Understanding social media addiction among adult athletes: psychological correlates in a cross-sectional study

Abstract Background Limited research has examined whether established psychological correlates of problematic social media use retain unique associations when considered together among adult athletes. This study examined the bivariate and adjusted associations of self-handicapping, psychological resilience, sportsmanship, anxiety, stress, and depression with self-reported addiction-related symptoms of social media use, with sportsmanship included as an exploratory extension. Methods A cross-sectional correlational design was employed with a convenience sample of 388 licensed adult professional athletes with professional sports contracts (183 females, 205 males; aged 18–59 years) recruited from sports clubs in Rize province and the surrounding area, Türkiye. Participants completed the Social Media Addiction Scale, Self-Handicapping Scale, Adult Resilience Measure (ARM), Sportsmanship Behavior Scale, and Depression Anxiety Stress Scale-21 (DASS-21). Scale scores were interpreted as continuous measures rather than clinical diagnoses. Pearson correlations and simultaneous six-predictor multiple linear regression were conducted. HC3 heteroskedasticity-robust standard errors were used in a sensitivity analysis. Results Addiction-related symptom scores were positively correlated with self-handicapping, anxiety, stress, and depression, and negatively correlated with psychological resilience and sportsmanship (all ps < 0.001). The regression model accounted for 48.7% of the variance within the present sample (adjusted R² = 0.479), F(6, 381) = 60.20, p < .001. Under conventional ordinary least squares inference, self-handicapping (β = 0.32, p < .001) and stress (β = 0.26, p < .001) retained significant positive associations, whereas psychological resilience (β = −0.23, p < .001) and sportsmanship (β = −0.14, p = .001) retained significant negative associations. Anxiety and depression did not retain significant adjusted associations. HC3 robust inference preserved this significance pattern, although diagnostic findings indicated potential functional-form misspecification. Conclusions The findings provide contextual evidence on psychological correlates of self-reported addiction-related symptoms among adult professional athletes. They may inform assessment and future research, but do not establish causal relationships, individual predictive performance, or effective intervention targets. Whether modifying these characteristics reduces problematic social media use requires experimental evaluation.