Justice-Oriented, Antiracist Validation: Continuing to Disrupt White Supremacy in Assessment Practices

ABSTRACT

Traditional validation approaches fail to account for the ways oppressive systems (e.g. racism, radical nationalism) impact the test design and development process. To disrupt this legacy of white supremacy, we illustrate how justice-oriented, antiracist validation (JAV) framework can be applied to construct articulation and validation, data analysis, and score reporting/interpretation phases of assessment design/development. In this article, we use the JAV framework to describe validation processes that acknowledge the role and impact of race/racism on our assessment processes—specifically construct articulation, analysis, and score reporting—on Black, Brown, Indigenous, and other students from historically marginalized populations. Through a JAV framework, we seek to disrupt inaccurate white supremacist approaches and interpretations that for too long have fuelled measurement practices.

KEYWORDS:

• Antiracist validity
• justice-oriented validity
• antiracist assessment

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ We use the phrase “Black, Brown, Indigenous, and other students from historically marginalized populations” with caution. All terms, including BIPOC, students of color, and minoritized students, that seek to capture the range of historically oppressed groups in the U.S. are always partial, contested, and shift over time. For example, we understand that Asian Pacific Islander Desi American (APIDA) and Middle Eastern/North African (MENA) students are not specifically named in our chosen term in this chapter.

² Our work aligns with other educational scholars who have called for the interruption of injustice in educational research (e.g., Souto-Manning & Winn, 2017).

3 We see construct articulation as defining, describing, and critically interrogating a construct. It requires one to determine prior to developing the assessment itself, what constellation of knowledge and skills define what we are trying to measure (and teach to, and learn). Construct validation includes critical appraisal of our construct of interest, analysis of construction representation, and analysis of cognitive data (including process data) to determine if the test items are in fact measuring the construct elements they were blueprinted to measure.