Table 2 Results of assessment of type I error and power for two approaches to testing the homogeneity of between-study variance
From: A likelihood ratio test for the homogeneity of between-study variance in network meta-analysis
Number of studies (N2A, A2A) | Evaluation method | Type I error | Power |
|---|---|---|---|
(66, 40) | Monte Carlo simulation | 4.8% | 88.9% |
(66, 40) | χ2 | 8.3% | 93.5% |
(330, 200) | Monte Carlo simulation | 4.4% | 100% |
(330, 200) | χ2 | 5% | 100% |