The association between physical activity and sleep quality among adolescents: the serial mediating roles of psychological resilience and academic stress and the moderating role of peer support.

Insufficient sleep and poor sleep quality among adolescents have become major public health concerns worldwide. Physical activity is widely associated with sleep quality, but the psychosocial factors related to this association require further examination. This cross-sectional study examined whether physical activity was indirectly associated with adolescents' sleep quality, as indexed by Pittsburgh Sleep Quality Index (PSQI) scores, through psychological resilience and academic stress, and whether peer support moderated the association between academic stress and PSQI scores. A total of 1,347 high school students from four districts in Shanghai completed paper-and-pencil questionnaires. Physical activity, sleep quality, psychological resilience, academic stress, and peer support were measured using the Physical Activity Questionnaire for Adolescents, the Pittsburgh Sleep Quality Index, the Resilience Scale for Chinese Adolescents, the Academic Stress Questionnaire for Middle School Students, and the Peer Support subscale of the Perceived Social Support Scale, respectively. Pearson's correlation analysis, hierarchical regression analysis, serial indirect-association analysis, and moderation analysis were conducted using SPSS 26.0 and PROCESS 4.2, with gender, grade level, household registration type, and only-child status controlled for. Higher PSQI scores indicate poorer sleep quality. Physical activity was negatively associated with PSQI scores and academic stress and positively associated with psychological resilience. Psychological resilience was negatively associated with PSQI scores and academic stress, whereas academic stress was positively associated with PSQI scores. Physical activity showed a significant direct association with PSQI scores (β = -0.202, 61.3%) and significant indirect associations through psychological resilience (β = -0.053, 16.0%), academic stress (β = -0.040, 12.1%), and the serial sequence of psychological resilience and academic stress (β = -0.035, 10.6%). Peer support moderated the association between academic stress and PSQI scores, such that this positive association was weaker among adolescents with higher peer support. These findings suggest that physical activity, psychological resilience, academic stress, peer support, and sleep quality are closely interrelated among adolescents. Given the cross-sectional design, the findings should be interpreted as statistical associations rather than evidence of temporal or causal mechanisms.