Holly Henrich and Dr. Chris Karpowitz, Political Science Department
Currently, the United States ranks as the 86th country in women’s representation in government,with 18.2 percent of seats in the House of Representatives and 20 percent representation in the Senate (Inter-Parliamentary Union, 2014). With a nation that is half comprised of women, questions are raised as to why women have historically been underrepresented in government. More coverage on the physical appearances and fashion choices of female political candidates has far exceeded the coverage given to their male counterparts (Rosholm, 2009). How female political candidates are initially perceived based on their physical attributes may be a contributing factor as to why there is little representation of women in government. Political scientist, Kathleen Hall Jamieson, discusses the relationship between femininity and competence in regards to women, claiming that these two characteristics serve as a double bind for women, particularly in the workplace. According to the theory, women that express high levels of femininity are perceived as less competent. Likewise, women that display low levels of femininity are also perceived as less competent (Jamieson, 1995). Achieving the perfect balance of femininity appears to be an important factor for how women are viewed by their peers. This research offers insight regarding the question of how differing levels of both femininity and perceived competence can affect attitudes about female political candidates.
This research involved an experimental approach using survey methods to gather results about voter attitudes. Four different conditions were created to modify the levels of femininity and competence between high and low in a female model used to project a female political candidate. Manipulating femininity and competence levels was done by adjusting makeup, dress attire, hair style, and items included in each condition such as visible degrees, candy bowls, and magazines. Each of these four conditions was paired with the same photo of a male model used to project the standard white male political candidate. After the four treatment types were created, we created a survey that randomized the selection of the conditions. The survey included questions regarding voting choice between the two candidates, character attributes associated with the two photos, (i.e. who is more trustworthy? likable? outgoing?), along with general questions about the participants’ political background. On August 21, 2014, the survey was launched using Amazon’s Mechanical Turk, an online resource for survey research, with 25 cents compensation for completing the survey.
A total of 516 participants completed all questions of the survey, with all surveys completed within the day of hosting the survey. The characteristics of the respondents did not correspond with a representative sample of the country as a whole, as is common with MTurk samples. The respondent group tended to be more liberal, younger (average age of 34) and skewed by gender with about 2/3 of respondents being male. These factors did not pose a particular problem for this research, however, because the emphasis was on how these voter choices carried across conditions. The main results hold even with controls for these characteristics.
Overall, the female candidate was more popular than the male candidate across the conditions. 71 percent of participants in the High Competence/Low Femininity condition preferred the female, whereas 56 percent of respondents in the High competence/High femininity condition selected the female. This difference was significant at a p=.02, indicating that respondents strongly preferred a less feminine female candidate with indicators of high levels of competence over a female candidate with both high femininity and competence levels. There were also important differences noted by the gender and partisanship of the respondents. Female respondents tended to be more enthusiastic about the female candidate, particularly the candidate with high competence/low femininity. In nearly every condition this difference between men and women was highly significant. Self identifying and leaning Republicans were most enthusiastic about female candidates projecting high femininity and low competency, and least enthusiastic about highly competent female candidates. Democrats were most enthusiastic about the High Competence/Low Femininity condition. In all conditions except Low competence/High Femininity, Democratic selection of the female candidate significantly exceeded that of Republican selection. In regards to character attributes, respondents tended to select the female candidate to be more likeable and outgoing, regardless of the condition. In the High Competence/ Low Femininity condition, the female candidate was more likely to be seen as intelligent and trustworthy.
The results of this research indicates that the heuristic of physical appearance is significant for female candidates, particularly because it influences different types of voters in different ways. Manipulation of femininity and competency suggests that things like amount of makeup and dress attire can be contributing factors in voter choice and attitudes about women. As indicated by the results, men and women respond differently to varying representations of women in politics, as do individuals identifying across the political spectrum. Because of this, it is of particular importance for female political candidates to seek a balance in physical appearance regarding femininity and competency when approaching different voting groups.
How one projects oneself during political campaigns is an important contributing factor to success rate. Because of socializing agents shaping the role women play in society, women face an uphill battle in regards to representation in the political arena. This study provided insight for some of these challenges.
Inter-Parliamentary Union. 2014. World Classification prepared by the Inter-Parliamentary Union. http://www.ipu.org/wmn-e/classif.htm. (Accessed September 8, 2014).
Rosholm, Joanna S. 2009. The Politics of Fashion: American Leaders and Image Perception. Washington, D.C. http://search.proquest.com.erl.lib.byu.edu/docview/ 304885469/1416C8A4B757867C458/6?accountid=4488. (Accessed October 29, 2013).