Institutional Responsibility and Racial Allyship: An Exploratory Study of Role and Gender at a PWI Utilizing the Anti-Racism Behavior Inventory Scale

Background and Purpose: Marginalized students at Predominantly White Institutions (PWI’s) rely on their institutions' protection when race-based harm occurs. Smith and Freyd’s (2014) Institutional Betrayal Theory (IBT) assert that harm occurs when institutions fail to prevent, respond to, or acknowledge transgressions in ways that protect their members. We viewed this study through an IBT lens to better understand students experiences policies and norms that sustain racial inequities.

This study aims to (1) determine to what extent institutional roles (a proxy for proximity to power within a PWI) explain variations in Anti-Racism and Behavioral Inventory (ARBI) scores for white participants and (2) determine to what extent and in what manner the intersection of institutional roles and gender explains variations in ARBI scores for white participants.

Methods: Data were collected at a Southeastern PWI via Qualtrics (May 2024–February 2025) using the ARBI, a 25-item Likert measure of anti-racism and allyship (scores 25–125; higher = greater engagement). An independent-samples t-test examined gender differences, while ANOVA with Tukey post hoc analysis assessed differences across institutional roles, gender, and their interaction. A hierarchical linear regression evaluated the effects of gender, role, and the interaction of role and gender on ARBI scores.

Results: Results from an independent-samples t-test (t(103) = -3.052, p < .003, two-tailed, 95% CI [-17.513, -3.715]) indicated a significant difference in mean ARBI scores between men (M = 93.82, SD = 13.49) and women (M = 104.43, SD = 17.68). Tukey post hoc analysis from a one-way ANOVA indicated that the mean increase from faculty to student (7.98, 95% CI [0.79 to 15.18]) was statistically significant (p = 0.026). A two-way ANOVA indicated the interaction effect between gender and institutional role was not statistically significant, F(5, 97) = 1.624, p = 0.202. Main effects indicated that institutional role was statistically significant F(5, 97) = 3.413, p = 0.037. Pairwise comparisons revealed that men faculty scored 11.91 points (95% CI, -21.85 to –1.97) lower than women faculty, a statistically significant difference (p = 0.019). Results from a hierarchical linear regression (F(2, 97) = 6.918, p < 0.001, adj. R2 = .10) indicated faculty role (-6.60, 95% CI [-12.98, -0.23]) (p = 0.04) and identifying as a man (-7.96, 95% CI [-15.22, -.70]) (p = 0.03), statistically significantly predicted ARBI scores. The interaction of gender and role was not statistically significant in the second model (-8.442, 95% CI [-22.97, 6.08]) (p = 0.25).

Conclusions and Implications: While limited to a PWI, this study highlights role-based differences in allyship and the risk of institutional betrayal. White male faculty, who often hold greater institutional power, reported significantly lower anti-racism awareness and behaviors than staff, students, and white women faculty. Although generalizability is limited, these findings reinforce reports from Black and marginalized students that white faculty frequently fail to recognize and intervene in racism and microaggressions. Such inaction, particularly those in positions of authority, may constitute institutional betrayal by signaling tolerance of harm and a failure to uphold the institution’s duty of care.