A Meta-analysis in Science Learning and Teaching in M.Ed. Theses in the Curriculum and Instruction Program, Mahasarakham University

Abstract

The purpose of the study was to synthesize M.Ed. theses in science learning and teaching in the curriculum and instruction program, Faculty of Education, Mahasarakham University, employing meta-analysis. The synthesis dealt with 81 theses published between 2012 and 2016. The instruments were a research quality assessment form and a research component report form. The data were effect sizes calculated according to Hunter’s approach, with 128 values. The data were analyzed with the descriptive analysis and one-way ANOVA in the effect size comparison. The results showed that the quality of the theses was at a very high level. The greatest number of theses was published in 2013. Constructivism was the most employed as the main principle of organizing for learning. Sampling involved 2 samples, with 61-90 subjects. Most of the subjects were in grade 10-12. The maximum duration of experiment was over 30 hours. The highest number of independent variables were 2 variables. The dependent variable most studied together with learning achievement was the attitude towards the course. The highest number of the tools used was of 4 types, with plans of organizing for learning as the most often used. Examination of the quality of the tool covered 4 aspects: validity, difficulty, discrimination and reliability. The basic statistics most used were percentage and the mean, and t-test was most often used to test the hypothesis. The effect size compared by using one-way-ANOVA indicated that the learning achievements of the students who studied with experiential learning and blended learning in Constructivism had no difference.