Critical Quantitative Methodology: Identifying and Remediating Racial (and Other) Forms of Measurement Bias with MIMIC Models

The Developmental Science Research Collaborative is hosting a talk by Dr. Matthew A. Diemer, Chair of Combined Program in Education and Psychology and Professor of Educational Studies in the Marsal Family School of Education at the University of Michigan.

Abstract:

Sound measurement is foundational to quantitative methodology in education and the social sciences. However, the racist history of quantitative methodology and measurement has led some to the conclusion that these approaches are irredeemable. Instead, acknowledging this history and repurposing quantitative methodology leads toward a more equitable, anti-racist, and critical quantitative perspective.

This presentation will explore measurement within the Critical Quantitative (CritQuant) framework. CritQuant is anchored by five guiding principles (i.e., foundation, goals, parity, subjectivity, and self-reflexivity) to mitigate racism (and/or other systems of oppression) and advance social justice.

Within the CritQuant framework, MIMIC (Multiple Indicator and MultIple Causes) models help researchers enact equity-oriented measurement by testing for latent mean differences and item intercept differences (i.e., DIF, or ‘differential item functioning’) across identity groups. These are important questions in educational and developmental sciences, such as whether a measure of perceived discrimination is equally precise across groups. By providing more fair measurement, this approach positions researchers to contribute to more equitable scholarship, policy, and practice.

Note: The Journal of Research on Educational Effectiveness paper accompanying this talk is available via https://osf.io/preprints/psyarxiv/8thpu_v1.