Methods: Based on a multistage random cluster sampling method, a total of 2001 adolescents aged 12-18 years (51.3% boys and 48.7% girls) was obtained from Hebei Province, China. Their mean age was 16.07 years (SD=1.38). In a cross-sectional research design, participants were invited to complete the physical abuse and physical neglect subscales in the Chinese version of the Childhood Trauma Questionnaire, anxious attachment and avoidant attachment subscales in the Adolescent-Parent Attachment Inventory (APAI), and the Ten-Item Internet Gaming Disorder Test. We performed descriptive statistics and bivariate correlation analysis in SPSS 26.0 and tested our hypotheses by using structural equation modeling (SEM) and multi-group analysis in SEM in AMOS 26.0.
Results: The results showed that physical abuse was directly associated with online gaming disorder (β=.107, p<.001). Only anxious attachment mediated the relationship between physical abuse and online gaming disorder (β=.053, bootstrapping 95% CI [.045, .098], p<.001), and between physical neglect and online gaming disorder (β=.029, bootstrapping 95% CI[.015, .038], p<.001). Specifically, both physical abuse (β=.309, p<.001) and physical neglect (β=.172, p<.001) were positively associated with anxious attachment, whereas physical neglect (β=.151, p<.001) was positively associated with avoidant attachment. However, the relationship between avoidant attachment and online gaming disorder was non-significant (β=.035, p=.129) in our sample while anxious attachment (β=.171, p<.001) positively predicted online gaming disorder of adolescents. The results of the multi-group analysis suggested the significant difference in the pathway coefficients between the two groups (Δχ2/(Δdf) =16.579/8, p<.05). Besides, significant difference was observed in the specific structural path between physical neglect and anxious attachment (CRD = −2.851, p<.05), with the effect being stronger for adolescents from single-child families (β=.279, p<.001) compared with those from non-single child families (β=.137, p<.001) in this relationship.
Conclusions and Implications: The findings suggest the linking mechanism of insecure attachment in the relationship between physical maltreatment and online gaming disorder. Also, the results of the multigroup analysis in SEM remind future practitioners to carefully evaluate the family conditions, especially the conditions of siblings when intervening in online gaming disorder of adolescents.