This study explored the cognitive structures that influence mobile information seekers’ intentions of promoting health behaviors. Using a nationwide self-reported survey (n = 1010) conducted by a national research institute in Korea, the present study conducted multiple regression analyses and moderated mediation tests for its analysis. The results showed that two of the three social cognitive components—i.e., health information orientation (HIO) and e-health literacy (EHL)—had a positive effect on users’ mobile health information seeking behaviors (MHIS), whereas informational social support (ISS) did not. Furthermore, the effects of the social cognition factors varied based on the type of health-promoting behaviors, and the types of health-related behavior predicted by the three factors also differed. Moreover, HIO was a factor that affected only nutritional intake among all of the health promotion behaviors, while ISS was a critical factor that predicted most health-related behaviors, except for the participants’ regular exercise (REX). The findings also indicated that the respondents’ levels of health efficacy and the extent of their health-related information seeking on mobile devices partially mediated the associations between the independent variables and their health-related behavioral outcomes. Among the mediated moderation effects of the three factors in social cognition theory (SCT), only the mediated effects of EHL were found to be significant. Based on these findings, the present study highlighted that the predictors of health-promoting behaviors may vary according to the features of the information seekers. The further implications of the study are also discussed.
1 January 2021 in American Journal of Health Behavior
Objectives: In this study, we aimed to find the influential factors in determining individuals' use and non-use of fitness and diet apps on smartphones. To this end, we focused on diverse groups of predictors that would significantly affect people's use and non-use of these apps. Methods: Overall, we considered 105 factors as potential predictors and included them in further analyses using a machine learning algorithm, XGBoost. The main reason for selecting this particular algorithm was that it had been known as one of the most accurate and popular algorithms for predicting consumer behaviors. Results: We found the accuracy score of those factors for predicting people's use and non-use of fitness and diet apps was approximately 71.3%. In particular, the most influential predictors were mainly related to social influence, media use, overeating, social support, health management, and attitudes toward exercise. Conclusion: These findings contribute to helping scholars and practitioners to develop more practical strategies of the implementation of fitness and diet apps.
31 January 2025 in Journal of Digital Contents Society