성인학습자의 무형식학습 참여 결정요인 분석: 랜덤 포레스트와 SHAP 적용

Abstract

The purpose of this study is to identify the key determinants that predict adult learners' participation in informal learning based on different types. The analysis was conducted on a sample of 9,949 adult learners extracted from the 2022 Korean Adult Lifelong Learning Survey. The types of informal learning were categorized into (a) traditional method-dependent learning and (b) online media-based learning. For the analysis, we applied explainable machine learning (XML) techniques, including random forest and SHapley Additive exPlanations (SHAP) algorithms. The results showed that the top determinants predicting participation in traditional method-dependent learning for adult learners were perceived lifelong learning effectiveness, job-related goal orientation, learning orientation, lifelong learning accessibility, and age. In contrast, for online media-based learning, the top determinants predicting adult learners' participation were age, job-related goal orientation, learning orientation, perceived lifelong learning effectiveness, and educational level. Lastly, while the relationships between each predictor and participation in informal learning were generally positive, a deeper interpretation is required depending on the type and measurement method of each variable. These findings suggest that adult learners' participation in informal learning can be considered a reflection of their demographic factors, as well as their perceptions and expectations regarding lifelong learning. Based on these results, we discussed implications for promoting lifelong learning participation among adult learners.