School-based relationships and problematic internet use amongst Chinese students

We know that adolescents across industrialized countries report spending a substantial quantity of daily time on the internet. Evaluations of ‘screen time’ indicate that Chinese adolescents spend 7.56 hours/day on viewing screens, American adolescents spend 7.22 hours/day and Australians adolescents spend 6.23 hours/day [1–3]. While the use of the internet by adolescents can be beneficial, the ‘problematic’ use of the internet has also been shown to affect academic performance [4], subjective wellbeing [5], anxiety [6], depressive symptoms [7], sleep disorders [8], reduced leisure activities [9], obesity [10], and suicidality [4].

Although there is no consensus about the definition of problematic internet use, a common definition is that PIU involves a strong urge to use the internet to the extent that it leads to significant social isolation, health issues and academic difficulty [11–13]. Recent research across six Asian countries (China, Hong Kong, Japan, South Korea, Malaysia and the Philippines) found the between 12.5% and 46.0% of 12–18 years old reported PIU [14]. In research using the Adolescent Pathological Internet Use Scale (APIUS) with Chinese adolescents [13], we previously found that the prevalence rate of PIU was 9.0% in Longhua district, Shenzhen [15]. These rates are similar to those found in past research with Chinese adolescents [16, 17]. Given the high prevalence rates of PIU among Chinese adolescents, it is important to identify risk and protective factors for PIU among Chinese adolescents to guide future longitudinal and interventional studies in China and similar cultures.

School-based relationships are likely to be a key domain of potential risk and protective factors for adolescent PIU. Based on social learning principles, adolescents may learn to use the internet to avoid stressful face-to-face interactions with peers, teachers, or other staff in school settings. For example, Italian high school students experiencing interpersonal difficulties report a preference for avoidant coping strategies as well as PIU [18]. Some emerging evidence from China suggests that negative school-based relationships are associated with PIU. For example, poor relationships with peers [16], and teachers, have predicted high levels of PIU in Chinese adolescents [19].

While these studies indicate that negative school-based relationships are risk factors for PIU, relationship quality is a multidimensional construct, and we know little about the dimensions of relationship quality that are related to adolescent PIU [20]. Further research is therefore needed on what aspects or characteristics of school-based relationships are associated with greatest risk of PIU. Given that interpersonal conflict has previously been linked with poor mental health in Chinese adolescents [21], this study aimed to examine the role of school-based interpersonal conflict upon PIU. To date, no study has specifically measured the association between school-based interpersonal conflict and PIU, however there is some emerging evidence that supports such a link. For example, a recent study conducted in China among 2666 middle school students found a significant relationship between experiences with verbal abuse from teachers and peer/online bullying and internet gaming disorder [22]. While indicative, this study was limited to measurement of one form of negative school-based relationship and measured a very specific form of PIU. Hence there is a need to explore whether similar relationships exist between a broader range of conflictual school-based relationships and more general experiences of PIU in Chinese adolescents.

A second reason that school-based relationships may be associated with PIU is derived from attachment theory [23]. This theory emphasizes the role of emotional connectedness and perceptions of safe/secure relationships with significant others as determinants of mental health outcomes [24]. In the school-based empirical literature, different terms have been used to capture emotional connectedness with others, such as school connectedness, perceived school climate, perceived school satisfaction, and school bonding. There is strong evidence that adolescents’ school connectedness is inversely associated with risky behaviors such as sexual risk taking [25], substance use [26], and antisocial behaviors [27]. There is also evidence that elevated school connectedness is positively associated with mental health [28] and a sense of well-being in adulthood [29]. Based on a sample of 2758 middle school students in southern China, Li et al. showed that in adolescents with lower school connectedness, higher levels of PIU were reported [30]. Likewise, classroom environment and climate, for example, perceptions of teacher receptivity, support and inclusiveness, is likely to be protective of emotional and behavioral problems in Chinese adolescents [31]. Very little previous research has tested the association of school connectedness and classroom atmosphere with adolescent PIU.

The focus of this study therefore was to evaluate whether school connectedness, classroom atmosphere as well as peer and teacher conflict are associated with PIU in a large Chinese adolescent sample. We were also interested in examining how these associations varied with age and gender. As adolescents move through puberty, peer networks become increasingly important to the adolescent [27] and adolescents’ sensitivity to conflict with peers increases [32]. In addition, older adolescents may become more vulnerable to anxiety and depressed mood compared with younger adolescents [33]. We therefore anticipated that school connectedness and interpersonal conflict would be more closely linked to PIU for older compared to younger adolescents. In terms of gender, previous findings show that Chinese adolescent females are subject to higher risk of depression associated with higher interpersonal conflict and lower school connectedness compared to Chinese adolescent boys [21]. Similarly, Stavropoulos and colleagues reported that being in a hostile classroom was associated with female’ PIU but not male PIU in adolescents [34]. Given these gender differences, it is likely that the links between both positive and negative school-based relationships and PIU may be stronger among adolescent girls compared to boys. It was hence also decided to also examine the role of gender and grade level moderators of associations between school-based relationships and PIU.

In summary, we aimed to test the associations between school-based conflictual relationships, school connectedness, and classroom atmosphere and PIU, and to examine how these effects varied by grade level and gender. The hypotheses were that:

The present study examined the associations between school-based relationships and PIU in a large sample of adolescents in China. The hypothesis of higher adolescent-peer conflict and adolescent-teacher conflict being associated with higher PIU was confirmed. That is, higher levels of adolescent-peer conflict and higher levels of adolescent-teacher conflict were associate with higher levels PIU. The hypothesis of higher school connectedness and classroom atmosphere being associated with lower PIU was also confirmed. Finally, the hypothesis that gender and grade level would moderate these relationships was partly confirmed. That is, the associations between conflict with peers and PIU was stronger in older adolescents compared with younger adolescents. However, there was no evidence of gender or grade level moderating the other associations identified.

Previous studies have looked at the associations between both negative peer-adolescent and teacher-adolescent relationships and PIU [16, 19]. However, prior studies used nonspecific definitions of peer and teacher relationship quality, leaving it unclear as to what aspects of relationship quality might be a potential prevention or intervention target. The present study highlights the role of conflictual relationships with peers and teachers as potential contextual factors for PIU. Given that Chinese adolescents spend a large part of their time in school, and academic performance is a major concern for them, it is understandable that school-based conflict with peers and teachers may be potent unique risk factors for PIU. Further, given that Chinese teachers are held in high esteem as part of Chinese Confucian culture [42], the negative impact for students who experience conflict with their teachers may be marked relative to other cultures. Future studies comparing the relative effects for Chinese and Western cultures are needed.

Our finding that both higher school connectedness and classroom atmosphere were related to lower PIU in Chinese adolescents is also novel. This finding is important in that it highlights that the perceived positive relationships within a classroom may add independent protective value to PIU in addition to broader school connectedness. As such both variables are important to consider when examining the associations between positive school-based relationships and PIU. However, this relationship needs to be further explored in other cultures. For example, a study by Jia et al. found that Chinese student’s perceptions of school climate were higher than their American counterparts. Given that there may be cultural differences in perceived school connectedness, our findings need to be replicated in other cultures [43].

Finally, to the best of our knowledge, no previous study has investigated the moderating effect of gender and grade level upon the associations between school-based relationships and PIU amongst adolescents. Contrary to Hypothesis 3, gender did not moderate any of the associations between the school-based relationships and PIU. However, we did find that grade moderated the association between conflict with peers and PIU, with this association being stronger in older students than younger students. One possible reason for the grade moderation effect in the present study may be that peer relationships carry more substantive meaning as they grow older [44], and conflictual peer relationships may therefore engender coping responses like PIU in older students, which in turn may serve to reduce the possibility of solving interpersonal conflict. Therefore, further research is needed to examine the reasons for this moderation effect between conflictual relationships with peers and PIU among Chinese adolescents.

One possible theoretical mechanism that may help explain our results is the use of avoidance as a coping strategy by adolescents. As shown in a cross-sectional study among Italian students, those who had interpersonal difficulties reported more avoidant coping, and in turn experienced more PIU [18]. In addition, lonely adolescents tend to prefer online communication and are more likely to use the internet as a coping strategy for stressful relationships leading to more PIU [45]. Similarly, recent research has shown that negative teacher-student relationships have an indirect effect on problematic mobile phone use through loneliness and motivation to escape negative emotions [46]. Conversely, it is likely that PIU may also have detrimental effects upon interpersonal relationships. For example, there is evidence that PIU can aggravate conflict between family members [47]. These authors suggest that a potential mechanism for such conflict may be that adolescent’s excessive internet use can lead to less face to face interaction with family, and less time studying, which then leads to increased anger among parents. These relationships are likely to be bidirectional as the increase in parental anger may result in more PIU as avoidance coping by the adolescent to escape dealing with distressed parents. Such bidirectional and cyclical school-based relationships need to be further examined in future studies.

There are a number of practical implications from our findings. Teachers, policy makers, psychologists, and public health workers should be cognisant of the interpersonal conflictual relationships with peers and teachers being potential risk factors for adolescent PIU, while positive and supportive classroom and school-based relations appear to be potential protective factors for adolescent PIU. This might involve training both adolescents and teachers in conflict resolution skills, as well as in prosocial interpersonal skills. In addition, teachers could be trained in creating a safe and inclusive classroom environment aligned with school’s values. While, there is some emerging evidence that family based therapy, aimed at improving relationships between adolescents and parents, may help reduce PIU in teenagers [48], there is a need to examine improving school relationships, to reduce PIU. Previous school-based programs have focused on either improving psychosocial competencies, e.g. problem solving, emotional regulation skills, or reducing other health behaviours, e.g. alcohol use, physical activity to reduce PIU [49].

Key areas for further research include investigating the relative role of family-based relationships as independent risk and protective factors for adolescent PIU in addition to school-based relationships. Second, there is a need to clarify the psychological mechanisms linking school-based relationships with PIU. For example, a future study could also include measures of loneliness, depression or anxiety, and avoidant coping strategies in order to examine the degree to which these variables mediated the associations identified in this study. Finally, future research should further explore other potential moderators in addition to gender and grade. For example, a Korean study found that higher school performance was positively associated with longer internet use for study compared to for general use. It suggested that the purpose of internet use could be potential moderating effect between the associations in this study.

This study has many strengths such as it is based on a large sample size and has a high response rate (99.2%) which results in limited bias associated with pathways to participation [50]. Moreover, the tested relationships between variables were controlled for using covariates that have previously been shown to be strong predictors of PIU, such as gender, grade family structure, parental education, and academic ranking [51]. Also, the study explored a diverse range of school-based relationships (positive and negative relationships) and PIU among Chinese adolescents. However, the study also had limitations that are important to be known. First, given the cross-sectional design of the study, conclusions about causal relationships are not possible. Future longitudinal studies are warranted to clarify the direction of these associations. Secondly, caution needs to be exercised when generalizing these findings to all of China as this survey was only conducted in one district in Shenzhen. Moreover, the self-reported data is purely reflective of the adolescents’ perspective. The strength of the validity of these findings could be improved by using multiple informants (e.g. parents, teachers, and peers) and multiple methods (e.g. interview, and observation). It is also worth noting that the findings of this study were limited to general problematic use of the internet rather than specific categories of problematic use (e.g. gaming addiction, smartphone addiction, social media addiction etc). In addition, the sample only included primary and secondary school students rather than vocational school students who might be more vulnerable to PIU [52].

This study revealed that conflict with peers and teachers were positively associated with PIU, while school connectedness and perceived classroom atmosphere were negatively associated with PIU in Chinese adolescents. The effect of conflict with peers on PIU was stronger in older than younger students. These findings point to the importance of effective conflict management, especially for older students, and the promotion of school connectedness and perceived classroom atmosphere to decrease PIU among Chinese adolescents.

The School of Public Health in Sun Yat-sen University undertook this study. We would like to thank Chuan-An Wu, Xiao-Na Yin, Guo-Ming Wen, Deng-Li Sun, Dan-Xia Xian, and other staff in Longhua Maternal and Child Health Care Center who offered much assistance in the field survey. We are also grateful to all students who participated in this research. This manuscript is based on PhD research conducted by the first author under the supervision of ES, WQC, and AK.

All relevant data are within the paper and itsfiles. Supporting Information

This study was supported in the form of funding by the Government of Longhua District, Shenzhen, China awarded to MH.