Identities, interest group coalitions, and intergroup relations

ABSTRACT

Interest groups are well known for lobbying and providing information to citizens. However, no extant scholarship explores how the actions of interest groups affect intergroup relations, even though these organizations represent a variety of social identities. I argue that the decisions of interest group leaders to work together or reject collaboration send signals to everyday members of the identity groups they represent about their relations to other groups. In a survey experiment with a nationally representative sample of African-Americans, I vary whether respondents receive information about a successful coalition or a rejected coalition between African-American interest groups and organizations representing another identity. I find that when African-American interest groups successfully form a coalition with a high-solidarity outgroup (e.g., Hispanics), individuals develop greater feelings of closeness with the outgroup and express greater support for policies that benefit that group. However, when leaders of the outgroup organizations reject the coalition, it creates a backlash effect of lower closeness and weaker policy support. This backlash effect does not occur for low-solidarity outgroups (e.g., atheists). These findings suggest that interest groups are an understudied source of elite influence on identity-based perceptions, which can either promote or obstruct harmonious intergroup relationships.

KEYWORDS:

• Identity
• interest groups
• intergroup relations
• public opinion
• political behavior
• political psychology
• race

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Terms such as closeness and commonality are sometimes used synonymously with others such as “superordinate identity” or “common identity” – i.e., a broader, overarching identity comprised of multiple identity groups (Craig and Richeson 2016; Gaertner et al. 1993; Wenzel, Mummendey, and Waldzus 2007). It is possible that interest group coalitions may create such superordinate identities. However, given that the boundaries between intergroup closeness and common identities are not always clear, I focus here on closeness and reserve the exploration of superordinate identities for further research.

2 https://www.beforthright.com/for-researchers.

3 Because I conduct many related tests, I include in the appendix a correction for multiple hypothesis testing. These adjusted results offer conclusions that are nearly identical to those I present here. Moreover, I offer alternate analyses utilizing interaction-based regression models in Table A.6 in the appendix (see Supplemental data). The conclusions of these analyses do not differ substantially from those presented here.

4 Although all respondents completed all outcome variables with respect to each of the three outgroups, I only present data from the “coalition” and “rejected coalition” treatment groups that correspond to the outgroups in their respective vignettes. For example, I do not analyze attitudes toward Hispanics among the coalition atheist group, nor do I present results regarding attitudes about individuals with disabilities in the rejected coalition Hispanic condition. A complete set of outcome variable means for all conditions is located in the appendix.

5 In the appendix, I present two additional series of results: the first comparing each treatment group to the working alone condition instead of the control, and the second pooling the control and working alone conditions as the comparison point. These tests produce largely similar results with differences in the expected directions, although the atheist conditions often do not differ significantly from the working alone group itself.

Additional information

Funding

This work was supported by The Graduate School at Northwestern University.