A Survey Study for The Comparison of Meta-Analysis Softwares

Abstract

In recent years, meta-analysis studies have become a popular field of study in order to quantitatively analyze the results obtained from more than one study carried out independently from the same subject and to interpret the results from a holistic perspective. Because of this, different licensed and open source software has been developed in the last 20 years. With this study, it is aimed to compare the existing programs and software in terms of meta-analysis in terms of quantitative and qualitative aspects. For this purpose, the results obtained from Jamovi, CMA, RStudio and Meta-Essentials programs were compared within the determined criterias. As a result of filtering the 1906 studies obtained from databases in accordance with the flow diagram, the number of studies to be included in the meta-analysis was determined to be 32. As a result of the study, forest plots obtained from different programs do not differ according to the program used. In addition, the open source and free of R, Jamovi and Meta-Essentials programs have been determined as the superiority of the programs compared to CMA. According to the findings, Jamovi has much more test for heterogeneity compaired to the other softwares. At the same time, the number of model prediction options is much higher for Jamovi compared to Meta-Essentials and CMA. While other software does not provide any information regarding model fit index and information criterion, Jamovi offers a rich output in this regard. It can be said that these features make Jamovi advantageous over other software.