Physical inactivity is a worldwide public health problem, and it is a risk factor for several chronic diseases. Square dancing nightly may be an efficient way to promote physical activity among Chinese residents. This study aims to explore factors that affect resident participation in square dancing on the basis of social cognitive theory constructs (outcome expectations, outcome expectancies, self-efficacy, self-control, and environment) that may provide a scientific basis for designing interventions to promote physical activity in the future.Forty squares near neighborhood communities in Chongqing were randomly selected. A sample of 1732 residents who came to these squares at 18:00-21:00 were interviewed using a social cognitive theory questionnaire jointly developed by researchers from Chongqing Medical University and Jackson State University.Among 1732 respondents, 279 (16.1%) were male and 1457 (83.9%) were female. A total of 939 (54.2%) of the respondents were square dancers. The mean age of the dancers was 58.6 (SD = 9.1) years. Of the dancers, 825 (87.9%) danced at least 150 minutes every week, and 792 (84.2%) indicated that they had danced for more than 1 year. All the constructs of social cognitive theory were significantly different between residents whether they danced or not (P < .001 for all). Women (OR = 1.54, 95% CI: 1.03-2.31) were more likely to dance. Nonretired residents (OR = 0.53, 95% CI: 0.39-0.74) were less likely to dance. Residents with income of more than CNY 4000 per month were less likely to dance (OR = 0.44, 95% CI: 0.23-0.86). Residents with high self-efficacy (OR = 2.01, 95% CI: 1.81-2.22), and high self-control (OR = 1.26, 95% CI: 1.18-1.34) were more likely to dance.This study provides salient implications for developing interventions to promote square dancing by using social cognitive theory. Gender, retirement, income, and self-efficacy would be the factors influencing whether residents participate in square dancing.