Computing the absolute measure of heterogeneity in meta-analysis

-- maintained by Tiejun Tong


This online calculator computes the new measure of heterogeneity, I A 2 , for quantifying the heterogeneity at the study population level (Yang et al., 2025). It ranges from 0 to 1, with higher values indicating greater heterogeneity between the studies. Moreover, I A 2 is not influenced by the study sample sizes, making it an absolute measure of heterogeneity between the studies.

In contrast, the well-known I 2 statistic (Higgins and Thompson, 2002; Higgins et al., 2003) measures the heterogeneity between the observed effect sizes and is therefore highly dependent on the study sample sizes. A major limitation of I 2 is that it increases rapidly toward 1 when the sample sizes become large, making it a relative measure of heterogeneity between the studies.

To facilitate implementation, we offer two Data Input options as follows, depending on whether the Q or I 2 statistic is already available. This online calculator accommodates both single-arm and two-arm studies and supports commonly used effect sizes for continuous outcomes, including the mean difference (MD) and the standardized mean difference (SMD).

Option 1: Individual-level input



   Reset           Calculate   
Absolute measure of heterogeneity I A 2 from Yang et al. (2025)

Option 2: Aggregate-level input


      or      

   Reset           Calculate   
Absolute measure of heterogeneity I A 2 from Yang et al. (2025)

References:

K. Yang, E. Lin, W. Xu, L. Zhu and T. Tong (2025), "An alternative measure for quantifying the heterogeneity in meta-analysis", Statistics in Medicine, 44: e70089.

J. P. Higgins, S. G. Thompson, J. J. Deeks and D. G. Altman (2003), "Measuring inconsistency in meta-analyses", British Medical Journal, 327: 557-560.

J. P. Higgins and S. G. Thompson (2002), "Quantifying heterogeneity in a meta-analysis", Statistics in Medicine, 21: 1539-1558.