Skip to main content

Table 8 Ratio of ANCOVA estimate of treatment effect to true treatment effect.

From: Parametric versus non-parametric statistics in the analysis of randomized trials with non-normally distributed data

Moderate positive skew: shift 0.9955
Moderate positive skew: ratio 0.9890
Moderate negative skew: shift 1.0003
Moderate negative skew: ratio 0.9823
Uniform: shift 1.0028
Uniform: ratio 1.0005
Extreme asymmetry: shift 1.0067
Extreme asymmetry: ratio 0.9122
Extreme negative skew: shift 0.9973
Extreme negative skew: ratio 1.0055