Digital literacy education and problematic Internet use among adolescents: a quasi-experimental study of digital self-efficacy and delay of gratification as cognitive–regulatory correlates

Problematic Internet use (PIU) has become an important concern in adolescent development, yet the psychological correlates through which educational programs may be associated with healthier Internet use remain insufficiently understood. This study examined whether a 6-week multi-component digital literacy and digital wellbeing program was associated with changes in adolescents’ PIU, digital self-efficacy (DSE), and delay of gratification (DGI), and whether the post-intervention data were statistically consistent with a theory-informed pathway linking DSE, DGI, and PIU. A two-arm pretest-posttest quasi-experimental design was conducted with 206 adolescents in Hubei Province, China. Participants were allocated to an intervention group ( n = 103) or a wait-list control group ( n = 103) using a quota-based procedure to improve comparability in age, gender, and daily Internet use time. The intervention group received the 6-week program, whereas the wait-list control group was offered the same program after posttest assessment. PIU, DSE, and DGI were measured at pretest and posttest. Data were analyzed using repeated-measures ANOVA and serial mediation analysis with demographic covariates and baseline scores controlled. Significant Group × Time interactions were observed for PIU, DSE, and DGI, indicating more favorable self-reported pre-to-post changes in the intervention group than in the wait-list control group. The post-intervention data were statistically consistent with the hypothesized DSE → DGI → PIU indirect pathway; however, alternative-model analyses indicated that DGI showed the more consistent association with PIU across specifications. These findings provide preliminary quasi-experimental evidence that participation in a multi-component digital literacy and digital wellbeing program is associated with healthier self-reported Internet use, stronger perceived digital competence, and greater delay of gratification. Because participants were not randomly assigned and the mediators and outcome were measured concurrently at posttest, the indirect associations should be regarded as hypothesis-generating rather than as evidence of a causal or temporal sequence.