An approach-Avoidance structural model of perceived social functioning in physical education.

PURPOSE: This study tested a cross-sectional approach-avoidance model integrating personal responsibility, interaction anxiety, fear of scrutiny, and perceived social functioning in university physical education (PE). METHODS: Participants were 215 students from six intact PE classes. Structural equation models were estimated using maximum likelihood and robust MLM with Satorra-Bentler correction because multivariate normality was violated. Model-based indirect associations were examined with 5000 bias-corrected bootstrap resamples, alongside measurement and sensitivity analyses. RESULTS: The primary model showed acceptable robust approximate fit (CFI = 0.977, TLI = 0.971, RMSEA = 0.069, 90% CI [0.054, 0.084], SRMR = 0.043), although the exact-fit null was rejected (Bollen-Stine p = .001). In the pooled primary model, personal responsibility was positively associated with social connectedness (β = 0.639, p < .001), fear of scrutiny was negatively associated with it (β = -0.461, p < .001), and interaction anxiety showed a positive conditional coefficient (β = 0.224, p = .005) despite a negative zero-order association (r = -0.260, p < .001), consistent with suppression. After covariate adjustment, personal responsibility remained positively associated with connectedness (β = 0.198, p = .002), whereas interaction anxiety (β = -0.025, p = .675) and fear of scrutiny (β = -0.121, p = .067) were not statistically significant. The three social dimensions showed weak discriminant validity and substantial overlap. CONCLUSION: The findings support a cautious, context-bounded structural account of perceived social functioning in university PE. The pooled upstream pattern was sensitive to course composition.