Depressive symptoms have a strong effect on appetite decline in older adults. There is also an association between social activity and decreased depressive symptoms, but the mechanism between appetite and social activity with respect to depressive symptoms is unclear. This cross-sectional study examined the direct and indirect effects of social activity on appetite, via depressive symptoms. A total of 259 community-dwelling older adults (mean age 75.1 ± 5.3 years; 55% female) completed the Council on Nutrition Appetite Questionnaire (CNAQ) and the 15-item version of the Geriatric Depression Scale (GDS). We used a cutoff score of 6 on the GDS-15 to determine whether participants displayed depressive symptoms. Social activity was assessed based on the number of activities participated in at least 1-3 times per month. The types of social activity consisted of local community, hobby, sports, citizen, industry, religion, volunteer, and others. Other assessed factors included age, sex, education, financial status, living situation, and comorbidities. To examine the relationships among social activity, depressive symptoms, and appetite, structural equation modeling (SEM) was used, adjusting for variables associated with depressive symptoms or appetite in multivariate analyses. SEM revealed that participation in more types of social activity had a significant indirect effect on higher appetite score via less having depressive symptoms (β = 0.04, p = 0.018). Our findings suggest that active participation in social activities may have a positive effect on good appetite via not having depressive symptoms among older adults. Further longitudinal or intervention studies are needed to confirm our findings.