The Effectiveness of the POE (Predict, Observe, Explain) Learning Model on Students' Learning Outcomes

Abstract

This study aims to examine the effect of the Predict, Observe, Explain (POE) learning model on students' learning outcomes in a specific subject. The research employed a quantitative approach with a quasi-experimental design, specifically a posttest-only control group design. The sample consisted of two eighth-grade classes at SMP Gajah Mada Bandar Lampung, selected through simple random sampling. The experimental class received instruction using the POE model, while the control class was taught using conventional methods. The research instrument was a multiple-choice test designed to assess learning outcomes in the cognitive domains of C3 (application) and C6 (evaluation). The test consisted of 20 items, of which 14 were valid, and had a reliability coefficient of 0.733. Data analysis using an independent samples t-test showed that the Sig. (2-tailed) value in Levene’s Test was 0.001, indicating a significant difference in learning outcomes between the experimental and control classes. Therefore, the research hypothesis was accepted, demonstrating that the use of the POE model has a significant effect on improving students’ learning outcomes. These findings suggest that innovative learning models such as POE are more effective than conventional methods in enhancing students’ higher-order thinking skills.