Skip to main content
. Author manuscript; available in PMC: 2008 Aug 15.
Published in final edited form as: Comput Stat Data Anal. 2007 Aug 15;51(12):6114–6122. doi: 10.1016/j.csda.2006.12.021

Table 2.

Type I Error, Statistical Power, and Coverage Probability in a Simulation Study of Quintile-Stratification based on a Misspecified Propensity Score: Correlation among propensity covariates, ρxT=0.20

Time Invariant Confounds Time-Varying Confounds Type I Error Statistical Power Coverage
x1 x2 x3 x4 x5 x6 x7 x8 α1 = 0 α2 = .22 α3 = .56 α1 = 0 α2 = .22 α3 = .56
1.25 1.00 1.25 1.00 1.25 1.00 1.25 1.00 0.06 0.88 1.00 0.94 0.95 0.95
1.25 1.00 1.00 1.00 0.06 0.86 1.00 0.94 0.95 0.95
1.00 1.25 1.00 1.00 0.06 0.87 1.00 0.94 0.95 0.95
1.00 1.00 1.25 1.00 0.07 0.90 1.00 0.93 0.95 0.93
1.00 1.00 1.00 1.25 0.05 0.88 1.00 0.95 0.94 0.95
1.25 1.25 1.00 1.00 0.05 0.86 1.00 0.95 0.95 0.95
1.00 1.00 1.25 1.25 0.07 0.91 1.00 0.93 0.94 0.92
1.25 1.00 1.25 1.00 0.06 0.90 1.00 0.94 0.95 0.95
1.00 1.25 1.00 1.25 0.05 0.87 1.00 0.95 0.94 0.94
1.25 1.00 1.00 1.25 0.06 0.88 1.00 0.94 0.94 0.94
1.00 1.25 1.25 1.00 0.05 0.91 1.00 0.95 0.95 0.93
1.75 1.00 1.75 1.00 1.75 1.00 1.75 1.00 0.07 0.84 1.00 0.93 0.95 0.96
1.75 1.00 1.00 1.00 0.07 0.82 1.00 0.93 0.94 0.93
1.00 1.75 1.00 1.00 0.06 0.83 1.00 0.94 0.94 0.94
1.00 1.00 1.75 1.00 0.13 0.94 1.00 0.87 0.86 0.72
1.00 1.00 1.00 1.75 0.09 0.87 1.00 0.91 0.92 0.91
1.75 1.75 1.00 1.00 0.07 0.83 1.00 0.93 0.94 0.94
1.00 1.00 1.75 1.75 0.18 0.96 1.00 0.82 0.77 0.53
1.75 1.00 1.75 1.00 0.14 0.95 1.00 0.86 0.85 0.67
1.00 1.75 1.00 1.75 0.08 0.85 1.00 0.92 0.91 0.89
1.75 1.00 1.00 1.75 0.09 0.88 1.00 0.91 0.91 0.89
1.00 1.75 1.75 1.00 0.16 0.92 1.00 0.84 0.83 0.69

Notes:

1) x1, x2, x5, and x6 are continuous variables

2) x3, x4, x7, and x8 are binary variables

3) x2, x4, x6, and x8 have been ignored in stratification

4) Treatment effects (αk) for dose k, are expressed in standard deviation units (i.e., effect size).