Meta-analysis on the Effect of Differentiated Instruction on Students' Mathematics Achievement

The present study generally aimed to synthesize the findings of research studies on the effect of differentiated instruction to students' mathematics achievement. In this context, the results acquired from the individual studies were combined through Glassian meta-analysis technique. The research studies that were included in the list were purposively selected, and have been subjected to the process of content analysis. A research summary form was developed to gather data from the number of research studies that contain three parts such as: the general data of the study, the methodological characteristics, and the substantive characteristics. Moreover, the methodological characteristics were determined which included the research design, students' profile, number and level of participants, and sampling technique. The substantive characteristics defined and identified were content differentiation, process differentiation, product differentiation, and learning environment differentiation. In addition, the research studies were carefully coded and recorded in the research summary form. The effect size was computed using the Glass' formula. Computed effect sizes were interpreted using Glass and Cohen‘s interpretation and Sawilowsky‘s interpretation. Findings showed that in terms of methodological characteristics, the following had exceptionally attained large mean effect sizes: 1) pre-test and post-test design; 2) involvement of regular secondary students; 3) the number of participants involved (61 to 100); and 4) the procedure on simple random sampling. Further, results concerning the calculated effect sizes according to the four types of modifications on differentiated instruction revealed that product differentiation had a huge mean effect size which implies that students‘ mathematics achievement was significantly higher when product differentiation was employed. Consequently, it was recommended that product differentiation be used more often in the mathematics classrooms, but nevertheless, content differentiation, process differentiation, and learning environment differentiation may also be used as other means to improve students' mathematics achievement.