rESCMA: A brief summary on effect size conversion for meta-analysis

Abstract

Effect sizes are highly relevant in quantitative research. They facilitate the comparison an quantitative synthesis of scientific studies. The main objective of this report is to present: a) a brief summary of the formulas used for conversion between the three main effect sizes used in the meta-analysis: the correlation coefficient, the standardized mean difference and the odds ratio; and b) the Rapid Effect Size Converter for Meta-Analysis (rESCMA), a open-source and browser-based app for efficiently converting and bulk-converting effect sizes and their variances based on the formulas proposed in this paper. In addition, a table summarizing the formulas is presented for easy accessibility and use.

Social support and psychological adjustment: A quantitative synthesis of 60 meta-analyses.

“…Effect sizes ( r s) were coded uniformly such that higher values indicate positive associations between social support and psychological adjustment (i.e., associations with outcomes reflecting poor adjustment were reversed). Effects reported as an odds ratio or Cohen’s d were converted to r (Villacura-Herrera & Kenner, 2020). Five meta-analyses did not provide sufficient information to determine the standard error of effects provided in their report (e.g., 95% confidence intervals).…”

Association of Personality Traits and Self-Care Behaviors in People With Type 2 Diabetes Mellitus: A Systematic Review and Meta-Analysis

“…There was methodological heterogeneity, with studies using different measures of associations. To harmonize the data for the meta-analysis, we converted the regression and/or correlation coefficients to standardized mean differences [ 55 ] and then converted them to log-odds ratios (logORs) and standard errors (SElogORs) [ 56 ]. For binary outcomes, ORs and SEs were transformed into logORs and SElogORs.…”