Gateways or hybrids: How do we actually perceive groups with dual identities?

Abstract

Studies of groups with dual identifications (who simultaneously belong to groups perceived as adversarial, for example) have shown that these groups can act as a gateway between the two groups they represent. If a person simultaneously belongs to perceiver’s ingroup and outgroup, perceiving them equally identified with both fosters positive attitudes towards outgroup. However, another line of research showed that dual identity is often recognized as a nonidentity, hybrid state, hence perceived as a threat to an ingroup identity. A potential cause of this gap can be the perception of compatibility between the two identities (bicultural identity integration (BII)). We hypothesized that varying the BII presentation would lead to a different perception of dually identified groups. In the experimental setting, participants read a bogus newspaper article describing children from Serb-Bosniak mixed marriages as being equally identified with both their parents’ ethnic groups, however perceiving them as (a) compatible (CID) or (b) incompatible (IID). In the control condition, participants read an article on an irrelevant topic. We measured if the groups differed in their perception of the dually identified group: (a) as a bridge between the two groups they represent and (b) as a potential traitor of the ingroup, both on 6-point Likert scale. Drawing from a convenient sample of 196 ethnic Serbs, aged 19–38 (M = 21.3(2.68)), we registered marginal main effect of the manipulation on perception of dual identity as a bridge (F(2, 193) = 2.752, p = .066, η2 = .028). Planned contrast test revealed that the effect was due to the difference between CID (M = 4.39, SD = 1.18) and IID (M = 3.98, SD = 0.97) (p = .026, 95% CI [0.029–0.715]); dual identity conditions together did not significantly differ from the control condition (p = .313, 90% CI [-0.488–0.161]). Similar pattern emerged for the perception of dual identity as a traitor (F(2, 193) = 3.038, p = .050, η2 = .031), where CID group scored significantly less (M = 1.58, SD = 1.12) than IID (M = 2.05, SD = 1.25) (p = .013, 95% CI [-0.881–-0.062]), with no significant difference between experimental and control group (p = .33, 90% CI [-0.523–0.184]). Our results confirm that identity integration leads to (non)acceptance of the dually identified groups. Strong dual identification itself is not always enough for a group to be a gateway group, and future studies should consider the perceived relations between the identities.