Social anxiety as a mediating factor in predicting problematic Internet use among lonely adolescents

In recent years, quite a few research studies investigating the consequences of Internet use among adolescents have been conducted world-wide (Caplan, 2003; Mythily et. al., 2008; Van den Eijnden, Meerkerk, Vermulst, Spijkerman, & Engels, 2008).  However, findings from these studies indicated that it is not likely that all adolescents who use the Internet suffer from similar psychological adjustment problems. According to theories in social and personality psychology, the problematic use of the Internet by a person is influenced and explained by factors in the immediate environment and the subjective individual characteristics of the person concerned. This implies that behavior is largely a function of the person and the situation (Ross & Nisbett, 1991). In view of this, this study intends to examine how individual differences in particular, social anxiety, mediate between loneliness and the problematic use of the Internet among a group of Asian adolescents. A survey was administered to 681 adolescents ranging between 13 and 18 years of age (M= 14.6).

Initial findings indicated significant correlations among the three variables-loneliness, social anxiety and problematic Internet use. Subsequently, the data was analysed according to Baron and Kenny's (1986) and Holmbeck's (1997) conceptual and statistical recommendations for assessing the presence of mediator effects. Three equations were tested using multiple regression analyses with social anxiety hypothesized as the mediator. Results from the analyses first indicated that loneliness was significantly associated with problematic Internet use (B = 0.52, ß = 0.28, p < .001). Subsequently, the hypothesized mediator, social anxiety, was also regressed on loneliness and the association was significant (B = 0.67, ß = 0.53, p < .001). Finally, problematic Internet use was regressed simultaneously on both social anxiety and loneliness. The result showed that social anxiety was significantly associated with problematic Internet use controlling for loneliness (B = 0.59, ß = 0.41, p < .001).  The effect of loneliness was found to be no longer significant on pathological internet use when social anxiety was included as a mediator (B = 0.12, ß = 0.07, ns), indicating complete mediation.