Özet:

In this study, it is aimed to synthesize the results obtained from the experimental studies investigating the effect of laboratory-based science education practices on academic achievements of the students in Turkey via meta-analysis method. For this purpose, articles, master's theses and doctoral dissertations having the statistical data that can be included in the meta-analysis study and appropriate to the research problem as a result of the related literature review made on the studies conducted in Turkey between 2013-2018 were analyzed by reviewing in Turkish and English languages from the national and international databases. As a result of the literature review, a total of 34 studies concerning the effect of student-centered teaching methods in science laboratories on academic achievements of the students in laboratory courses and including appropriate data for the coding protocol were included in the meta-analysis. In these individual studies with totally 2237 participants moderator analyses were performed for the variables of educational level, publication types, teaching methods, science areas and application regions whereas, meta-regression analyses were performed for the variables of sample size and application durations. It was determined that all the studies included in the study had a positive effect size value. While the effect size calculations of the studies were performed using heterogeneity, publication bias and intervening variable analyses in the “Comprehensive Meta-Analysis Version 2.0” statistical software. A heterogeneous distribution was determined in the studies included (Q=79.41 p<0.05, I2=58.44). Overall effect size value of student-centered teaching methods concerning academic achievements of the students in laboratory courses was found to be d=0.94 (95% CI, SE=0.07) at the confidence interval of 0.80 and 1.09 using random effects model. According to the classification by Cohen, Manion & Morrison (2011), it was determined that laboratory-based learning approach affected academic achievements of the students moderately. As a result of the moderator analyses, it was found that the effect sizes did not vary according to the variables of educational level, teaching methods, publication types, science areas, and application regions.

Anahtar Kelimeler: