Leisure-time physical activity participation and subjective class identity among Chinese residents: evidence from...

Subjective class identity is shaped not only by objective socioeconomic resources, but also by everyday experiences through which individuals evaluate their bodies, lives, and social relations. Although physical activity is often studied as a health behavior, leisure-time exercise, sports, and fitness may also function as socially meaningful lifestyle practices.

Using data from the 2021 China General Social Survey, this study examines the association between regular leisure-time physical activity and subjective class identity among Chinese residents, and further assesses whether self-rated health, subjective well-being, and social interaction help account for this association. The baseline analytical sample includes 5,688 respondents, while pathway analyses use 3,802 respondents with complete information on self-rated health, subjective well-being, social interaction, and all covariates.

Ordered logit models show that regular physical activity is positively associated with higher subjective class identity after adjustment for demographic and socioeconomic covariates, including gender, age, marital status, household registration, ethnicity, region, education, political status, and personal income (b = 0.231, robust SE = 0.052, p < 0.001; OR = 1.26). The association remains stable across alternative measurements of subjective class identity and physical activity, ordered probit models, generalized ordered logit diagnostics, survey-weighted models, province fixed effects, clustered standard errors, propensity-score weighting, entropy balancing, and income-coding sensitivity analyses. Pathway models and Karlson-Holm-Breen decomposition indicate that self-rated health, subjective well-being, and social interaction jointly account for 73.20% of the total association in the pathway sample, with subjective well-being contributing the largest share. Heterogeneity analysis shows broadly positive associations across social groups, with the clearest subgroup difference appearing by marital status.

Given the cross-sectional design, the findings should be interpreted as robust associations and statistical pathways rather than causal effects. The study contributes to social stratification and health sociology by showing that subjective class identity is connected to embodied health, affective evaluation, and relational experiences embedded in everyday lifestyle practices.