This preprint examined the relationship between screen time and depressive symptoms in a large sample of Chinese college students. Using self-report questionnaire data from 10,999 participants, the authors report a statistically significant positive association between screen time and depressive symptom scores (r = 0.16, p < 0.01). Regression analysis indicated that screen time significantly predicted higher levels of depressive symptoms (β = 0.712, p < 0.001).
The authors emphasize that extended screen exposure was associated with greater depressive symptomatology in this cross-sectional sample. As a cross-sectional design, these data demonstrate associations but do not on their own establish causation.
Correlation analyses reported in the study show several relationships among the variables measured. Screen time correlated positively with worse sleep quality (r = 0.15, p < 0.01) and negatively with emotion regulation ability (r = −0.13, p < 0.01). Sleep quality showed a positive correlation with depressive symptoms (r = 0.31, p < 0.01), while emotion regulation showed a negative correlation with depressive symptoms (r = −0.42, p < 0.01).
In regression models, screen time significantly predicted:
Additionally, sleep quality and emotion regulation independently predicted depressive symptoms in the regression models: sleep quality positively predicted depressive symptoms (β = 1.318, p < 0.001) and emotion regulation negatively predicted depressive symptoms (β = −0.424, p < 0.001).
These reported effect sizes and directions are taken directly from the source preprint and indicate consistent associations between higher screen exposure, poorer sleep, compromised emotion regulation, and elevated depressive symptoms.
To examine potential mechanisms linking screen time with depressive symptoms, the study tested a serial mediation model using the PROCESS macro (Model 6) with bootstrapping to estimate indirect effects. The mediation analysis reported three key indirect pathways:
These findings, as reported, suggest both independent and sequential mediating roles for sleep quality and emotion regulation in the association between screen time and depressive symptom levels in this sample.
The study used a stratified cluster sampling method to recruit 10,999 college students for a cross-sectional questionnaire survey. Data collected included self-reported screen time, sleep quality, emotion regulation ability, and depressive symptoms. Statistical procedures included descriptive statistics, correlation analyses, and multiple regression using SPSS 26.0. The serial mediation model was tested with the PROCESS macro (Model 6) and bootstrapping to evaluate indirect effects.
The source reports ethical oversight: the study protocol received approval from the ethics committee at Nantong Municipal Health Commission (approval number 2024(1)), and participant consent procedures were confirmed. The authors declared no competing interests and acknowledged funding from the Postgraduate Research & Practice Innovation Program of Jiangsu Province.
The source does not provide further detail in the abstract about the specific instruments used to measure sleep quality, emotion regulation, or depressive symptoms, nor does it report demographic breakdowns, effect size interpretation beyond the coefficients and correlations cited, or sensitivity analyses in the portion of the text provided.
The authors conclude that higher screen time is significantly associated with greater depressive symptoms among college students, and that this relationship is partially explained by poorer sleep quality and reduced emotion regulation ability. The serial mediation result reported suggests a pathway where prolonged screen exposure may impair sleep, which then weakens emotion regulation capacity, contributing to higher depressive symptoms.
Because the report is a preprint, the authors note that the findings have not been peer-reviewed and should not by themselves be used to guide clinical practice. The large sample size strengthens statistical power for detecting associations, and the mediation analyses provide hypotheses about psychological mechanisms that could be tested in longitudinal or experimental designs.
This work is reported as a preprint on medRxiv and has not undergone peer review. The study design is cross-sectional; therefore, directionality and causality cannot be established from the analyses presented. The abstract does not specify measurement instruments, timing of data collection beyond the posting date, or detailed participant demographics, so those details were not reported in the source material provided here.
Ethical approval and informed consent were reported, and the authors declared no competing interests. Further peer review and additional methodological detail would be required to evaluate generalizability, measurement validity, and causal interpretation of the proposed mediation pathways.