The effect of STEM education on students' science achievement: A meta-analysis study

In this research, it was aimed to determine the effect of STEM education on students' science achievement. In this direction, the meta-analysis method was used. 67 data from 54 studies were included in the scope of meta-analysis. Data of semi-experimental pattern type studies on STEM education of 3704 students were used. The data were obtained from the Council of Higher Education Thesis Center, Google Academic, DergiPark, ULAKBİM, EBSCO databases. The included studies were limited to 2018-2022. The Comprehensive Meta-Analysis program was used for the effect size values of individual studies and all analyzes performed on these values. At the end of the study, the overall effect magnitude value was determined at the 'wide' level (Hedge's g=1.038). When determining the overall effect size, the value predicted by the random effect model was used. In this study, publication bias was also tested and the results were reported. The studies were analyzed according to the publication year, publication type, class level, sample size, subject area, application time, geographical regions where applications are made and study method moderator variables. The general effect size values of the groups formed in terms of the specified moderator variables were obtained and the differences between the groups were examined. A statistically significant difference was found between the groups formed by class level, sample size and geographic regions. It was observed that the groups of other moderator variables did not differ. In addition, in moderator analyzes, the effect model suitable for each group was determined and the general effect size was decided. According to the research findings, it can be said that STEM education is very effective. For this reason, it is recommended that teachers actively use the STEM education approach in science lessons.