Consumer-Driven Racial Stigmatization: The Moderating Role of Race in Online Consumer-to-Consumer Reviews

The paper investigates how race, as signaled by reviewers’ profile pictures, moderates the impact of online consumer-to-consumer (C2C) review valence (positive vs. negative) on readers’ perceptions of source credibility and information adoption. Motivated by persistent racism in markets and the growing influence of C2C interactions, the study extends research on marketplace racism—traditionally focused on firm-driven communications and algorithms—into the consumer-driven realm. Drawing on Critical Race Theory (CRT), the authors posit that racial cues shape how consumers interpret information, potentially embedding racial bias into judgments of credibility and usefulness of reviews. The research problem is operationalized via hypotheses that reviewer race moderates the effects of review valence on credibility and adoption, with expectations of lower credibility and adoption for Black reviewers (especially for negative reviews) and higher credibility/adoption for White reviewers, with Asians in-between.

Theoretical background integrates three streams: (1) Valence, credibility, and adoption of online reviews: Source credibility influences perceived review quality and purchase intentions. Prior findings are mixed on whether positive or negative reviews have higher credibility and adoption, but peripheral cues (e.g., reviewer information) matter. (2) Profile avatars/pictures: Presence of profile pictures increases perceived credibility and can amplify review effects; however, prior work largely examines presence/absence rather than racial content of images. (3) Critical Race Theory (CRT): Race is a social construct shaping inferences about moral and mental qualities; online platforms can mediate racial discrimination. Prior marketing research documents racial bias in marketplaces (e.g., sharing economy discrimination, ad delivery bias). The authors argue that revealing reviewer race via profile pictures may alter perceived credibility (moral) and information adoption (competence). Hypotheses: H1—Reviewer race (via picture) moderates the impact of review valence on source credibility and information adoption compared to no picture. H2—Black reviewers’ negative reviews will show lower credibility and adoption than their positive reviews and than White/Asian reviewers. H2a—White reviewers’ reviews (positive and negative) will show higher credibility and adoption than Asian and Black reviewers.

Two experimental, between-subjects studies using mock TripAdvisor hotel reviews with controlled content and reputational cues. Study 1: 2 (Valence: positive vs. negative) × 2 (Profile avatar: Identified Black vs. Unidentified/no photo) design; four scenarios. Stimuli: identical review texts across conditions; same helpfulness badges. Sample: N=200 Prolific participants (cell n≈50; 35% female; mean age ~25 years), racially diverse (White 39.5%, Black 20.8%, Asian 19.5%, Caribbean 7.5%, mixed 7%, Hispanic 4%, other 1.7%). Screening ensured TripAdvisor usage (M=4.50/5). Scenario realism M=6.05/7. Measures: Source credibility and information adoption adapted from Wu & Shaffer (1987). Controls: attitude toward checking online reviews (Donthu & Gilliland, 1996; Qiu et al., 2012) and perceived TripAdvisor credibility (Qiu et al., 2012). Manipulation checks confirmed valence and race/avatar perceptions (χ² p<.001). Psychometrics: convergent/discriminant validity satisfactory (AVE>0.5; loadings/reliabilities >0.7). Analysis: MANOVA after checking assumptions (Levene p>.05; Box’s p=.820). Study 2: 2 (Valence: positive vs. negative) × 3 (Profile avatar: White vs. Black vs. Asian) design; six scenarios. Stimuli controlled for identical content and neutral facial expressions. Pretest of photos (n=20) confirmed equal expressions and race as the only salient difference. Sample: N=300 Prolific participants (cell n≈50; 40% female; mean age ~26 years), racially diverse (White 38.3%, Black 29.5%, Asian 19.7%, Pacific Islander 6.5%, Latino/Hispanic 4.7%, mixed 1%, other 0.3%). TripAdvisor usage M=4.16/5; realism M=6.00/7. Same dependent variables and controls; manipulations successful (χ² p<.001). Validity and reliability met thresholds. Analysis: MANOVA after assumption checks (Levene p>.05; Box’s p=.850).

Study 1: Significant interaction between review valence and reviewer profile picture (Wilks’ λ=.833, F(2,122)=10.160, p<.001), significant on both source credibility and information adoption (p<.001). Means: - Source credibility: Black reviewer—positive 4.00 vs. negative 2.61; Unidentified—positive 2.00 vs. negative 4.56. - Information adoption: Black reviewer—positive 3.00 vs. negative 1.80; Unidentified—positive 2.00 vs. negative 3.50. Interpretation: Racial cues altered evaluations. Black reviewers’ negative reviews were least credible and least adopted; positive Black reviews were more credible than their negative ones, while no-photo reviewers’ negative reviews were more credible/adopted than their positives, indicating activation of racial signifiers. Study 2: Significant interaction between valence and reviewer race (Wilks’ λ=.944, F(4,584)=8.895, p<.001), significant on both outcomes (p<.001). Means: - Source credibility (positive): White 5.53, Asian 5.20, Black 4.10. (negative): White 5.25, Asian 4.20, Black 2.30. - Information adoption (positive): White 5.00, Asian 4.60, Black 3.10. (negative): White 5.12, Asian 4.00, Black 2.00. Findings support H2 and H2a: White reviewers’ opinions (positive and negative) had the highest credibility and adoption; Asian reviewers were intermediate; Black reviewers were lowest, especially for negative reviews. Positive reviews from Black reviewers increased perceived credibility relative to their negative reviews (expectancy violation effect), but adoption remained lower than for White and Asian reviewers. Overall, results evidence consumer-driven racial stigmatization in C2C reviews and a racial hierarchy in perceived credibility and usefulness.

The findings demonstrate that race functions as a negatively charged meaning-maker in C2C online reviews, shaping how readers assess source credibility and whether they adopt shared information. This extends the literature beyond profile-picture presence to the racial content of avatars, revealing an epistemic form of racism: consumers filter and justify knowledge based on racialized assumptions. The results align with CRT by showing how racial biases prevalent in society are enacted in unregulated C2C communication spaces, thereby influencing marketplace norms. Black reviewers’ negative reviews were particularly discounted, while White reviewers’ opinions—regardless of valence—were most trusted and adopted, with Asians evaluated closer to Whites than Blacks. These patterns explain and quantify how race moderates the persuasive impact of review valence, addressing the research question and underscoring the broader relevance to marketing, platform governance, and societal equity.

This research empirically establishes that reviewer race, signaled via profile pictures, moderates the effects of review valence on perceived source credibility and information adoption in C2C platforms. Across two experiments, White reviewers were consistently evaluated as more credible and influential than Asian and Black reviewers; Black reviewers, especially when posting negative reviews, were least credible and least adopted. The work advances CRT within marketing by documenting consumer-driven racial stigmatization in online reviews and reframing marketplace racism as an epistemic bias in information processing. Practical implications include consumer education on unconscious bias, platform design choices (e.g., reducing race-revealing avatars or adding counter-bias measures), and managerial strategies to engage with and reinforce credible content from stigmatized groups. Future research should examine additional identity cues (age, gender, status), cross-attribute filtering, and the interplay of platform algorithms with user-level biases.

- Stimuli omitted other identity cues (e.g., age, social status, education, personality) to avoid homophily; future work should manipulate and test these as moderators/mediators. - Only one gender (male) and one facial expression (neutral) were used; future studies should incorporate female/male avatars and varied expressions. - Further research is needed on source credibility effects when reader and reviewer attributes are filtered/matched, and how this impacts racial stereotyping. - Findings are based on mock TripAdvisor scenarios and Prolific samples; generalizability to other platforms, contexts, and behaviors should be tested.