Skip to main content
. 2009 Dec 11;85(6):862–872. doi: 10.1016/j.ajhg.2009.11.001

Table 3.

Type I Error Rates (×104) at the Nominal Significance Level of 10−4 for Testing Equal Odds Ratios When Control Samples Are Shared Between Two Case-Control Studies

No. of Cases
Original Data
Summary Results
Study 1 Study 2 No. of Controls Robusta Naiveb Robusta Naiveb
1000 1000 1000 0.95 0.001 1.65 0.001
1000 1000 2000 0.98 0.027 1.36 0.027
1000 1000 3000 0.99 0.091 1.25 0.091
1000 1000 4000 1.00 0.173 1.21 0.173
1500 500 1000 0.92 0.004 1.42 0.004
1500 500 2000 0.93 0.084 1.19 0.084
1500 500 3000 0.93 0.177 1.11 0.177
1500 500 4000 0.94 0.265 1.08 0.265
a

Robust variance estimator used.

b

Naive variance estimator used.