Development, implementation, and evaluation of a belief-targeted non-face-to-face leisure-time physical activity promotion program for South Korean University students

Abstract Background Despite a sharp decline in physical activity levels among university students since the COVID-19 pandemic, physical activity programs for this population remain limited. Our previous research examined key beliefs influencing physical activity behavior among university students at a Korean university during the COVID-19 pandemic. Accordingly, this study aimed to develop, implement, and evaluate a non-face-to-face leisure-time physical activity (NFTF-LTPA) promotion program for university students, focusing on three beliefs—health, friends, and laziness—identified as key determinants of physical activity behavior in our previous study. Methods Undergraduate students were newly recruited from the same university and departments (i.e., the College of Social Sciences and the School of Business) as our prior study. Participants were stratified and randomly assigned to intervention or control groups and a belief-targeted intervention was implemented for three months. Process evaluations were conducted at mid-, and post-intervention, and outcomes were assessed at pre-, mid-, and post-intervention. A one-month follow-up survey was also conducted after the intervention ended. A 2 × 4 repeated measures generalized linear model was conducted and subgroup analyses examined participants with high intervention adherence (PHIA) compared with controls. Results Among the total analytic sample, significant group × time interaction effects at the α = 0.10 level were observed for outcome evaluation of behavioral beliefs ( F = 2.70; p = 0.072; $$\:{\eta\:}^{2}\:$$ = 0.05) and motivation to comply with normative beliefs ( F = 2.54; p = 0.085; $$\:{\eta\:}^{2}\:$$ = 0.05). However, no significant group × time interaction effects were observed for control beliefs, intention, or LTPA participation. In an exploratory subgroup analysis, a significant group × time interaction effect was observed for the control belief of laziness ( F = 3.85; p = 0.031; $$\:{\eta\:}^{2}$$ = 0.18). Post hoc analyses indicated lower perceived laziness at the program midpoint in the PHIA group compared with the control group. Conclusions Although the intervention produced changes in selected belief components among the total analytic sample, these changes did not translate into significant improvements in intention or behavior. Exploratory subgroup analyses suggested potential changes in control beliefs among participants with higher intervention adherence. These findings provide preliminary insights into the development, implementation, and evaluation of belief-targeted NFTF-LTPA promotion programs for university students. Trial registration The trial was retrospectively registered at ClinicalTrials.gov (Identifier: NCT07423403; 19 February 2026).