Physical activity and marital satisfaction among teachers: a mediation model of self-esteem, spirituality, and psychological distress

Background/objectives Marital satisfaction is influenced by various intrapersonal and environmental factors. The present study explores the associations between physical activity and marital satisfaction in a sample of Tunisian married teachers, examining the mediating roles of body esteem, sexual self-esteem, and psychological distress, with spirituality acting as a psychosocial resource. Methods A cross-sectional study was conducted with 924 married teachers (52% women) from six Tunisian regions. Participants completed the Arabic versions of the Ironson-Woods Spirituality/Religiousness Index (IWSRI), Body Esteem Scale (BES), Sexual Self-Esteem Scale (SSES), Kessler Psychological Distress Scale (K10), and the Dyadic Adjustment Scale (DAS-4). The hypothesized pathways were examined using Structural Equation Modeling (SEM). Results The Arabic IWSRI demonstrated strong psychometric properties ( α = 0.961) with a confirmed four-factor structure. SEM revealed a significant positive association between physical activity and marital satisfaction ( β = 0.34, p < 0.001) which was partially mediated by body esteem ( R 2 = 0.48), sexual self-esteem ( R 2 = 0.52), and reduced psychological distress ( R 2 = 0.24). Spirituality was positively correlated with both physical activity ( β = 0.41, p < .001) and marital satisfaction ( β = 0.29, p < .001). The overall model accounted for 63% of the variance in marital satisfaction. Conclusions In this specific population of Tunisian teachers, physical activity is associated with higher marital satisfaction through both physiological and psychosocial dimensions. Spirituality appears to be a significant resource correlated with mental balance and resilience. These preliminary results support an integrative view of health—encompassing physical, psychological, and spiritual facets—for understanding factors linked to stable marital relationships. However, due to the cross-sectional design, causal directions cannot be inferred.