The sex of a congressional candidate can influence voting choices, but does candidate sex also influence the timing of those choices? This paper examines that question in light of other information that voters weigh in making their decisions. Using a national survey from the 2006 election, and a unique dataset of political informants, we find that the sex of the candidate conveys ideological information that permits voters to make swifter judgments. Additionally, it reduces the probability of a delayed decision by supplying information helpful to the choice between candidates—even in the absence of ideology. In fact, the impact of candidate sex rivals other variables that are traditionally used to explain the time-to-decision. Consistent with the literature on sex stereotypes, we find a stronger influence for Democratic than Republican female candidates.
This is a preview of subscription content, log in to check access.
Buy single article
Instant access to the full article PDF.
Tax calculation will be finalised during checkout.
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
Tax calculation will be finalised during checkout.
The exact proportion of campaign and late deciders is dependent on how these terms are defined. These numbers represent the percent of the electorate who had not reached a decision by the end of the conventions (see Table 1 Box-Steffensmeier and Kimball 1999). Late deciders, those who make their vote choice decision in the last 2 weeks of the campaign still represent a sizable proportion of the electorate, ranging from 9.8 % to 25.8 % from 1948 to 1996.
Surveys were conducted in the final weeks of the campaign across all districts, but individuals were not interviewed all on the same day. However, the time/date of the interview is not correlated with geographic region or district. Individuals interviewed on any given day are randomly distributed across all districts.
Our analysis utilizes the CCES pre-election vote intention question. Although there is a post-election questionnaire, over 20% of the pre-election sample did not participate in the post-election, due to the inability to contact and refusals. It remains unclear whether the non-respondents in the post-election survey were systematically different from the participants. Truncating the sample to only include post-election voters would significantly reduce our sample size. In general, our results remain robust when replicated using the constrained sample. The strength of the coefficients weakens slightly and the standard errors are inflated. However, the variables remain similarly-sized and signed and the major relationships are retained.
The CCES asked constituents to position the candidates ideologically on a 100-point scale, with lower values representing more liberal and higher values representing more conservative, political views. In addition, a 1,000–respondent subsample was asked to rate the candidates on a second ideological question using a seven-point scale. To make the constituent survey consistent with a set of informant-based ideology questions (described later in this section), which are also scaled on seven-points, we use an ordered logit to generate the predicted probability of falling into a given ideological category. The candidates are placed into one of seven categories, in which the candidate has the greatest predicted probability, based on the ordered logit.
Of the 12,976 respondents who received the vote decision question, 6,135 and 5,567 respondents could not place the Republican and Democratic candidate, respectively.
Measured as no high school (1) to post-graduate degree (6).
Ranges from 18 to 95.
Respondents were asked the partisan identification of their member of Congress, Governor and two Senators. The knowledge measure reflects the average number of questions answered correctly.
Individuals positioned themselves ideologically, on a 100-point scale and on a five-point scale. Because a 1,000-respondent subsample was asked to rate themselves on a third ideological question using a seven-point scale, we use an ordered logit to generate the predicted probability of falling into a given ideological category to make the scales comparable to the ones used in the informant survey (described later in this section).
The seven-point scale is also folded into an ideological strength variable, ranging from independent (0) to strong ideologue (3).
Ranging from strong Democrat (1) to strong Republican (7).
The seven-point partisanship measure is also folded into a partisan strength variable, varying from independent (0) to strong partisan (3).
Measured from strongly disapprove (1) to strongly approve (4).
In other specifications of our models we included controls for income (not shown), but have omitted this variable in the models presented due to missing data. Including the income variable reduces our number of observations by 1,400. The substantive interpretation of the models does not change regardless of whether the variable for income is included.
Our data include 109 races with two men, six races featuring two women and 40 mixed-sex races.
The response rate for the informant study was 21%.
All analyses reported in the paper include a weight variable that takes into account the size of the district sample, and clusters by district.
Expected vote is measured through a question that asks informants to rate the winning congressional candidate’s expected vote percentage on six point scale, ranging from 50% or less (1) to 91% or more (6). Using the midpoint for each category, we convert this variable into a simulated vote percentage.
We control for the candidates’ ideological distance to the constituent using a question that asks the informants to rate the candidates’ ideology on a seven-point scale, ranging from very liberal (1) to very conservative (7). We construct a measure of candidate proximity to the constituent by taking the absolute value of the difference between the constituent’s self-reported ideology subtracted by the informant-based candidate ideology score.
To capture valence, we tap the informants’ ratings of the candidates on seven items: personal integrity, ability to work well with others, competence, grasp of the issues, ability to find solutions to problems, qualifications to hold office, and overall strength as a public servant. Informants rated the candidates from extremely weak (1) to extremely strong (7).
The informant data also includes variables for campaign visibility, which allows us to test the hypothesis that the shorter decision times associated with women candidates result because women are covered differently in the news media than men. We included these variables in our models, however they did not have a significant influence on late-decisions, nor did they impact the direction, magnitude or significance of the other variables in the model. Importantly, campaign visibility appears to be uncorrelated with candidate sex, as the inclusion of the visibility variable does not attenuate the size, sign or significance of the sex variables. The presence of a female Democrat continues to erode the time-to-decision, independent of campaign visibility. Thus, we do not believe that the differential media coverage of women candidates that may occur during the course of an election accounts for the relationships we observe and report in our paper. Female Democratic candidates reduce decision times, regardless of the media coverage they attract. For the sake of parsimony, we omit these variables from the models.
Ranging from unshared (−1) to shared (1).
These data come from the Almanac of American Politics (Barone and Cohen 2006).
Previous research shows that women are more likely to run in winnable races (Fulton et al. 2006). Therefore, voters may arrive at quicker decisions when a female candidate is present because the district is already favorably predisposed to the female candidate. To see if district partisan predispositions pose an alternative explanation to the reduced time-to-decision evident when women candidates run, we re-run our models including a variable for district partisanship (measured as the Democratic presidential candidate’s two-party vote in the previous election). Our results suggest that district partisanship does not exert an independent effect on the time-to-decision. In addition, the inclusion of district partisanship does not alter the direction, magnitude or significance of any of the variables included in the model—most importantly, female Democratic candidate. Therefore, we do not think that women’s propensity to run in safe districts poses a plausible alternative explanation for our finding that the presence of a female Democrat reduces the time-to-decision.
In 2006, Republican vulnerability stemmed from voters’ disenchantment with the ongoing wars in Iraq and Afghanistan, the deficient government response to Hurricane Katrina, and a handful of high profile political scandals (Jack Abramoff, Mark Foley and Robert Ney). Democrats wound up gaining seven seats in the Senate and 32 in the House, granting them majority party status in both chambers for the first time in 14 years (Barone and Cohen 2008).
Overall, the informants and the voters perceived the candidates’ ideology similarly. Like the voters, the informants perceived male Democratic candidates to be significantly to the right of female Democratic candidates; but did not perceive any significant difference between male and female Republicans. Although the informants viewed female Democrats as being more liberal than the voters (2.321 vs. 2.538), this difference is not statistically significant.
We ran multivariate models treating the ability to place the candidates as a dependent variable. Regardless of specification, voters are much better able to place female Democratic candidates due to inferences about ideology, issue competency and personality traits. To keep the paper streamlined, we continue to report the bivariate results.
Theoretically, we believe that it is the presence of a female Democrat (regardless of the sex of the Republican) that conveys additional information to aid in the decision-making process. While we expect the presence of a woman Democratic candidate will reduce the time-to-decision compared to when both of the candidates are men, we do not expect that two women will further erode the time-to-decision any more than when one woman Democrat runs. As a robustness test, we experimented with specifications that included an interaction between Democratic female * Republican female, and all of the models remained stable. The inclusion of the interactive term for Democratic female * Republican female does not alter the size, sign or significance of the coefficients of interest. Moreover, the interactive term is insignificant. This is consistent with our expectation that it is the presence of a female Democrat that promotes swifter decisions, regardless of the sex of her competitor.
To illustrate these effects, Supporting Information C depicts the effect of the winning candidate’s expected vote and the probability of a late decision in open and incumbent dominated races based on Table 2, Model 1. All other variables in the model are held at their mean. The relationships depicted in Supporting Information C are consistent with our hypothesis that learning about the candidates is easier in more competitive contexts, where information about the candidates is rich, which helps voters arrive at a decision. Alternatively, Supporting Information C illustrates that individuals in the least competitive congressional districts receive a paucity of information about their alternatives, which causes them to delay their decision until more information becomes available. We also believe that constituents in uncompetitive congressional districts may lack energy and enthusiasm about the upcoming election, and therefore postpone learning about the candidates and committing to a decision. The gap between open and incumbent dominated seats depicted in Supporting Information C reflects our expectation that individuals are more likely to delay their decisions in open seat contests—where both candidates are unfamiliar—as opposed to incumbent dominated ones—where at least one of the candidates is recognizable. Again, this gap is evident when race competitiveness is held constant.
We analyzed the unit effect of candidate sex because it is inappropriate to directly interpret the effects of interaction terms from the results table. As Brambor et al. (2006) explain, the coefficients and significance of those coefficients traditionally reported in results tables are only correct when all the constituent terms are set to 0. They recommend calculating the predicted probabilities or the marginal effects (the first difference of the predicted probabilities) to assess the relationships of interest. Given the binary nature of our constituent variablesthe unit effect is the appropriate quantity of interest.
Alexander, D., & Anderson, K. (1993). Gender as a factor in the attribution of leadership traits. Political Research Quarterly, 46(3), 527–545.
Barone, M., & Cohen, R. E. (Eds.). (2006). The almanac of American politics. Washington, DC: National Journal Press.
Barone, M., & Cohen, R. E. (Eds.). (2008). The almanac of American politics. Washington, DC: National Journal Press.
Berelson, B. R., Lazarsfeld, P. F., & McPhee, W. N. (1954). Voting: A study of opinion formation in a presidential campaign. Chicago: University of Chicago Press.
Box-Steffensmeier, J. M. & Kimball, D. (1999). The timing of voting decisions in presidential campaigns. Paper presented at the Annual Meeting of the Midwest Political Science Association, Chicago.
Brambor, T., Clark, W. R., & Golder, M. (2006). Understanding interaction models: Improving empirical analyses. Political Analysis, 14(1), 63–82.
Burden, B. C. (2004). Candidate positioning in U.S. congressional elections. British Journal of Political Science, 34, 211–227.
Campbell, A., Converse, P., Miller, W., & Stokes, D. (1960). The American voter. New York: Wiley.
Center for American Women Politics (CAWP). (2010). Summary of women candidates for selected offices 1870–2010. http://www.cawp.rutgers.edu/fast_facts/elections/documents/can_histsum.pdf. Last Accessed 10/8/2012.
Chaffee, S. H., & Choe, S. Y. (1980). Time of decision and media use during the Ford-Carter campaign. Public Opinion Quarterly, 44(1), 53–69.
Chaffee, S., & Rimal, R. N. (1996). Time of vote decision and openness to persuasion. In D. Mutz, P. Sniderman & R. Brody (Eds.), Political persuasion and attitude change (pp. 267–291). Ann Arbor: University of Michigan Press.
Conover, P. J., & Feldman, S. (1989). Candidate perception in an ambiguous world: Campaigns, cues and inference processes. American Journal of Political Science, 33(4), 912–940.
Delli Carpini, M. X., & Keeter, S. (1996). What Americans know about politics and why it matters. New Haven: Yale University Press.
Dolan, K. (1997). Gender differences in support for women candidates: Is there a glass ceiling American politics? Women & Politics, 17(2), 27–41.
Dolan, K. (2004a). The impact of candidate sex on evaluations of candidates for the U.S. House of Representatives. Social Science Quarterly, 85(1), 206–217.
Dolan, K. (2004b). Voting for women. Boulder, CO: Westview.
Dolan, K., & Sanbonmatsu, K. (2009). Gender stereotypes and attitudes toward gender balance in government. American Politics Research, 37(3), 409–428.
Downs, A. (1957). An economic theory of democracy. New York: Harper Row.
Fournier, P., Nadeau, R., Blais, A., Gidengil, E., & Nevitte, N. (2004). Time-of-voting decision and susceptibility to campaign effects. Electoral Studies, 23(4), 661–681.
Fox, R. L., & Lawless, J. L. (2005). It takes a candidate: Why women don’t run for office. Cambridge, MA: Cambridge University Press.
Fulton, S. A. (2012). Running backwards and in high heels: The gendered quality gap and incumbent electoral success. Political Research Quarterly, 65(2), 303–314.
Fulton, S. A., Maestas, C. D., Maisel, L. S., & Stone, W. J. (2006). The sense of a woman: Gender, ambition and the decision to run for Congress. Political Research Quarterly, 59(2), 235–248.
Gertzog, I. N. (2002). Women’s changing pathways to the U.S. House of Representatives: Widows, elites and strategic politicians. In C. S. Rosenthal (Ed.), Women transforming Congress. Norman: University of Oklahoma Press.
Gopoian, J. D., & Hadjiharalambous, S. (1994). Late-deciding voters in presidential elections. Political Behavior, 16(1), 55–78.
Groseclose, T. (2001). A model of candidate location when one candidate has a valence advantage. American Journal of Political Science, 45(4), 862–886.
Hayes, D. (2005). Candidate qualities through a partisan lens: A theory of trait ownership. American Journal of Political Science, 49(4), 908–923.
Hayes, B. C., & McAllister, I. (1996). Marketing politics to voters: Late deciders in the 1992 British election. European Journal of Marketing, 30(10/11), 127–139.
Huddy, L., & Terkildsen, N. (1993). Gender stereotypes and the perception of male and female candidates. American Journal of Political Science, 37(1), 119–147.
Irwin, G. A., & Van Holsteyn, J. J. M. (2008). What are they waiting for? Strategic information for late deciding voters. International Journal of Public Opinion Research, 20(4), 483–493.
Kahn, K. F. (1994). Does gender make a difference? An experimental examination of sex stereotypes and press patterns in statewide campaigns. American Journal of Political Science, 38(1), 162–195.
Kahn, K. F. (1996). The political consequences of being a woman. New York: Columbia University Press.
Koch, J. W. (2000). Do citizens apply gender stereotypes to infer candidates’ ideological orientations? Journal of Politics, 62(2), 414–429.
Koch, J. W. (2002). Gender stereotypes and citizens’ impressions of House candidates’ ideological orientations. American Journal of Political Science, 46(2), 453–462.
Lewis-Beck, M. S., Norpoth, H., Jacoby, W. G., & Weisberg, H. F. (2008). The American voter revisited. Ann Arbor: University of Michigan Press.
Lodge, M., Steenbergen, M. R., & Brau, S. (1995). The responsive voter: Campaign information and the dynamics of candidate evaluations. American Political Science Review, 89(2), 309–326.
McDermott, M. L. (1997). Voting cues in low-information elections: Candidate gender as a social information variable in contemporary United States elections. American Journal of Political Science, 41(1), 270–283.
McDermott, M. L. (1998). Race and gender cues in low-information elections. Political Research Quarterly, 51(4), 895–918.
Mondak, J. J. (1995). Competence, integrity, and the electoral success of congressional incumbents. Journal of Politics, 57(4), 1043–1069.
Nir, L., & Druckman, J. N. (2007). Campaign mixed-message flows and timing of the vote decision. International Journal of Public Opinion Research, 20(3), 326–346.
Petrocik, J. R. (1996). Issue ownership in presidential elections, with a 1980 case study. American Journal of Political Science, 40(3), 825–850.
Popkin, S. L. (1991). The reasoning voter: Communication and persuasion in presidential campaigns. Chicago: University of Chicago Press.
Rahn, W. (1993). The role of partisan stereotypes in information processing about political candidates. American Journal of Political Science, 37(2), 472–496.
Sanbonmatsu, K. (2002). Gender stereotypes and vote choice. American Journal of Political Science, 46(1), 20–34.
Sanbonmatsu, K., & Dolan, K. (2009). Do gender stereotypes transcend party? Political Research Quarterly, 62(3), 485–494.
Schneider, M. C., & Bos, A. L. (2011). The interplay of party and gender stereotypes in evaluating political candidates. Manuscript presented to the New Research on Gender and Political Psychology Conference, New Brunswick, NJ.
Sides, J. (2006). The origins of campaign agendas. British Journal of Political Science, 36(3), 407–443.
Sniderman, P. M., Brody, R. A., & Tetlock, P. E. (1991). Reasoning and choice: Explorations in political psychology. Cambridge: Cambridge University Press.
Stokes, D. E. (1963). Spatial models of party competition. American Political Science Review, 57(2), 368–377.
Stokes, D. E. (1992). Valence politics. In E. Kavanaugh (Ed.), Electoral Politics. New York: Oxford University Press.
Stokes, D. E., & Miller, W. E. (1966). Party government and the saliency of Congress. A. Campbell et al. (Ed.), Elections and political order. New York: Wiley.
Stone, W. J., & Simas, E. N. (2010). Candidate valence and ideological positions in U.S. House elections. American Journal of Political Science, 54(2), 371–388.
Verba, S., Burns, N., & Schlozman, K. L. (1997). Knowing and caring about politics: Gender and political engagement. Journal of Politics, 59(4), 1051–1072.
Whitney, D. C., & Goldman, S. B. (1985). Media use and time of vote decision: A study of the 1980 presidential election. Communication Research, 12(4), 511–529.
Wright, J. R., & Nemi, R. G. (1983). Perceptions of candidates’ issue positions. Political Behavior, 5(2), 209–224.
We are grateful to Walter J. Stone for his guidance on this paper and for generously providing the data used in this analysis and Susan Welch for her insightful comments that helped shape the paper.
Electronic supplementary material
Below is the link to the electronic supplementary material.
About this article
Cite this article
Fulton, S.A., Ondercin, H.L. Does Sex Encourage Commitment? The Impact of Candidate Choices on the Time-to-Decision. Polit Behav 35, 665–686 (2013). https://doi.org/10.1007/s11109-012-9214-0
- Candidate sex
- Sex stereotypes
- Voting behavior
- Women in politics