LAUSR

There is a current explosion in data science techniques that have led to numerous new ways of classifying data and its structure. In our study [Park et al., 2019], we applied the Schmid–Leiman transformation, verified the appropriateness of the second-order factor models of the World Health Organization Disability Assessment Schedule II (WHODAS-II), and we provided the item-level R values for general disability factor and group (subscale) factors. On the other hand, Williams [2019] applied McDonald’s omega in his additional analysis to determine whether the subscale scores were interpretable as a measure of a single factor or not (focused on the domain level). Williams’ conclusions are appropriate and well founded. We also note that the item-level R values from our Schmid–Leiman transformation also indicated that the items in cognition and participation subscales provided little additional information beyond the overall disability factor (see Table 1). So, we agree with the authors conclusion that the cognition and participation subscales may not contain enough information beyond the overall score to be standalone dimensions using our reported sample size. Nevertheless, we also contend that the