The Inner Workings of the Intersectional Discrimination Index: (re)assessing the Internal Validity of the Anticipated, Day-to-day, and Major Discrimination Measures.

Of the three Intersectional Discrimination Index (InDI) scales, only the anticipated discrimination measure (InDI-A) has been subjected to factorial analysis to date. This study expands upon such evaluation by assessing the internal validity of the lifetime day-to-day (InDI-D) and major discrimination (InDI-M) scales, in addition to further examining the anticipated discrimination measure. For each InDI scale, administered to diverse Canadian (n = 1,065) and U.S. (n = 1,518) samples in 2016, we determined the number of underlying factors, estimated item loadings, and assessed residual correlations. We also examined the correlation between the InDI-D and InDI-M, and the degree to which the three InDI measures are invariant across Canada and the United States. We show that the InDI is characterized by three good-fitting unidimensional measures, moderate to strong loadings, and five pairs of potentially redundant items. There is considerable but not excessive overlap between the InDI-D and InDI-M, and estimates of anticipated, day-to-day, and major discrimination are invariant across countries. Given these findings, we suggest that the InDI measures have good internal validity both in Canada and the United States. To further develop the scales, future studies should evaluate their scalar properties and external validity concerning health. Additionally, these studies should determine whether the measures provide discrimination estimates that are meaningfully comparable across groups at the intersections of different social statuses and positions.