**Principal Investigator(s):** Joshua Kalla University of California, Berkeley Email: [kalla@berkeley.edu][1] Home page: [http://polisci.berkeley.edu/people/person/joshua-kalla][2] Dawn Teele University of Pennsylvania Email: [teele@sas.upenn.edu][3] Home page: [http://www.dawnteele.com/dt/d.l.t..html][4] Frances Rosenbluth Yale University Email: [frances.rosenbluth@yale.edu][5] Home page: [http://politicalscience.yale.edu/people/frances-mccall-rosenbluth][6] Rachel Silbermann Yale University Email: [rachel.silbermann@yale.edu][7] Home page: [https://sites.google.com/site/rachelsilbermann/][8] **Sample size:** 2144 **Field period:** 10/21/2014-12/30/2014 **Abstract:** Despite making major inroads into national legislatures over the last 20 years, female politicians still do not attain parity either in national legislatures or in “P-Suite” positions, that is, as party leaders, president, prime minister, chair of powerful committees, or members of the premier’s cabinet. In this paper we investigate the determinants of low female advancement in politics, asking whether it is due to experience, discrimination, or preferences for particular skill sets. Drawing on an original survey of national legislators in 26 countries, politicians at all levels of government in the United States, and a sample of American voters, we use conjoint experiments to examine the barriers to promotion for women in power. Our results overwhelmingly show that sex-based discrimination does not account for the political glass ceiling. All else equal, most respondents prefer female candidates to male candidates. However, politicians and voters, especially men, discriminate among career profiles and age – preferring leaders who are younger and have professional as opposed to teaching backgrounds. Because female politicians tend to be older and are more likely to be educators, our results suggest that the gender gap in the attainment of political positions may be due to a subtle bias against characteristics that are, in practice, divided by gender. **Hypotheses:** H1: Faced with an opportunity to promote two people who are exactly the same on all dimensions but sex, the discrimination thesis predicts that the man is chosen over the woman. H2: Candidates who already have regional, federal, or executive political experience will be supported over those whose political experience is local or non-executive. H3: In choosing whom to promote into the highest political offices, politicians are biased toward particular skill sets that are associated with sex-based occupational sorting. They prefer male skill sets over female skill sets. H4: Men and women will reveal different preferences for promotion, with each more supportive of members of their own gender H5: Sex-based discrimination and preferences for male-based skill sets will depend on a respondent’s ideology. **Experimental Manipulations:** Conjoint experiment that manipulated: Gender, Age, Number of Children, Current Occupation, Number of Years in Politics, Spouse's Occupation. **Key Dependent Variables:** Randomly assigned election to be: City council, Congress, Governor. Outcome #1: Based on the limited information above, which of the two candidates would you be more likely to support in the ELECTION primary? Outcome #2: Please indicate the extent to which you feel favorable or unfavorable toward each of the two candidates. **Summary of Findings:** Our results overwhelmingly show that sex-based discrimination does not account for the political glass ceiling. All else equal, most respondents prefer female candidates to male candidates. However, politicians and voters, especially men, discriminate among career profiles and age – preferring leaders who are younger and have professional as opposed to teaching backgrounds. Because female politicians tend to be older and are more likely to be educators, our results suggest that the gender gap in the attainment of political positions may be due to a subtle bias against characteristics that are, in practice, divided by gender. **Additional Information:** This study included 3 samples: survey of global legislators, survey of American elected officials, TESS survey of American voters. **Findings from this project:** Teele, Dawn Langan, Joshua Kalla, and Frances Rosenbluth. "The ties that double bind: social roles and women's underrepresentation in politics." *American Political Science Review* 112.3 (2018): 525-541. [1]: mailto:kalla@berkeley.edu [2]: http://polisci.berkeley.edu/people/person/joshua-kalla [3]: mailto:teele@sas.upenn.edu [4]: http://www.dawnteele.com/dt/d.l.t..html [5]: mailto:frances.rosenbluth@yale.edu [6]: http://politicalscience.yale.edu/people/frances-mccall-rosenbluth [7]: mailto:rachel.silbermann@yale.edu [8]: https://sites.google.com/site/rachelsilbermann/