Impact of within-case variability on Tau-U indices and the hierarchical linear modeling approach for multiple-baseline design data: A Monte Carlo simulation study

This study focuses on two different measures to quantify intervention effectiveness for multiple-baseline design data. The influence of 216 design conditions on the estimate of the Tau-U index (with three different variants) and the regression-based effect size is empirically investigated through a Monte Carlo simulation study. Results demonstrate that the magnitude of the within-case variance influences the Tau-U indices the most when the true intervention effect is small relative to large within-case variance. The within-case variance does not systematically influence the regression-based effect size measure. The Critical Tau-U is introduced to help determine if Tau-U estimates indicate evidence in support of an intervention effect and address underlying issues with currently used benchmarks.