Problem solving based learning in math educati̇on: Meta - analysis study

In this master's thesis, meta-analysis was conducted on experimental work discussing the impact of problem solving-based learning on achievement and attitude in mathematics education. Meta-analysis is a method of combining the results of multiple studies which are performed on a specific issue independently and making the statistical analysis of the results of findings. For that purpose, the effect size between variables has been demonstrated by assembling numerically the findings of experimental studies have been carried out in Turkey and abroad between the years 2001-2015. 54 , and 97 master's and doctoral dissertations at home and abroad respectively ; 117 domestic, 146 overseas articles and papers have been identified in order to determine the impact of problem solving-based learning on achievement and attitude. From these studies only pretest-posttest control group studies and the studies revealing all the necessary data for calculating the effect size were included in the meta-analysis process. As a consequence, 62 studies were combined with meta-analysis method. Considering meta-analysis findings, it is found that studies on academic achievement and attitude are performed in 2006 in Turkey, whereas studies focusing on just achievement issue are mostly performed in 2010 with respect to problem solving-based learning. Considering in terms of topics, "Equations and Inequalities" has been mostly the subject field of studies included in meta-analysis of the effectiveness of problem solving-based learning in math class in Turkey in terms of academic achievement. Following this topic, the geometry is the second topic on most of the studies. However, no significant subject field was found for the attitudes. Nonetheless, it is seen that "Numbers" and "geometry" topic fields are mostly included in overseas experimental model of empirical research. Regarding the findings of studies carried out according to their education level, the studies on the effectiveness of problem solving-based learning on academic performance and attitude at secondary level in math class in Turkey, constitutes 50 % of all studies included in the meta- analysis. In other words, it was found that the majority of research was done at the secondary school stage. Moreover, we have the same results at overseas undergraduate level studies. Ankara is seen as hosting most studies on academic achievement and attitude among of all the cities included in meta-analysis researches in Turkey. When the countries are analyzed according to number of meta- analysis study: the United States of America (USA) in terms of academic achievement; Pakistan in terms of attitude have been identified as countries that do most work. When the experimental studies included in the meta-analysis are examined in terms of implementation periods, studies are generally applied between 1-4 weeks in Turkey, whereas the application of studies abroad spread as long as 5-6 weeks. This study is based on classification system according to Cohen (1988). The effect sizes attained represent a large effect size for Turkey about academic achievement, and a medium effect size for abroad studies. Among all of the domestic and foreign studies included in the meta- analysis, it is shown that 96.55 % of them have a positive effect size. It is found that, the overall effect size of the study included in the meta-analysis for Turkey is E. S. = 0,512; for abroad it is E.S.=0,775 in terms of attitude. Those both values are regarded as medium effect size according to Cohen (1988). This result signifies that problem solving-based learning compared to classical syllabus practices improve students' attitudes towards Math class 40 % in Turkey and 58 % abroad positively. At the end of research, researchers are suggested to plan their research before starting their work, take into consideration workload, have alternative ways to get over the difficulties and continue their works intensely. Key Words: Problem Solving Based Learning, Mathematics, Academic Performance, Attitude, Meta-Analysis