Download data and study materials from OSF
Miami University of Ohio
Sample size: 1017
Field period: 03/24/2013-08/02/2013
Campaigns increasingly use campaign materials that invoke an identity appeal to specific groups of people, including women. Yet, we know very little about how women respond to these appeals, the causal mechanism driving responses, or whether male and female candidates can use these ads with equal effectiveness. We argue that voters will support a candidate who uses an identity-based advertisement because the ad primes identity. Using an experimental design that provides women with an identity-based appeal, our findings suggest that female candidates, but not male candidates, are able to prime female voters' gender identity. Male candidates' appeals are successful because they persuade female voters of their positive traits. We discuss the implications of these findings for descriptive representation.
What are the effects on female voters of a message targeted towards them based on appeals to their identity? How do these effects differ if the appeal is from a male or a female candidate?
H1: Among women, the likelihood of voting for a candidate with an identity-based targeted message will be greater than the likelihood of voting for a candidate with an issue-based message (H1).
H2: An identity-based message from an outgroup source, i.e. a male candidate, will be more successful than an issue-based message. At the same time, we expect that an identity-based message from an ingroup member (a female candidate) will be more successful than an issue-based message from an outgroup member.
H3: An identity-based message from a female candidate will prime group-based social identity, increasing vote support among women. However, an identity-based message from a male candidate will not similarly prime group-based social identity among women.
Male candidate, identity-targeted ad; Female candidate, identity-targeted ad; Male candidate, issue-targeted ad; Female candidate, issue-targeted ad. We developed a template of the website ascribed to "Congresswoman/man Patricia/Tom Johnson," and varied the content from an identity-targeted appeal to women (Treatment) to an issue-targeted appeal on transportation (Control). The identity-targeted appeal used the frame of domestic violence and actions by the Congresswomen/man to limit violence against women. Both sets of content resemble actual website content from a congressional campaign.
The subjects answered a series of questions about their impression of the candidate, their feelings towards a set of groups, and their attitudes about policies. As a manipulation check, we asked the respondents to identify which issue (violence against women, transportation, or healthcare) was featured on the website. Our key variable of interest is a vote variable where respondents rated how likely they would be to vote for the candidate, on a five-point scale, from very unlikely (1) to very likely (5).
We test the effects of identity-based messages targeted towards women from both female and male candidates, among female voters. We find that these ads are effective among female voters. However, not all of our hypotheses were supported. We predicted that the identity-based message from the female candidate would be more effective than the male candidate's message, yet the treatments were equally as effective in responses from female voters.
We find alternative mechanisms at work among the female responses to the male and female candidates. Identity-based appeals from the female candidates prime identity and the candidate's ability to handle violence against women policies, while appeals from the male candidate simply persuade the female respondents that the male candidate is a stronger leader and more compassionate. Thus, it seems as though only female candidates can prime identity. Female candidates should be able to attract the attention of female voters because of their shared identity and interests, which presents an opportunity to encourage women to vote for women.
We have a second paper that examines the effect of targeted messages on the outgroup (in this case, men). Results available from the author.
Holman, Mirya R., Monica C. Schneider, and Kristin Pondel. 2015. "Gender Targeting in Political Advertisements." Political Research Quarterly 68(4): 816-829.
"Implicit and Explicit Gender Targeting in Political Advertisements" presented at the 2013 Southern Political Science Association Meeting in Orlando, FL. Winner of the Marion Irish Outstanding Women and Politics Paper at SPSA 2013.
"Implicit and Explicit Gender Targeting in Political Advertisements" presented at the 2013 Midwest Political Science Association Meeting in Chicago, IL.