We prepared data for use as raw ordinal data in the statistical modeling package Mx (Neale et al. 2003). This approach assumes that the ordinal categories are representative of an underlying normal distribution of liability, with thresholds in liability discriminating between categories. Measures of MD, RUC and RUS were coded 0 or 1; NDC was grouped into five categories coded from 0 to 4; NDS was grouped into four categories coded from 0 to 3.
