The first primer for the QuantCrit-curious critical race theorist or psychologist: On intersectionality theory, interaction effects, and AN(C)OVA/regression models

Abstract

Moderated general linear modeling (MGLM) is a highly popular statistical approach in the social sciences, as it allows analysts to examine the separate and interactive effects of 2+ variables on a numerically-measured outcome. Despite correspondences between MGLM and intersectionality theory, interdisciplinary cross-communication is rare. Quantitative research can be strengthened when vetted through a critical race theory (CRT) framework. Also, qualitative intersectionality work can be complemented with statistics. To promote greater appreciation and usage of MGLM in CRT-informed psychological research, it is argued that readers, reviewers, and editors should familiarize themselves with the basics of QuantCrit. Have all variables been accurately measured? Has the dataset been properly structured? Have all statistical assumptions been met? What data tables and figures are reported? How are the results interpreted? This primer addresses these questions while minimizing MGLM technicalities. After covering the historical context of QuantCrit, data from a houselessness dataset are examined to demonstrate the QuantCrit protocols. Limitations of MGLM, as well as QuantCrit-based guidelines for reporting MGLM results, are discussed.