. Author manuscript; available in PMC: 2017 Nov 1.

Published in final edited form as: Biom J. 2016 Aug 23;58(6):1538–1551. doi: 10.1002/bimj.201500201

Table 4.

Simulations under differential error when the true relationship with Y is null, N = 1200 simulated datasets of size N = 2000.

Predictor

r_YW

r_YG

r_WG

\frac{r_{YG} r_{WG}}{r_{YW}}

Expected relationship

Estimate of β_W or β_W_|_G (SE)

% p < 0.05

Overall

−0.400

−0.219

0.895

0.490

|β_W| < |β_W|G|

−0.155 (0.008)

100%

−0.211 (0.014)

100%

−0.223 (0.017)

100%

Overall

−0.200

−0.219

0.850

0.931

|β_W| > |β_W|G|

−0.077 (0.008)

100%

−0.058 (0.015)

96.8%

−0.060 (0.019)

92.8%

Overall

−0.004

−0.219

0.807

45.4

N/A

−0.002 (0.009)

4.4%

0.068 (0.015)

99.7%

0.073 (0.017)

98.9%

Overall

0.010

−0.219

0.803

−18.2

|β_W| < |β_W|G|

0.004 (0.008)

5.8%

0.076 (0.014)

99.9%

0.081 (0.017)

99.7%

Overall

0.200

−0.219

0.761

−0.832

|β_W| < |β_W|G|

0.077 (0.009)

100%

0.182 (0.013)

100%

0.190 (0.016)

100%

AR, acceptable reporters; UR, underreporters.