Effect of Traditional Methods in Geometry and Numbers Learning Domains on Academic Achievement: A Meta-Analysis Study

Abstract

Nearly all studies aiming to determine the effect of modern teaching methods on academic success using an experimental design contain control groups in which traditional methods are used. Effect of traditional methods in geometry and numbers learning domains on academic achievement has not been conclusively studied by researchers yet. On the other hand, it is claimed that almost all experimental designed studies aiming to determine the effect of “modern” teaching methods, have utilized traditional methods in their control group. While there are some review studies within the literature, for experimental groups’ effects on academic achievement, no review studies on control groups’ effect have been detected so far. Consequently, our aim is to systematically review the studies’ control group findings on traditional methods in experimental researches. The purpose of this meta-analysis study is to calculate the overall effect of traditional methods in Geometry and Numbers Learning Domains (G & N LD) on academic achievement. With this in mind, data was collected from the master and doctoral theses submitted in Turkey, to indirectly answer the following research question: “Does traditional methods in (G & N LD) effect students’ academic achievement?” A meta-analysis aims to compare and combine the findings from various independent studies on a subject and determine their overall effect. Data sources of the study are studies giving pretest and posttest values for their control groups from the studies with pretest and posttest experimental and control group designs on (G & N LD).Included studies were retrieved from Advanced Thesis Search Database of Council of Higher Education (YÖK), using keywords search “geometry”, “number”, “mathematics” and “control” (both in Turkish and in English). The theses on (G & N LD) and using middle school (5th to 8th grades) as the sample were included into the meta-analysis considering the inclusion criteria. The results show that traditional methods differ by an average of 0.83 standard deviations, and it can be interpreted that traditional methods increase success in mathematics teaching. As a result of analysis, it was seen that this success was not by chance (p<0.001). As a recommendation, it is important to note that there is no loss of data since testing will not reveal whether the experienced method is different from the conventional method by testing one-sample t-test for the value obtained in their experimental study with a point increase corresponding to the 0.83 standard deviation.