영농교육이 귀농인의 경제적 성과에 미치는 영향: 로짓과 GBM을 활용한 성향점수매칭 분석

The Effects of Agricultural Training on the Economic Performance of Returning Farmers: A Propensity Score Matching Analysis Using Logit and GBM

초록

This study examines the effect of agricultural training on the economic performance of returning farmers in South Korea using Propensity Score Matching(PSM). The analysis is based on data from the 2024 Survey of Returning Farmers and Rural Migrants conducted in Changnyeong-gun and Gokseong-gun, which include 471 returning farmers. To mitigate selection bias arising from self-selection into agricultural training, propensity scores were estimated using both logistic regression(Logit) and a machine learning–based Gradient Boosting Machine(GBM). Outcome variables include subjective changes in economic living conditions, monthly household income after returning to farming, and changes in household income before and after returning to farming. The results indicate that agricultural training had no statistically significant effect on subjective evaluations of economic living conditions. However, it consistently showed positive treatment effects on average monthly household income after returning to farming and on changes in household income before and after returning to farming, despite variation in statistical significance across model specifications. These findings suggest that agricultural training positively impacts the income of returning farmers, highlighting the need for policies that expand opportunities to participate in agricultural training programs.

키워드

귀농인; 영농교육; 성향점수매칭; 경제적 성과; GBM; Returning Farmers; Agricultural Training; Propensity Score Matching; Economic Performance; Gradient Boosting Machine
제목
영농교육이 귀농인의 경제적 성과에 미치는 영향: 로짓과 GBM을 활용한 성향점수매칭 분석
제목 (타언어)
The Effects of Agricultural Training on the Economic Performance of Returning Farmers: A Propensity Score Matching Analysis Using Logit and GBM
저자
노은준; 허정회
DOI
10.36464/jrd.2026.49.3.003
발행일
2026-09
유형
Y
저널명
농촌경제
권
49
호
3
페이지
49 ~ 73