A growing body of work on candidate traits shows that people with a given social characteristic tend to prefer candidates or leaders who share that characteristic (Campbell and Cowley 2014; Cutler 2002). However, the existing evidence for whether women vote for women is mixed. For example, Kathleen Dolan found that candidate sex was a driver of voting behavior for the U.S. House of Representatives in 1992, but not in 1994 or 1996 (Dolan 1998, 2001, 2004). Eric Smith and Richard Fox used pooled U.S. data from 1988 to 1992 and found that well-educated women were more inclined to support women candidates in House but not Senate races (Smith and Fox 2001), and others have found that women are more likely to vote for women candidates only when they are perceived as being pro-feminist (Plutzer and Zipp 1996). By contrast Fulton (2014) found that women are not more likely to vote for women candidates in the United States, but that male Independents are somewhat less likely to vote for them. Others have found little evidence whatsoever of an association between candidate gender and vote choice (McElroy and Marsh 2010).