🤖 AI Summary
The conventional I² statistic is overly sensitive to sample size and reflects sampling variability in observed effect sizes rather than true between-study heterogeneity. Method: This paper formally defines “absolute heterogeneity across study populations” and proposes IA²—a novel, sample-size-invariant heterogeneity measure—deriving exact estimators for both mean difference (MD) and standardized mean difference (SMD). Contribution/Results: Through simulation studies and empirical analyses, IA² is shown to be asymptotically unbiased, scale-invariant, and fully independent of sample size. Crucially, IA² reliably detects scenarios where I² is spuriously inflated despite genuine effect consistency across studies. By providing a theoretically grounded criterion and practical tool for assessing fixed-effect model appropriateness, IA² enhances the scientific rigor and robustness of heterogeneity evaluation and model selection in meta-analysis.
📝 Abstract
Quantifying the heterogeneity is an important issue in meta‐analysis, and among the existing measures, the I2$$ {I}^2 $$ statistic is most commonly used. In this article, we first illustrate with a simple example that the I2$$ {I}^2 $$ statistic is heavily dependent on the study sample sizes, mainly because it is used to quantify the heterogeneity between the observed effect sizes. To reduce the influence of sample sizes, we introduce an alternative measure that aims to directly measure the heterogeneity between the study populations involved in the meta‐analysis. We further propose a new estimator, namely the IA2$$ {I}_{mathrm{A}}^2 $$ statistic, to estimate the newly defined measure of heterogeneity. For practical implementation, the exact formulas of the IA2$$ {I}_{mathrm{A}}^2 $$ statistic are also derived under two common scenarios with the effect size as the mean difference (MD) or the standardized mean difference (SMD). Simulations and real data analyses demonstrate that the IA2$$ {I}_{mathrm{A}}^2 $$ statistic provides an asymptotically unbiased estimator for the absolute heterogeneity between the study populations, and it is also independent of the study sample sizes as expected. To conclude, our newly defined IA2$$ {I}_A^2 $$ statistic can be used as a supplemental measure of heterogeneity to monitor the situations where the study effect sizes are indeed similar with little biological difference. In such scenario, the fixed‐effect model can be appropriate; nevertheless, when the sample sizes are sufficiently large, the I2$$ {I}^2 $$ statistic may still increase to 1 and subsequently suggest the random‐effects model for meta‐analysis.