Comprehensive Evaluation of The Support For Leadership Through Mentoring and Learning Program in Timor-Leste Elementary Schools

Abstract

The ALMA (Apoiu Lideransa liu husi Mentoria no Aprendizajen) or Support for Leadership through Mentoring and Learning Program was launched in 2016 as a bilateral collaboration between the Australian and Timorese governments with the objective of strengthening teacher professional competencies and supporting the implementation of Student-Centered Learning (SCL) in elementary schools across Timor-Leste. Despite its broad mandate, empirical evidence regarding the program's systemic effectiveness from resource allocation to long-term student learning outcomes remains limited. This study presents a comprehensive evaluation of ALMA using a Logic Model evaluation framework integrated with Partial Least Squares Structural Equation Modeling (SmartPLS) to assess the directional relationships among the program's structural components: Input, Process, Output, and Outcome. A descriptive-evaluative research design was adopted involving 100 participants from eight elementary schools, comprising 80 teachers, 6 mentors, 8 school principals, and 6 Ministry of Education officials. Data were collected through structured questionnaires, focus group discussions (FGD), and in-depth interviews, and analyzed using descriptive statistics and SmartPLS. The outer model results confirmed satisfactory convergent validity (all AVE > 0.50), discriminant validity (Fornell–Larcker criterion satisfied), and composite reliability (all ρc > 0.70). The inner model demonstrated that Input significantly predicted Process (β = 0.716, t = 15.317, p < 0.001) and Process significantly predicted Output (β = 0.738, t = 17.881, p < 0.001). The relationship between Output and Outcome was marginally significant and negative (β = −0.224, t = 1.957, p = 0.050), suggesting that while immediate training outputs are strong, their translation into long-term student learning outcomes is constrained by systemic factors. Goodness-of-Fit (GoF = 0.444) indicated a robust structural model. Qualitative findings corroborated quantitative results, highlighting workload concerns and the need for sustained institutional support. The findings underscore the importance of reinforcing post-training follow-up mechanisms and aligning program governance structures with the long-term capacity-building goals of educational reform in Timor-Leste.