32(4), 208–217. https://doi.org/10.1016/j.appdev.2011.05.001

12.

Holfeld

Baitz

(2020). The mediating and moderating effects of social support and school climate on the association between cyber victimization and internalizing symptoms. Journal of Youth and Adolescence, 49(11), 2214–2228. https://doi.org/10.1007/s10964-020-01292-0

13.

Huang

Hong

J. S.

Espelage

D. L.

(2013). Understanding factors associated with bullying and peer victimization in Chinese schools within ecological contexts. Journal of Child and Family Studies, 22(7), 881–892. https://doi.org/10.1007/s10826-012-9647-4

14.

Ivarsson

Broberg

A. G.

Arvidsson

Gillberg

(2005). Bullying in adolescence: Psychiatric problems in victims and bullies as measured by the Youth Self Report (YSR) and the Depression Self-Rating Scale (DSRS). Nordic Journal of Psychiatry, 59(5), 365–373. https://doi.org/10.1080/08039480500227816

15.

Kaltiala-Heino

Rimpelä

Rantanen

Rimpelä

(2000). Bullying at school—an indicator of adolescents at risk for mental disorders. Journal of Adolescence, 23(6), 661–674. https://doi.org/10.1006/jado.2000.0351

16.

Kumpulainen

Räsänen

(2000). Children involved in bullying at elementary school age: Their psychiatric symptoms and deviance in adolescence: An epidemiological sample. Child Abuse and Neglect, 24(12), 1567–1577. https://doi.org/10.1016/S0145-2134(00)00210-6

17.

Leadbeater

B. J.

Sukhawathanakul

Thompson

Holfeld

(2015). Parent, child, and teacher reports of school climate as predictors of peer victimization, internalizing and externalizing in elementary school. School Mental Health, 7(4), 261–272. https://doi.org/10.1007/s12310-015-9154-z

18.

(2005). Mind or virtue: Western and Chinese beliefs about learning. Current Directions in Psychological Science, 14(4), 190–194. https://doi.org/10.1111/j.0963-7214.2005.00362.x

19.

Litwack

S. D.

Aikins

J. W.

Cillessen

A. H. N.

(2012). The distinct roles of sociometric and perceived popularity in friendship: Implications for adolescent depressive affect and self- esteem. The Journal of Early Adolescence, 32(2), 226–251. https://doi.org/10.1177/0272431610387142

20.

Long

Zhou

(2020). Relational victimization and internalizing problems: Moderation of popularity and mediation of popularity status insecurity. Journal of Youth and Adolescence, 49(3), 724–734. https://doi.org/10.1007/s10964-019-01177-x

21.

National School Climate Council (2007). The school climate challenge: Narrowing the gap between school climate research and school climate policy, practice guidelines and teacher education policy. https://schoolclimate.org/wp-content/uploads/2021/05/school-climate-challenge-web.pdf.

22.

Olenik-Shemesh

Heiman

Eden

(2012). Cyberbullying victimisation in adolescence: Relationships with loneliness and depressive mood. Emotional and Behavioural Difficulties, 17(3-4), 361–374. https://doi.org/10.1080/13632752.2012.704227

23.

Olweus

(2001). Peer harassment: A critical analysis and some important issues. In Juvonen

Graham

(Eds.), Peer harassment in school: The plight of the vulnerable and victimized (pp. 3–20). Guilford Press.

24.

Rose

M. G.

(1985). Back-translating to recover form. International Journal of Translation, 31(1), 6–11.

25.

Salmivalli

(2018). Peer victimization and adjustment in young adulthood: Commentary on the special section. Journal of Abnormal Child Psychology, 46(1), 67–72. https://doi.org/10.1007/s10802-017-0372-8

26.

Swearer

S. M.

(2001). The Bully Survey. Unpublished manuscript. Department of Educational Psychology, The University of Nebraska – Lincoln, Nebraska, United States.

27.

Swirsky

J. M.

Xie

(2021). Peer-related factors as moderators between overt and social victimization and adjustment outcomes in early adolescence. Journal of Youth and Adolescence, 50(2), 286–297. https://doi.org/10.1007/s10964-020-01313-y

28.

Waasdorp

T. E.

Pas

E. T.

O’Brennan

L. M.

Bradshaw

C. P.

(2011). A multilevel perspective on the climate of bullying: Discrepancies among students, school staff, and parents. Journal of School Violence, 10(2), 115–132. https://doi.org/10.1080/15388220.2010.539164

29.

Wang

Boyanton

Ross

A. S. M.

Liu

J. L.

Sullivan

Anh Do

(2018). School climate, victimization, and mental health outcomes among elementary school students in China. School Psychology International, 39(6), 587–605. https://doi.org/10.1177/0143034318805517

30.

Wang

La Salle

K. A.

Sullivan

K. E.

(2018). School climate and parental involvement buffer the risk of victimization on suicidal thoughts and behavior among Asian American middle school students. Asian American Journal of Psychology, 9(4), 296–307. https://doi.org/10.1037/aap0000138

31.

Wang

La Salle

Liu

J. L.

(2021). Do parental involvement and adult support matter for students’ suicidal thoughts and behavior in high school? School Psychology Review, 51(3), 1–14. https://doi.org/10.1080/2372966X.2021.1873058

32.

Wang

Liu

J. L.

Zhu

Yao

Boyanton

(2020). Covitality moderates the relationship between victimisation and loneliness. Australian Journal of Psychology, 73(1), 34–45. https://doi.org/10.1080/00049530.2021.1904494

33.

Werth

J. M.

Nickerson

A. B.

Aloe

A. M.

Swearer

S. M.

(2015). Bullying victimization and the social and emotional maladjustment of bystanders: A propensity score analysis. Journal of School Psychology, 53(4), 295–308. https://doi.org/10.1016/j.jsp.2015.05.004

34.

Xie

Bear

Yang

Marshall

Gong

(2016). Reliability and validity of the Chinese version of Delaware bullying victimization scale-student. Chinese Journal of Clinical Psychology, 23(4), 594–596. https://doi.org/10.16128/j.cnki.1005-3611.2021.04.010

35.

Yang

Bear

G. G.

Chen

F. F.

Zhang

Blank

J. C.

Huang

(2013). Students’ perceptions of school climate in the US and China. School Psychology Quarterly, 28(1), 7–24. https://doi.org/10.1037/spq0000002

36.

Yang

Sharkey

J. D.

Reed

L. A.

Chen

Dowdy

(2018). Bullying victimization and student engagement in elementary, middle, and high schools: Moderating role of school climate. School Psychology Quarterly, 33(1), 54–64. https://doi.org/10.1037/spq0000250

37.

Yin

Xue

Zhang

Liao

(2021). Analysis of Beijing’s population situation. In Yin

Xue

Dong

(Eds.), Blue book of Beijing population: Report on Beijing’s population development in 2021 (pp. 1–21). Social Sciences Academic Press(China).

38.

Zhao

(2005). Character education in China today. Social Studies and the Young Learner, 17(3), 11–14.

	Main effect model		Main effect model		Interaction model
	b(SE)	β		β	b(SE)	β
Constant	−1.71 (.06)		1.75(.05)***		1.75(.05)***
Gender (1 = boy, 0 = girl)	−0.09 (.07)	−.05	−0.18 (.06)**	−.10	−0.19 (.06)**	−.10
Grade level (0 = 7^th and 1 = 8^th)	−0.11 (.07)	−.06	−0.06(.06)	−.03	−0.06(.06)	−.03
School climate	−0.02 (.05)	−.01	.11(.04)**	.08**	.10(.04)*	.08*
Physical Victimization			.08(.07)	.07	.08(.07)	.07
Verbal Victimization			.13(.05) *	.13*	.12(.05) *	.12*
Relational Victimization			.44(.07)***	.37***	.41(.07)***	.35***
Cyber Victimization			.04(.07)	.03	.09(.07)	.06
School Climate*Physical Victimization			__	__	.04(.08)	.03
School Climate* Verbal Victimization			__	__	−.09(.07)	−.07
School Climate*Relational Victimization			__	__	.22(.09)*	.14*
School Climate*Cyber Victimization			__	__	−.05(.09)	−.03
F(df)	1.24 (3,683)		39.67 (7, 679)***		26.25 (11, 675)***
R²	.01		.29		.30
Δ R²			.28		.01