The strongest, clearest story in these data is not that social media is destroying GPAs — it is that procrastination is, and that social media addiction sits alongside procrastination as part of a broader self-regulation picture that is more tightly bound to anxiety than to grades.
Procrastination is the more direct predictor of GPA
APS-S scores correlated negatively with cumulative GPA (r = −.301, p = .005) and remained a significant predictor in the regression model. BSMAS, by contrast, was not significantly related to GPA at the bivariate level (r = −.092) and did not uniquely predict GPA in regressions. This pattern is consistent with the view that self-regulation failures — not screen time per se — drive academic costs (Rozgonjuk & Elhai, 2019; Shannon et al., 2022).
Social media addiction is more about anxiety than grades
BSMAS correlated positively with GAD-7 (r = .326, p = .003) and with APS-S (r = .453, p < .001). This aligns with prior work linking problematic phone and social media use to depression, anxiety, and distress (Demirci et al., 2015; Elhai et al., 2017; Keles et al., 2020). Mechanistically, this fits a picture in which addiction-like use, push notifications, and compulsive checking erode attention and emotional regulation (Kim et al., 2016) more reliably than they erode any specific exam score.
No significant sex interactions
Female students reported higher anxiety on the GAD-7 (M = 11.16 vs. 6.39), echoing general epidemiological patterns. But none of the BSMAS × sex or APS-S × sex interactions reached significance for GPA or anxiety. The associations between addiction-like social media use, procrastination, and outcomes were broadly similar across male and female students in this sample. This contrasts with Chen et al.'s (2017) finding of different predictors by gender, and may reflect the smaller sample, the U.S. undergraduate context, or the narrower range of predictors used here.
Limitations
- Self-report and recall bias. All measures, including GPA, were self-reported.
- Cross-sectional design. The data cannot establish whether procrastination causes lower GPA, or whether struggling academically encourages procrastination.
- Sample size and composition. N = 84 limits power to detect small interactions; the binary male/female grouping excluded other gender identities.
- No behavioral data. Actual screen time, app use, and notification volume were not measured.
Implications
For students and advisors, the practical implication is that interventions targeting procrastination and self-regulation may do more for GPA than blanket social media restrictions. For mental health, however, addressing addiction-like social media use is well-justified given its consistent link with anxiety in this and prior work. Future studies should pair self-report scales with passive sensing of phone use, span longer time windows, and include broader gender identities to test these patterns more rigorously.