The purpose of this study is to calculate the effect size, by running a meta-analysis, of the experimental studies carried out in Turkey between 2004 and 2014 that investigate the effect of learning styles on academic achievement, attitude, retention, and to define whether the academic achievement shows a significant difference in terms of learning styles model, experimental design and course type. For this purpose, a metaanalytical review method was employed to combine the outcome of the independent experimental studies. The studies included in this review were collected from CoHE National Thesis Archive (2015), ULAKBIM (2015), Google Academic (2015), ERIC (2015) and EBSCO (2015) databases. As a result of the searching process, 402 studies were assessed according to the inclusion criteria and 30 experimental studies were included in this study. Cohen's d coefficient was calculated for the effect size in this study. Because there was a high amount of heterogeneity (Q > x[superscript 2], p < 0.05) among the effect sizes of the studies, the common effect size was calculated according to the random effect model. As a result of meta-analysis, it was determined that the instructional designs based on the learning styles model had a large effect on the academic achievement (d = 1.029), attitude (d = 1.113) and retention (d = 1.290). Moreover, the academic achievement did not show any significant difference according to learning style model, course type and experimental design.