Efektivitas Desain Pembelajaran Proyek IPAS Terintegrasi Pendekatan STEM terhadap Keterampilan Pemecahan Masalah Kolaboratif di SMK Negeri 1 Kemangkon

Abstract

Integrated STEM and project-based learning are increasingly used to connect scientific concepts with authentic, collaborative problem contexts. This study examined the effectiveness of an IPAS project-learning design integrated with STEM at SMK Negeri 1 Kemangkon. The study compared learning gains between an experimental class receiving the intervention and a control class receiving conventional learning. A quasi-experimental pretest-posttest control-group design was used with 64 students, comprising 32 students in the experimental group and 32 in the control group. Data were analyzed using descriptive statistics, normality and homogeneity tests, paired-samples t-tests, an independent-samples t-test on gain scores, normalized gain, and Cohen's d. The experimental group increased from a mean pretest of 54.19 to a mean posttest of 86.19, while the control group increased from 56.94 to 66.28. Mean gain was 32.00 in the experimental group and 9.34 in the control group. The between-group difference in gain was statistically significant, t(62)=9.98, p<.001, with a 95% confidence interval of 18.12–27.20 and a large effect size (d=2.49). Mean normalized gain was 0.69 in the experimental group and 0.20 in the control group. The intervention was substantially more effective than the comparison learning condition. However, the uploaded dataset operationalizes the outcome through a conceptual-understanding test on solar-system content; therefore, the empirical findings should not be interpreted as a direct measurement of collaborative problem-solving competence unless a validated collaborative problem-solving instrument is also administered.