Are people more prejudiced towards groups that are perceived as coherent? A meta‐analysis of the relationship between out‐group entitativity and prejudice

Abstract

A meta-analysis was conducted to examine the relationship between out-group entitativity and prejudice. A quantitative analysis of 85 effect sizes from 33 independent samples showed a significant positive relationship between entitativity and prejudice (Fisher's z = .414, 95% CI [0.272, 0.557], p < .0001). Three possible moderators of the relationship between entitativity and prejudice were tested: conceptualization of the entitativity (essence-based entitativity scale, agency-based entitativity scale, common entitativity scale), the target of the prejudice, and the measures of prejudice (attitudes, emotions, behaviour towards out-group). Results demonstrated that out-group entitativity correlated with prejudice only when entitativity was conceptualized as an essence-based or common-based scale, and prejudice was measured as the attitude to the out-group. The target of prejudice does not moderate the relationship between entitativity and prejudice.

Modeling stereotypes and negative self‐stereotypes as a function of interactions among groups with power asymmetries

“…It is generally assumed that these tendencies would explain the development of positive stereotypes about the ingroup, negative stereotypes about the outgroup, and discrimination among social groups (Abrams & Hogg, ; Scandroglio, López Martínez, & San José Sebastián, ; Tajfel & Turner, /2001). Importantly for our current work, it has been found that negative attitudes towards the outgroup depend on it being perceived as a homogeneous entity (i.e., entitativity; Agadullina & Lovakov, ), which is consistent with the view that stereotypes are conceptual (Macrae & Bodenhausen, ), and not merely attitudinal phenomena.…”

Air Pollution, State Anxiety, and Unethical Behavior: A Meta-Analytic Review

“…To account for statistical dependencies, we further conducted PET and PEESE meta-regression with RVE using the R package robumeta (Fisher et al, 2016). “Because PET underestimates nonzero effects and PEESE overestimates null effects” (Agadullina & Lovakov, 2018, p. 712), a two-step conditional PET-PEESE procedure is recommended: If PET finds a significant effect, then the PEESE estimate is preferred; if PET does not find a significant effect, then the PET estimate is preferred (Agadullina & Lovakov, 2018; Stanley & Doucouliagos, 2014). Analyses revealed that both PET, b = 3.98, SE = 1.86, t (5.32) = 2.14, p = .08, and PEESE, b = 25.64, SE = 15.64, t (1.69) = 1.64, p = .26, were not significant.…”

“…Because the studies used different analytical strategies, we converted all effect sizes to the same metric, Pearson’s r , and applied Fisher’s z transformation. When r was unavailable but the standardized regression coefficient (β) was available, we followed the commonly used β-to- r imputation formula: r = β + .05λ, where λ equals 1 when β is nonnegative, and λ equals 0 when β is negative (Peterson & Brown, 2005; for a recent meta-analysis example, see Agadullina & Lovakov, 2018, p. 711). For comparison purposes, we present effect sizes that account for population as a key covariate (e.g., crime rate = number of crimes divided by population).…”

“…However, this approach results in a loss of information and fails to account for statistical dependency across effect sizes (Friese et al, 2017). Therefore, we used robust variance estimation (RVE; Hedges, Tipton, & Johnson, 2010) in the R package robumeta (Fisher, Tipton, & Hou, 2016), which has been used in recent meta-analyses (Agadullina & Lovakov, 2018; Bediou et al, 2018; Friese et al, 2017; Kurdi et al, 2019).…”

Group cohesion benefits individuals who express prejudice, but harms their group

“…Finally, our results provide new evidence of how entitativity can exacerbate intergroup conflict. Whereas much work shows that viewing a group as entitative can stoke stereotyping and prejudice (Agadullina & Lovakov, 2018;Brewer & Harasty, 1996;Er-rafiy & Brauer, 2013;Spencer-Rodgers, Hamilton, & Sherman, 2007), the present work shows it can license individual members of that group to express prejudice against others.…”