Improving standardised university exam scores through problem-based learning

“

" The key conclusion was that problem-based learning improved standardised ETS exam scores when prior ability was accounted for (N = 135). Nevertheless, SAT (prior ability) was the most important predictor of MFTB score, with a large effect size of 59% in the GLM regression. GPA had a much smaller effect size of 18% in GLM. Both gender and pedagogy were statistically significant but they had smaller effect sizes as compared to SAT and GPA. Problem-based learning improved MFTB scores in all five test groups. Males scored higher on the ETS exam, which was consistent with the findings in the literature. The other predictors frequently reported in the literature such as age, ethnic culture and student transfer status were not significant in this study. In particular, the ANCOVA model supported the hypothesis that problem-based learning pedagogy helped students improve their MFTB scores. The ANCOVA results confirmed SAT was the most important predictor of MFTB score (PETA2 = 45%), but most importantly, pedagogy (group treatment) was the second most influential factor in the model (PETA2 = 12%), followed by gender (PETA2 = 4%) and GPA (PETA2 = 4%), which (all factors combined together) captured 51% (adj. R2 ) of the variance on MFTB score." (p. 298)

”