Clips of Sarah Palin's interviews with Katie Couric have generated lots of buzz about whether Palin is sufficiently well-informed about national and international affairs to be an effective vice president. Palin fans will be tuning in to tomorrow night's vice presidential debate eager to see her allay those doubts, while skeptics will be viewing in much the same spirit as the people who watch NASCAR races hoping to witness a crash.
But do voters really care how much Palin doesn't seem to know? After all, similar concerns were raised about George W. Bush. Does anyone remember how many world leaders he couldn't name in 2000? Nonetheless, Bush managed to get elected and reelected--and he was at the top of the ticket, not the understudy. While voters naturally prefer knowledgeable candidates to ignorant ones, it is not something they seem to care a lot about.
Surveys conducted by the National Election Study team in each presidential year since 1980 have asked prospective voters to rate the presidential candidates on a variety of specific traits. In 2000, for example, survey respondents were asked how well the phrases "moral," "really cares about people like you," "knowledgeable," and "provides strong leadership" described Bush and Al Gore. The biggest difference in perceptions of the two candidates was that people saw Bush as considerably less "knowledgeable" than Gore -- by 11 points on a 100-point scale. (They also saw Bush as less empathetic, but most considered him a stronger leader.)
How much did that matter? My analysis suggests that an undecided voter who saw Bush as 11 points less "knowledgeable" than Gore was only about 1.3% less likely to vote for Bush as a result. Comparable differences on the other trait dimensions were three to five times as consequential. Clearly, voters in 2000 were much more concerned about electing someone who was strong, empathetic, and moral, with "knowledgeable" a distant fourth. And they weren't just giving the genial anti-intellectual Bush a pass -- much the same pattern has held in other recent elections.
These estimates of the impact of trait perceptions allow for the fact that each party's loyalists are very likely to see their own candidate as superior in every way. For example, very conservative Republicans in 2000 saw Bush as being about 20 points more knowledgeable than Gore on the 100-point scale, while very liberal Democrats saw Gore as being about 40 points more knowledgeable. Those differences suggest that, regardless of what Governor Palin actually says tomorrow night, fans and skeptics will both find plenty of ammunition to support their preconceptions. But the people who matter -- the voters whose minds are still not made up -- will mostly not care whether Palin can rattle off the names of Supreme Court cases or world leaders. If she comes across as strong and empathetic, that may be enough.
Larry M. Bartels directs the Center for the Study of Democratic Politics in Princeton University's Woodrow Wilson School of Public and International Affairs. He is the author of Unequal Democracy: The Political Economy of the New Gilded Age.