Multidimensional Associations Between Health-Related Quality of Life and Digital Inclusion, Socioeconomic, and Agricultural Characteristics Among Rural Farmers

บทคัดย่อ

Health-related quality of life (HRQoL) among rural farmers reflects a complex interplay of demographic, socioeconomic, agricultural, household, and digital characteristics. However, their multivariate relationships with multidimensional HRQoL remain insufficiently understood. This cross-sectional study examined these relationships among 311 farmers in Nakhon Si Thammarat Province, Thailand, using canonical correlation analysis (CCA). HRQoL was assessed using a researcher-developed 60-item questionnaire covering six domains: physical health, mental well-being, social relationships, environmental quality, work participation, and economic welfare. The questionnaire was reviewed by three experts for content validity (IOC = 0.80–1.00) and pilot-tested among 30 farmers. Internal consistency was acceptable (Cronbach's α = 0.89; domain-level α = 0.78–0.86). CCA yielded six canonical functions, of which only the first was statistically significant (Wilks' λ = 0.396, Rc = 0.583, Rc² = 0.340, p < 0.001), indicating a moderate multivariate association between the predictor and HRQoL variable sets. The largest positive structure coefficients among predictors were observed for digital training score (rs = 0.716), internet accessibility (rs = 0.579), farm size (rs = 0.571), and family relationship (rs = 0.535), while age showed a negative coefficient. (rs = −0.309). Among HRQoL domains, mental well-being (rs = 0.847), work participation (rs = 0.829), physical health (rs = 0.802), and economic welfare (rs = 0.774) showed the largest coefficients. The predictor canonical variate accounted for 21.3% of the variance in HRQoL, whereas the HRQoL canonical variate accounted for 2.9% of the predictor variance. These findings indicate a meaningful multivariate association rather than causal effects and should be interpreted within the study context.