Meta-Analysis: Comparative Effect of Project-Based Learning and Problem-Based Learning on Scientific Literacy Skills in Science Learning

Abstract

The low level of science literacy among Indonesian students in PISA is a major concern, highlighting the need for innovative learning models such as Project-Based Learning (PjBL) and Problem-Based Learning (PBL). However, the comparative effectiveness of these two models has not yet been extensively synthesized quantitatively. This study aims to compare the effects of PjBL and PBL on high school students’ science literacy in science education using a meta-analysis approach. This meta-analysis follows the PRISMA 2020 protocol, with a literature search conducted across five databases (Google Scholar, ResearchGate, ERIC, ScienceDirect, and Portal Garuda) covering the period 2016–2026. Inclusion criteria included: a population of high school students; PjBL or PBL interventions; experimental designs with control groups; and the availability of complete statistical data (mean, SD, n, or t-values). Effect sizes were calculated using Hedges’ g with a random-effects model. A total of 14 studies met the inclusion criteria (10 PBL studies and 4 PjBL studies). The average effect size for PBL was 1.76 (very large category) with high heterogeneity (I² = 91.45%). The average effect size for PjBL was 0.77 (large category) with low heterogeneity (I² = 0%). The difference in effect size between PBL and PjBL, which was 0.99 points, indicates the quantitative superiority of PBL. The PBL model proved to be more effective than PjBL in improving high school students’ science literacy. It is recommended that teachers prioritize PBL, while the implementation of PjBL should be designed with a stronger emphasis on scientific inquiry.