Factors influencing the implementation of STEAM instruction in lower secondary local education: A hierarchical component PLS-SEM approach

Abstract

Implementing STEAM education in mountainous lower secondary schools is constrained by limited resources and uneven teacher capacity in northern Vietnam and beyond. This study aims to examine teachers’ readiness and the feasibility of applying the STEAM model in teaching the Local Education subject in mountainous contexts, and to validate a hierarchical component PLS-SEM model of factors influencing STEAM implementation. Employing a deductive approach within a quantitative cross-sectional design, this study surveyed 285 lower secondary teachers. Data were analyzed using descriptive statistics to assess the current state of STEAM implementation, and a two-stage partial least squares structural equation modeling (PLS-SEM) approach after establishing measurement quality via Cronbach’s alpha, composite reliability, and average variance extracted (AVE). Results show that both perception and implementation criteria are rated at a fairly good level (mean approximately 3.5/5), with satisfactory internal consistency for the main constructs (Cronbach’s alpha > 0.88). The structural model indicates that Enabling Conditions (DB) and STEAM Element Specification (XD) have significant effects on the Implementation construct and jointly explain 82.3% of its variance (R² = 0.823). By contrast, teachers’ knowledge, specific pedagogical measures, and existing procedures do not exhibit significant direct effects. These findings imply that improving STEAM instruction in mountainous lower secondary schools should prioritize strengthening foundational enabling conditions (infrastructure, school-level support, and teaching resources) and designing and standardizing clear, feasible STEAM procedures adapted to local teaching contexts, supported by coherent implementation strategies and adequate structural backing rather than relying solely on individual teacher competence.