Beyond Group Mean Differences: A Demonstration With Scale Scores in Psychology

In this paper, we present a model for comparing groups on scale score outcomes. The has number of features that make it desirable analysis scores. is based ordinal regression, hence able to capture the shape data even when are highly discrete, or display marked ceiling floor effects. Additionally, incorporates hierarchical modelling create accurate summaries differences in scores across groups. Statistically, assumes ordinal, and hierarchically estimates entire distribution each group using factor smooths. Substantively, capable of: estimating location-based, dispersion-based descriptives group; uncertainty about these estimates; performing pairwise comparisons different estimates. estimation method Bayesian, however, have created GUI-based application users may install their computer run model, reducing barrier applying method. takes raw user input, runs returns multiple model-based graphical patterns data, as well tables more precise reporting. We also share code allows extend additional research contexts.