Understanding how entrepreneurial intentions transform into entrepreneurial actions: A two-stage PLS-SEM analysis and binary logistic analysis from a mountain emerging economy.

<ns5:p> Background Entrepreneurial intention does not automatically translate into entrepreneurial action, and the mechanisms governing this conversion remain insufficiently theorised and rarely tested in peripheral, institutionally constrained contexts. This study examines the determinants of entrepreneurial intention and its conversion into actual entrepreneurial decision among young people in Lao Cai, a multi-ethnic border mountain province in northern Vietnam representative of institutionally thin, peripheral emerging economies in Southeast Asia. Methods Drawing on an integrated framework combining the Theory of Planned Behaviour with resource-based and entrepreneurial ecosystem perspectives, the study employed a two-stage design: Partial Least Squares Structural Equation Modelling (PLS-SEM) to estimate the effects of seven antecedents on entrepreneurial intention, and binary logistic regression to assess the conversion of intention into actual decision. Survey data were collected from 307 young people (196 active entrepreneurs, 111 with confirmed intention) across all nine districts of the province. Results The structural model explained 74.8% of the variance in entrepreneurial intention (R <ns5:sup>2</ns5:sup> = 0.748; adjusted R <ns5:sup>2</ns5:sup> = 0.742). Entrepreneurship education (β = 0.348), experience/readiness (β = 0.242), and perceived behavioural control (β = 0.210) were the strongest predictors, followed by ecosystem support (β = 0.195) and attitude (β = 0.130); subjective norms (β = 0.054, p = 0.357) and social capital (β = 0.019, p = 0.717) were non-significant. Entrepreneurial intention strongly predicted actual decision (B = 2.137; Exp(B) = 8.472; p < 0.001), an odds ratio roughly twice the meta-analytic benchmark, though discriminative power was modest (AUC = 0.604, 95% CI [0.537, 0.671]). Conclusion These findings provide rare quantitative evidence from an underrepresented mountainous, multi-ethnic Southeast Asian context, propose a replicable two-stage analytical framework for intention-to-action research, and inform policy on entrepreneurship education, experiential capital, ecosystem investment, and inter-ethnic entrepreneurial networks. </ns5:p>