Triggering competence may protect multiple minority members from hiring discrimination

The study examines how intersectional identities shape hiring evaluations under varying informational conditions, focusing on Portuguese perceptions of Portuguese versus Brazilian male applicants who are either straight or gay. Grounded in the stereotype content model (warmth, competence) and theories of intersectionality and cognitive compartmentalization, the research asks whether triggering explicit competence reduces discriminatory judgments and whether multiple-minority status (gay Brazilian men) can mitigate negative stereotypes associated with a single minority status (Brazilian nationality) when competence cues are absent. Portugal offers a pertinent sociocultural context with relatively inclusive LGBTQ+ policies and a complex post-colonial relationship with Brazilian immigrants. The work aims to clarify when and how stereotypes influence warmth, competence, and salary decisions in hiring-like evaluations.

The stereotype content model (SCM) posits that stereotypes cluster along warmth and competence, predicting distinct emotions and behaviors toward outgroups. Prior work shows unipolar stereotyping (e.g., men as more competent in STEM; Black men as less warm/threatening; disabled individuals as less competent) and highlights added complexity for intersectional targets. Intersectionality research reports mixed patterns: integrative processes can benefit some multiple-minority targets (e.g., gay Black men sometimes viewed as more competent leaders than straight Black men), whereas dominance processes and double-jeopardy can harm others. In occupational contexts favoring prototypical masculinity/whiteness, some intersectional targets gain (e.g., gay Black men offered higher salaries than straight Black men; lesbians perceived as better collaborators), but other findings conform to additive disadvantages (e.g., female Maghreb candidates face more negative evaluations; backlash against older women showing masculine behaviors). Ambiguity and cognitive load can amplify reliance on stereotypes. Portugal’s context includes implicit prejudice toward Brazilians (e.g., accent-based discrimination), despite anti-discrimination norms, and relatively favorable attitudes and legal protections toward LGBTQ+ people. The literature suggests that individuating information (e.g., competence cues) can reduce stereotyping, whereas ambiguity encourages heuristic, stereotype-based judgments.

The project comprised a pilot study and two experiments with Portuguese university participants. Pilot study: N=118 (aged 18–56, M=21.02, SD=5.71; mostly women; all students). Explicit attitudes were measured with adapted Attitudes Toward Gay Men (ATG; 10-item subset) and a parallel Attitudes Toward Brazilian Men (ATB; 10 items), rated 1–7. Confirmatory factor analyses supported unifactorial solutions (ATG: χ²(33)=123.20, p<0.001, CFI=0.89, TLI=0.85, RMSEA=0.15, SRMR=0.07; ATB: χ²(33)=46.18, p=0.064, CFI=0.97, TLI=0.96, RMSEA=0.06, SRMR=0.06); reliabilities were ω=0.88 (ATG) and ω=0.80 (ATB). Implicit preferences were measured via two Portuguese IATs (sexuality: straight vs gay; nationality: Portugal vs Brazil) using iatgen. The sexuality IAT showed a weak preference for straight over gay (M=0.11, SD=0.40), t(116)=3.14, p=0.002, d=0.29, α=0.75; the nationality IAT showed a moderate preference for Portuguese over Brazilian (M=0.40, SD=0.36), t(117)=12.03, p<0.001, d=1.11, α=0.73. Explicit–implicit correlations were negligible; ATG–ATB correlated r=0.65, p<0.001; IAT d-scores correlated weakly r=0.19, p=0.04. Experiment 1 (high competence CVs): N analyzed=77 (47 men, 29 women; ages 19–29, M=22.13, SD=1.87; mostly heterosexual). Between-subjects 2×2 design: nationality (Portuguese vs Brazilian) × sexual orientation (straight vs gay). CVs indicated high explicit competence via strong education/experience; nationality manipulated via names, universities, and personal statements; sexual orientation via leisure/organization cues. Participants evaluated one randomly assigned CV and rated warmth (friendliness, threat), competence (competence, hirability), masculinity/femininity, and proposed a starting salary. Measures showed acceptable reliability (e.g., friendliness α=0.74, threat α=0.73, competence α=0.79, hirability α=0.68). Analyses used 2×2 ANOVAs (with initial checks for participant gender effects) and correlations among subscales. Experiment 2 (low competence CVs): N=120 (65 men, 52 women, 3 unspecified; ages 19–59, M=25.15, SD=6.03; mostly heterosexual). Same 2×2 design and manipulations, but CVs provided less objective information and reflected less-experienced applicants (lower implied competence). Masculinity/femininity combined into one scale. Participants rated warmth (friendliness, threat), competence (competence, hirability), masculinity/femininity, and proposed a starting salary. Reliabilities were lower than Experiment 1 (friendliness α=0.25; threat α=0.63; competence α=0.57; hirability α=0.65), interpreted as a byproduct of increased ambiguity. Analyses paralleled Experiment 1. A mega-analysis (three-way ANOVA) compared type of CV (high vs low competence) × nationality × sexual orientation on warmth, competence, masculinity, and salary.

Pilot study: Explicit attitudes toward gay and Brazilian men were low (item means generally <2), with acceptable reliability. Implicit measures showed a weak pro-straight bias (M d=0.11, SD=0.40, p=0.002) and a moderate pro-Portuguese bias (M d=0.40, SD=0.36, p<0.001). ATG and ATB correlated strongly (r=0.65), and sexuality and nationality IAT d-scores correlated weakly (r=0.19). Experiment 1 (high competence): No significant main effects or interactions on warmth or competence ratings. Warmth means: Portuguese M=4.03 vs Brazilian M=4.15, p=0.161; straight M=4.09 vs gay M=4.09, p=0.980. Competence means: Portuguese M=5.29 vs Brazilian M=5.32, p=0.908; straight M=5.30 vs gay M=5.32, p=0.921. Salary proposals did not differ significantly by nationality (Portuguese M=1201.08 vs Brazilian M=1148.60) or sexual orientation (straight M=1143.69 vs gay M=1204.74), interaction p=0.063, with straight Brazilians descriptively lowest. Gay candidates were rated more feminine than straight (F(1,73)=6.76, p=0.011, η²=0.08). Competence correlated positively with salary (r=0.43, p<0.001); warmth correlated negatively with salary (r=-0.25, p=0.030). Experiment 2 (low competence): Warmth showed main effects of nationality and sexual orientation, and a significant interaction: straight Portuguese were rated warmer than all other groups (F(1,116)=17.54, p<0.001, η²=0.10). Competence showed main effects and an interaction: straight Brazilians were rated significantly less competent (M=3.82) than straight Portuguese (M=5.64), gay Portuguese (M=5.45), and gay Brazilians (M=5.47); all others did not differ (F(1,116)=74.24, p<0.001, η²=0.24). Masculinity/femininity: gay targets rated more feminine than straight; Brazilian targets more feminine than Portuguese; no interaction. Salary: Brazilian candidates offered less than Portuguese; gay offered slightly more than straight; interaction showed straight Brazilians received the lowest salaries (M=1063) vs straight Portuguese (M=1278.33), gay Portuguese (M=1220), and gay Brazilians (M=1215), all p<0.001 (F(1,116)=20.59, p<0.001, η²=0.13). Warmth and salary were positively correlated (r=0.45, p<0.001); competence and salary remained positively correlated (r=0.26, p=0.004). Mega-analysis: CV type moderated effects. Straight Portuguese were perceived warmer in low- vs high-competence CVs; straight Brazilians were rated more competent in high- than low-competence CVs. Gay Portuguese and gay Brazilians were perceived more feminine in low- than high-competence conditions. Salary did not show a significant three-way interaction, though straight Brazilians’ lowest offers emerged in the low-competence experiment.

Findings show that when competence is triggered via individuating information, evaluations of warmth, competence, and salary proposals converge across nationality and sexual orientation, minimizing stereotyping in hiring-like judgments. In contrast, when competence cues are absent and judgments are more complex or ambiguous, stereotype-consistent patterns emerge: straight Brazilian applicants are viewed as less competent and receive the lowest salary proposals, whereas gay Brazilian applicants are insulated from these penalties, aligning with integrative/compartmentalized accounts of intersectional stereotyping wherein sexual orientation cues mitigate negative national-origin stereotypes. These effects mirror Portugal’s implicit prejudice patterns—explicit egalitarian norms but implicit pro-ingroup biases—and underscore the role of cognitive processing strategies: ambiguity encourages heuristic reliance on stereotypes; individuating competence information anchors judgments and reduces bias. The results contribute to debates on when multiple-minority status protects versus harms targets in organizational contexts and demonstrate the boundary condition of competence salience in a European, post-colonial setting.

The paper demonstrates that triggering competence can protect minority and multiple-minority applicants from hiring-related discrimination: under high-competence CVs, Portuguese vs Brazilian and straight vs gay men were rated similarly in warmth, competence, and salary. Under low-competence CVs, straight Brazilians faced competence and pay penalties, while gay Brazilians did not, suggesting that multiple-minority status can buffer against negative stereotypes tied to nationality in ambiguous evaluations. These findings extend intersectionality and SCM scholarship to the Portuguese context and highlight the practical value of emphasizing competence cues in selection processes. Future research should: (1) test with professional hiring managers and larger, more diverse samples; (2) experimentally trigger warmth as well as competence; (3) vary task complexity and information recall to unpack processing mechanisms; (4) examine additional intersections (e.g., race, SES, gender identity) and cross-cultural generalizability; and (5) evaluate interventions (e.g., structured CVs, debiasing training) that foreground individuating competence information.

Key limitations include: (1) inconsistent effect sizes across experiments, with small η² in Experiment 1 raising risks of false negatives/positives; (2) student samples and roleplay may limit external validity compared to professional decision-makers; (3) limited numbers of LGBTQ+ participants precluded robust moderator analyses by participant sexual orientation; (4) only competence was experimentally triggered, not warmth; (5) lower reliability for some scales in Experiment 2, likely due to increased ambiguity; (6) salary question order could have influenced stereotype activation; and (7) cultural and target specificity (nationality, not race) constrain generalizability to similar sociocultural contexts.