Skip to main content

Table 3 Summary of the simulation setup (confidence intervals).

From: Parametric methods outperformed non-parametric methods in comparisons of discrete numerical variables

Confidence intervals

U: the Welch U (T adjusted for unequal variances)

 

PB: percentile bootstrap (2000 samples)

 

Bt: bootstrap-t (2000 samples)

Outcome scales

{0, 1, 2}, {0, 1, 2, 3,}, {0, 1, 2, 3, 4}, {0, 1, 2, 3, 4, 5}

Distributions

Uniform, normal, u-shaped, linear trend, step, skewed

Sample sizes (m, n)

(10, 10), (50, 50), (25, 10), (100, 25)

Nominal confidence level

95%

Replications

10 000

Programming language

Matlab [10] with the Statistics Toolbox