Skip to main content

Table 2 Type 1 error rates for the strategies at the simulated sample sizes

From: Using Bonferroni, BIC and AIC to assess evidence for alternative biological pathways: covariate selection for the multilevel Embryo-Uterus model

 

Sample size

Strategy

400

800

1600

3200

6400

Tests with sub-model not pre-specified

Found-in-Either

8.7%

7.4%

7.8%

8.2%

7.5%

Found-in-Both

2.6%

2.3%

2.3%

2.7%

2.5%

Global

7.0%

5.2%

4.9%

5.6%

4.9%

Bonferroni

4.7%

3.5%

3.8%

4.5%

3.9%

Tests with sub-model pre-specified

Uterus sub-model

5.8%

4.9%

5.1%

5.3%

5.1%

Embryo sub-model

5.4%

4.7%

4.9%

5.7%

5.0%

  1. Estimated simulation error (95% CI width) is ±0.55%.