Exploring Predictors of Older Adults’ Intention to Participate in Age-Friendly Campus Programs Using Random Forest
Korean Council on Social Welfare Education · 30.06.2026 · 73 (2) · 67-94
DOI: 10.31409/kjswe.2026.73.2.67 ↗Kurz gefasst
Fachartikel in der Zeitschrift Korean Council on Social Welfare Education, online erschienen am 30.06.2026. Verfasst von Ra-Ra Nam und 1 weiteren Autor:innen.
Themen: Gerontologische Pflege
Abstract (Originalsprache)
This study aimed to explore the predictors influencing older adults’ intention to participate in Age-Friendly Campus programs. To this end, survey data were collected from 230 older adults residing in rural areas and small-sized cities. A total of 21 variables, including sociodemographic characteristics, psychosocial characteristics, and digital literacy, were included in the analysis. The analysis was conducted in two stages. In the first stage, Random Forest analysis was used to examine the relative importance of the variables. In the second stage, hierarchical logistic regression analysis was conducted to test the statistical significance of individual variables. The results showed that the 5-fold cross-validation AUC was .744, and the classification accuracy of the final model was 71.7%. In the Random Forest variable importance analysis, digital literacy showed the highest contribution, followed by social participation, age, life satisfaction, and self-mastery. In contrast, the hierarchical logistic regression analysis identified age, residential area, and disability status as statistically significant predictors. Age consistently showed a negative association across all models, and older adults residing in small-sized cities showed higher odds of participation intention than those residing in rural areas. Digital literacy showed the highest variable importance in the Random Forest analysis but was not statistically significant in the logistic regression analysis. This result can be interpreted as reflecting the methodological difference between the two approaches: Random Forest is well suited to capturing nonlinear branching structures, whereas logistic regression tests the independent linear effect of each variable after controlling for other variables. This study confirmed that age, residential area, and disability status are major statistical predictors of older adults’ intention to participate in Age-Friendly Campus programs, while digital literacy and social participation may serve as important exploratory predictive cues. These findings suggest that local universities should consider rural older adults’ accessibility, age-specific participation strategies, disabilityfriendly learning environments, and digital literacy support when designing Age-Friendly Campus programs.
Zitieren
APA
Nam, R., & Lim, J. (2026). Exploring Predictors of Older Adults’ Intention to Participate in Age-Friendly Campus Programs Using Random Forest. Korean Council on Social Welfare Education, 73(2), 67-94. https://doi.org/10.31409/kjswe.2026.73.2.67
Vancouver
Nam R, Lim J. Exploring Predictors of Older Adults’ Intention to Participate in Age-Friendly Campus Programs Using Random Forest. Korean Council on Social Welfare Education. 2026;73(2):67-94. doi:10.31409/kjswe.2026.73.2.67
BibTeX
@article{Nam2026,
author = {Nam, Ra-Ra and Lim, Jin-Seop},
title = {Exploring Predictors of Older Adults’ Intention to Participate in Age-Friendly Campus Programs Using Random Forest},
journal = {Korean Council on Social Welfare Education},
year = {2026},
volume = {73},
number = {2},
pages = {67--94},
doi = {10.31409/kjswe.2026.73.2.67}
}
EUVIMED