Purpose This study probes the determinants of human capital investment among farming households in Agam Regency, Indonesia, challenging classical economic assumptions by testing the relative influence of psychosocial assets against conventional financial factors. Design/methodology/approach We employed a survey of 391 farmer households, selected via proportional random sampling across four key agricultural districts. Data were analyzed using structural equation modeling (SEM) with LISREL 8.80 to test a model integrating educational variety, internal locus of control, household income and social capital. Findings Our analysis reveals a nuanced picture. While social capital and educational variety significantly promote human capital investment, household income shows no statistically significant effect. Contrary to initial expectations, internal locus of control also fails to emerge as a direct driver in this specific context. The results underscore social capital as the pivotal factor shaping educational investment decisions in these homogeneously poor communities. Practical implications The findings urge a strategic pivot in policy. Rather than focusing primarily on financial subsidies, interventions should be designed to strengthen community social capital – such as through trusted networks and cooperative structures – while simultaneously improving the accessibility and perceived value of diverse educational pathways. This psychosocial approach is likely to be more effective in fostering sustainable human development in marginalized rural settings. Originality/value This research breaks new ground by directly comparing economic, social and psychological factors within low-income farming communities, providing empirical evidence for a psychosocial asset-based model of human capital investment.
Hastuti et al. (Mon,) studied this question.