Li, T.
A note on computing confidence intervals for the coefficient of variation in meta-analysis
Previous research in meta-analysis has suggested that the coefficient of variation (CV) is a reasonable index to represent the magnitude of between-study heterogeneity in meta-analysis. The present research aims to compute CV and the associated confidence interval with two methods: multivariate delta method and gamma function-based method. In addition, the width of the confidence interval is examined using a simulation study. The true parameters used in the simulation were based on a real-world dataset. The related R script is provided in Appendix A. This note on computing a CV intends to bring more studies to fully examine the properties of various points estimates in meta-analysis. It promotes the idea that the confidence intervals should be reported along with the point estimates in practice.
Testing, Psychometrics, Methodology in Applied Psychology, 2022, Vol. 29, pp. 201-211, DOI: 10.4473/TPM29.2.3
