Passing TPACK through the Meta-science circle: A Meta-analytic Systematic Review of the application of the Model

This systematic review and meta-analysis evaluates the impact of the Technology Pedagogical Content Knowledge (TPACK) model on student learning outcomes. A total of 126 empirical studies met our inclusion criteria, substantially more than in previous meta-analyses (e.g., 8 and 59 studies), due in part to our broader and more rigorous selection approach, which avoided biasing results based on study outcomes. The included studies involved more than 15,000 students, with a strong bias toward university settings (75.77%), particularly in teacher education programs (89.57%). Despite the widespread adoption of TPACK as a framework for technology integration in education, our findings indicate that evidence of its effectiveness remains limited, methodologically uneven, and insufficient to establish its superiority over other models. A thorough analysis showed a more than plausible publication bias. Applying a Robust Bayesian Meta-analysis allowed us to infer the effect after correcting for possible publication bias, resulting in a small and statistically non-significant size. The methodological quality of the included studies was very poor with important shortcomings; many studies lacked solid research designs, such as pretest-posttest configurations with active control groups, and instead abounded in designs without a control group. We conclude that more rigorous and specific research is urgently needed to clarify the real impact of the model on learning and to guide evidence-based decisions in teacher education.