Issue knowledge and perceptions of agreement in the 2004 presidential general election.

Kenski, Kate ; Jamieson, Kathleen Hall

The importance of political information in the democratic process has long been debated. While some view political information as an indispensable component of a healthy democracy (Delli Carpini and Keeter 1996), others maintain that those with minimal political information can still use heuristics to make reasonable choices (Lupia and McCubbins 1998). In the context of this debate on the importance of political knowledge, we ask four questions about the 2004 presidential election: To what extent did voters in 2004 learn the presidential candidates' issue positions? Did those citizens with higher levels of political knowledge make different voting decisions from their less informed counterparts? On those issues where there was a mismatch between citizens' perceptions of agreement and actual agreement on the issues, could citizens have learned the candidates' actual issue positions from the presidential debates or the press? And, were candidate ambiguity and heuristic confusion possible culprits in the instances in which there was a mismatch?

An analysis of data from the National Annenberg Election Survey (NAES) shows that on the seven issue positions examined, citizen understanding of the presidential candidates' positions was relatively high compared to 2000. However, as in 2000, when people made mistakes in attributing issue positions to the candidates, the results helped Republican incumbent George W. Bush and hurt Democratic candidate John Kerry. Mistakes, in other words, were not randomly distributed between the candidates. Moreover, the more information citizens acquired, the more they tended to support Kerry over Bush, after controlling for six demographic variables as well as party identification and ideology. In short, political information matters. Content analyses indicate that citizens could have learned about the candidates' positions from the debates as well as press coverage.

Because U.S. citizens vote for candidates, not policy positions, understanding where the candidates stand on issues is important. Unsurprisingly, policy preferences influence, in part, how people vote (Page and Jones 1979). While the causal direction of the issue proximity and vote choice relationship has been the subject of some debate, it is clear that "issue proximities covary with voting behavior" (Brody and Page 1972, 457). When voters fail to see a difference between the candidates on an issue, the issue has no impact on the voting decision (Page and Brody 1972). Candidate ambiguity, therefore, inhibits policy voting.

Although voter knowledge levels are far from optimal (Delli Carpini and Keeter 1996; Waldman and Jamieson 2003), failures to detect differences between two candidates are not necessarily the fault of the voters. In addition to the limitations of humans' internal information-processing capacities (Miller 1956), four environmental factors may constrain citizens from perceiving issue differences between candidates. First, if the media environment does not provide significant coverage of an issue--either because reporters are not interested in the topic, cover it in ways that minimize learning (Cappella and Jamieson 1997), or do not see the issue featured in candidate communication-then media reports will not help voters see issue distinctions between the candidates. Second, if the candidates take clear stands on an issue but their stated issue positions seem similar to voters, then the issue will not receive prominent news treatment, and the similarity, even if noted in news and by voters, will not invite voter scrutiny. For example, Page and Brody (1972) explained the inability of the Vietnam War issue to drive voters toward a choice of either Richard Nixon or Hubert Humphrey as a byproduct of the similarity in these candidates' stated positions on Vietnam.

Third, if candidates agree on a policy goal (e.g., a prescription drug benefit) but disagree about the means of achieving it (e.g., mainly through private insurers or mainly through the government), the citizenry may have difficultly in distinguishing between the candidates. For example, Waldman and Jamieson (2003) attribute voter confusion in the 2000 election to the fact that in their general election ads, George W. Bush and Al Gore embraced similar goals (i.e., education reform, a prescription drug benefit for seniors, Social Security reform) but different means of achieving them. And fourth, if candidates take positions that conflict with party stereotypes, then citizens may have a difficult time learning those candidates' stances. For example, voters may have difficulty processing the issue stances of a Democrat who is pro-life or a Republican who is pro-choice on abortion because the stances violate stereotypic expectations of typical Democratic and Republican placement.

Candidates adopt ambiguous positions when their take on an issue differs from the center of public opinion and "when public opinion is spread across the spectrum" (Campbell 1983, 290). Ambiguity is not always the wisest strategy for a candidate to follow. From the candidate's perspective, the success of strategic use of ambiguity depends in part on the environmental context, that is, the extent to which the citizenry is risk averse or risk acceptant (Shepsle 1972). In a risk-averse environment, ambiguity can hurt a candidate's chances of winning. Yet, even if a candidate wants to clearly elucidate his or her policy stands, "there are limits on the number of messages the candidates can transmit or that the average voter can or will receive" (Page 1976, 748).

In addition to influencing vote choice, citizen learning is important to the democratic process because elections have effects beyond the vote. "Campaigns do more than choose leaders; each is a vehicle for the citizenry to assess the past and set an agenda for the future," argue Waldman and Jamieson (2003, 161). This essay builds on that work (2003), which found that public ignorance of the candidates' actual positions in 2000 was particularly damaging to Democratic nominee Al Gore because voters were unaware that their positions were closer to his than to those of Republican nominee George W. Bush.

To examine citizens' understanding of the presidential candidates, we analyzed post-election panel (second wave) data from the NAES collected between November 4 and December 28, 2004. The data were weighted to take into account household size and number of telephone lines into the residence and to adjust for variation in the sample relating to geographic region, sex, race, age, and education. Three types of questions were analyzed: perceptions of the candidates' issue positions on seven items, self-placement on the corresponding issues, and presidential vote choice.

Understanding of the Candidate Issue Positions

Table 1 presents the responses to seven candidate issue positions. Respondents were asked to name which candidate--Bush, Kerry, both, or neither--favored each. On average, the correct candidate was identified 71.0 percent of the time, suggesting that political knowledge acquisition was high overall. In 2000, citizens responding to the NAES only answered 57.0 percent of candidate knowledge items correctly, as shown in Table 2. (1) Identifying Kerry as the candidate who "favored eliminating tax breaks for overseas profits of American corporations and using the money to cut taxes for businesses that create jobs in the United States" gave respondents the most trouble, but well over a majority (58.2 percent) of citizens confronted with four choices (Bush, Kerry, both, or neither) nevertheless provided the correct answer. Importantly, on questions on which party heuristics forecast the candidates' positions, knowledge was high. Nearly 79 percent of respondents knew that Kerry was the candidate who "wanted to make additional stem cell lines from human embryos available for federally funded research on diseases like Parkinson's." Over 75 percent of citizens identified Bush as the candidate who favored "an amendment to the Constitution saying that NO state can allow two men to marry each other or two women to marry each other" (77.3 percent), "laws making it more difficult for a woman to get an abortion" (77.5 percent), and "making the recent tax cuts permanent" (75.3 percent).

Citizens' Lack of Knowledge Benefited Bush More Than Kerry

While knowledge overall was quite high, further analyses reveal that some citizens did make mistakes in identifying candidate issue positions, and these mistakes led to discrepancies between perceptions of agreement and actual agreement with the candidates. Table 3 shows that more people agreed with Kerry's actual issue positions than Bush's positions on five of the seven issues analyzed. The two issues on which people agreed with Bush more than Kerry were: "making the recent tax cuts permanent" and "allowing workers to invest some of their Social Security contributions in the stock market."

Figures 1 and 2 compare how much citizens perceived that they agreed with Bush and Kerry and how much they actually agreed with them. Perceived agreement was calculated by matching a respondent's self-placement on an issue with her assessment of where the candidate stood on that same issue. Respondents were asked whether they favored or opposed a set of issues positions. They were later asked about the same issues in the form of candidate knowledge questions. If a respondent reported, for example, that she favored "making it more difficult for a woman to get an abortion" and, when later asked which candidate (Bush, Kerry, both, or neither) favored "making it more difficult for a woman to get an abortion," said that Kerry favored that position, then the respondent was said to perceive agreement with Kerry on that issue. Note that perceived agreement is not synonymous with actual agreement. Because Kerry did not actually favor this position, the respondent would have been coded as being in perceived agreement but not actual agreement with Kerry on the abortion issue.

[FIGURES 1-2 OMITTED]

As shown in Figure 1, citizens' perceptions of agreement with Bush were higher than their actual levels of agreement on four issues: the constitutional amendment banning homosexual marriage, additional federal funding for stem cell research using human embryos, reimportation of drugs from Canada, and elimination of tax breaks for overseas profits of American corporations and use of such breaks domestically for job-creating corporations. (2) The discrepancy on perceived versus actual agreement was particularly high on the latter two, where the difference was 9.1 percent on drugs from Canada and 10.7 percent on corporate tax breaks. There were two issues on which people actually agreed with Bush more than they thought they did: making the recent federal tax cut permanent (9.3 percent difference) and allowing workers to invest some of their Social Security contributions in the stock market (11.1 percent difference).

The differences between perceived and actual agreement were more consequential for Kerry. (3) Figure 2 shows that there were only two issues on which people thought that they agreed with Kerry more than they actually did: "making the recent federal tax cut permanent" (5.4 percent difference) and not "allowing workers to invest some of their Social Security contributions in the stock market" (5.1 percent difference). On the remaining five issues, citizens actually agreed with Kerry more than they thought they did, and these discrepancies were statistically significant (as noted previously). The largest discrepancies were on reimporting drugs from Canada (18.7 percent) and eliminating tax breaks for overseas profits of American corporations (17.8 percent).

Does Political Knowledge Affect Vote Choice?

One might question whether political knowledge has much impact on vote choice when a respondent's party identification is taken into account. To that end, we ran a logistic regression model of the two-party vote (Bush = 1, Kerry = 0) to determine whether people with varying levels of political knowledge voted differently when demographic variables, party identification, and ideology were controlled. As shown in Table 4, party identification operated as expected. Republicans were significantly more likely to support Bush (p < .001), while Democrats were more likely to support Kerry (p < .001). Political knowledge was significantly associated with vote choice (p < .001). The greater one's knowledge, the less likely one was to vote for Bush. Each candidate issue knowledge item that was answered correctly was associated with a 16.8 percent decrease in the odds (4) of voting for Bush, after taking the effects of six demographic variables, party identification, and ideology into consideration. This roughly translated into a 4.6 percent decrease in the probability of voting for Bush for each one-unit increase in political knowledge. (5)

Could Voters Have Learned Candidate Issue Positions from the Presidential Debates or the Press?

Because an individual's understanding of the candidates' issue positions was significantly associated with vote choice in the 2004 presidential election, it is important to examine where citizens could have learned about the candidates. On those issues where perceived agreement differed from actual agreement by 10 percent or more, we investigated the extent to which citizens could have learned the candidates' issue positions from the three presidential and one vice presidential debates or could have learned the information in the newspaper. In fall 2004, three presidential debates (September 30, October 8, and October 13) and one vice presidential debate (October 5) were held. The sizes of the presidential debate audiences were larger in 2004 than in 2000 (Kenski and Stroud 2005). Transcripts from these debates (Commission on Presidential Debates 2004) were analyzed to determine whether the candidates articulated their issue positions on changing the recently passed Medicare prescription drug law to allow reimporting drugs from Canada, redeploying tax breaks for overseas profits of American corporations, and allowing workers to invest some of their Social Security contributions in the stock market. In addition, a Lexis-Nexis search on these issues was conducted to find out the extent to which they were covered in the Los Angeles Times, New York Times, and Washington Post between October 1 and November 1, 2004. (6)

Allowing Prescription Drugs from Canada

Allowing prescription drugs from Canada was a prominent issue in the Kerry campaign and discussed in his campaign advertising. Yet, after the campaign was over, over a third of citizens could not correctly place or differentiate the major party candidates' positions on this issue. Did citizens' lack of understanding stem from lack of discussion in the presidential debates or coverage in the press? Our analysis of the presidential debate transcripts and press coverage demonstrates that the presidential candidates' positions were covered frequently in both venues. The correct answer was available or strongly implied in two of the three presidential debates as well as in the vice presidential debate. Both candidates' issue positions on importing drugs from Canada were noted in two Los Angeles Times, three New York Times, and four Washington Post articles. In addition, there were four Los Angeles Times, eight New York Times, and seven Washington Post articles that articulated Bush's position but did not articulate Kerry's position. And, there were three New York Times and two Washington Post articles that noted Kerry's position on drugs from Canada but did not mention Bush's.

Kerry and his vice presidential running mate John Edwards noted their position on prescription drugs from Canada in three debates. In the vice presidential debate held on October 5, Edwards stated, "And we're also going to finally do something about the cost of prescription drugs. They've blocked allowing prescription drugs into this country from Canada. We're going to allow it." In the second presidential debate held on October 8, Kerry maintained, "I'm fighting to let you get those drugs from Canada, and I'm fighting to let Medicare survive." He also stated his position on this issue in the third presidential debate on October 13.

With Kerry and his running mate stating his position on prescription drugs from Canada in such high-profile events, what could have led to confusion of the candidates' positions on this issue? The confusion may have stemmed in part from the nuance in the president's position. On October 8, in the second presidential debate, conducted as a town hall format, Bush was asked specifically about his position on drugs from Canada from a citizen questioner, John Horstman.

Horstman: Mr. President, why did you block the reimportation of safer and inexpensive drugs from Canada, which would have cut 40 to 60 percent off of the cost?

Bush: I haven't yet. Just want to make sure they're safe. When a drug comes in from Canada, I want to make sure it cures you and doesn't kill you. And that's why the FDA and that's why the surgeon general are looking very carefully to make sure it can be done in a safe way. I've got an obligation to make sure our government does everything we can to protect you. And what my worry is is that, you know, it looks like it's from Canada, and it might be from a third world. And we've just got to make sure, before somebody thinks they're buying a product, that it works. And that's why we're doing what we're doing. Now, it may very well be here in December you'll hear me say, I think there's a safe way to do it.

The president's response left open the possibility that he would support reimportation from Canada in the future, provided that U.S. safety standards were met, which complicates the question of the extent to which Bush opposed reimporting drugs from Canada. Although the administration opposed the idea in the fall of 2004, Bush did not foreclose reimportation from Canada altogether. Candidate ambiguity was thus present.

The candidates' issue positions from the debates also were articulated in press coverage. The fact that Bush's position had become more ambiguous in the debate was noted in some of those stories. In the New York Times (October 17), for example, Robert Pear observed, "The Bush administration has consistently opposed efforts to allow imports of prescription drugs from Canada and other countries. But in the presidential debate on Oct. 8, Mr. Bush hinted that he might be open to the idea, if the government could somehow ensure the safety of imported drugs" (Section 1, p. 18).

The role of advertising in prompting clarifying news accounts is evident in an article published on October 17 (p. A06), where Washington Post staff writer Howard Kurtz reprinted the text of a Kerry ad that blamed Bush's policies for the flu vaccine shortage. That ad said:

 Three years ago, medical experts warned George Bush that a
 dangerous shortage loomed. Instead of fixing the problem,
 production of the vaccine was sent to a factory overseas-the
 vaccines were contaminated. Now Bush wants Canada to help,
 even though his own policies make it illegal for us to import
 medicine from Canada.

The U.S. flu vaccine shortage in the fall of 2004 provided a venue beyond debates and ads for the Kerry campaign to drive its position on prescription drugs from Canada into news space. A Washington Post article (October 19) noted,

 Although preoccupied with Iraq, Kerry continued to try to saddle
 Bush with the flu-vaccine shortage, accusing him of putting the
 interest of the drug companies above the needs of consumers by
 not developing a health plan that allows the importation of
 cheaper drugs from Canada. (Milbank and Romano 2004, A08)

So too did summary articles on candidate positions. An article entitled "Candidates' Proposals" in the Washington Post's front section on October 22 noted that Kerry "[w]ould allow legal importation of prescription drugs from other countries, such as Canada, and permit federal health officials to negotiate for bulk drug prices" (p. A06).

Eliminating Tax Breaks for Overseas Profits of American Corporations and Using Money for Tax Breaks for Corporations that Create Jobs at Home

Another issue central to Kerry's campaign was outsourcing of jobs and elimination of tax breaks for overseas profits of American corporations that Kerry argued engaged in such outsourcing. Kerry contended that the current tax code encouraged companies to send jobs overseas. The topic was emphasized by the Democratic presidential team in two of the three presidential debates as well as the vice presidential debate. On October 5, Edwards explained the Democrats' position, stating, "We believe in a strong middleclass in this country. That's why we have a plan to create jobs, getting rid of tax cuts for companies outsourcing your jobs; give tax cuts to companies that'll keep jobs here in America."

In the second and third presidential debates, Kerry explained his position on corporate tax cuts and outsourcing. In the town hall-style presidential debate on October 8, citizen Jane Barrow asked a question about the future of the American worker.

Barrow: Senator Kerry, how can the U.S. be competitive in manufacturing given--in manufacturing, excuse me--given the wage necessary and comfortably accepted for American workers to maintain the standard of living that they expect?

Kerry: Jane, there are a lot of ways to be competitive. And unfortunately again I regret this administration has not seized them and embraced them. Let me give you an example. There is a tax loophole right now. If you're a company in St. Louis working, trying to make jobs here, there is actually an incentive for you to go away. You get more money, you keep more of your taxes by going abroad. I'm going to shut that loophole, and I'm going to give the tax benefit to the companies that stay here in America to help make them more competitive. Secondly, we're going to create a manufacturing jobs credit and a new jobs credit for people to be able to help hire and be more competitive here in America.

Kerry addressed the issue again in the third presidential debate on October 13:

 Today, if you're an American business, you actually get a benefit
 for going overseas. You get to defer your taxes. So if you're
 looking at a competitive world, you say to yourself, "Hey, I do
 better overseas than I do here in America." That's not smart. I
 don't want American workers subsidizing the loss of their own job.
 And when I'm president, we're going to shut that loophole in a
 nanosecond and we're going to use that money to lower corporate
 tax rates in America for all corporations, 5 percent.

Despite the fact that the issue was central to Kerry's campaign and emphasized in the debates, citizens understood the candidates' positions on this issue the least of the seven issue knowledge items examined. Of citizens, 58.2 percent correctly identified Kerry as the candidate supporting redeploying tax breaks now given for overseas profits of American corporations to corporations that created jobs at home.

Compared to drugs from Canada, tax breaks for overseas profits received less coverage. There were no articles that fully explained both candidates' issue positions on the topic within the same article. However, Kerry's full position was explained in one Los Angeles Times article and one Washington Post article. While few articles articulated the position in full, explaining both parts of Kerry's plan (eliminating taxes for overseas profits of American corporations and using the money to cut taxes for businesses that create jobs in the United States), there were several articles that explained a part of the plan, thus allowing citizens to infer the candidates' positions. Four Los Angeles Times, five New York Times, and two Washington Post articles articulated pieces of the Kerry position. For example, writing for the New York Times on October 28, Jim Rutenberg explained, "Mr. Kerry has proposed closing a loophole that allows companies to defer paying taxes on profits earned overseas, although economists give mixed assessments of how much that would reducing the flow of work to other countries" (p. A25). One New York Times article partially explained Bush's position on the matter.

Where the ambiguity in Bush's forecast of possible future support for reimportation may have confused some about his position on that issue, here we have an issue that calls on conflicting heuristic inferences. An uninformed voter would expect that the Republican would defend tax breaks for corporations both in the United States and abroad. Using the party heuristic would prompt the correct inference that Kerry would want to eliminate the tax break for corporations abroad but would invite the incorrect inference when faced with determining whether he would reduce corporate taxes for those that create jobs at home. Resistance to the notion that a Democrat might favor corporate tax breaks in any setting may have pushed low-information voters to an incorrect answer here. Even when faced with Kerry's statements and press accounts of the Democratic position, those relying strongly on the party heuristic may have gotten this question wrong.

Investing Some Social Security Contributions in the Stock Market

The third issue on which some respondents were confused was allowing workers to invest some of their Social Security contributions in the stock market, an issue that was featured in the 2000 presidential campaign as well. In 2004, respondents actually agreed with Bush more than they thought they did by 11.1 percent. In 2004, around two thirds (66.9 percent) of citizens correctly identified Bush as the candidate who favored allowing workers to invest some of their Social Security contributions in the stock market. Although this issue has become a major focus of the second Bush term, it was discussed in only one presidential debate. In the October 13 debate, both candidates clearly addressed their positions on the issue. Bush stated,

 I believe that younger workers ought to be allowed to take some of
 their own money and put it in a personal savings account, because I
 understand that they need to get better rates of return than the
 rates of return being given in the current Social Security trust.
 And the compounding rate of interest effect will make it more
 likely that the Social Security system is solvent for our children
 and our grandchildren. I will work with Republicans and Democrats.
 It'll be a vital issue in my second term.

Kerry countered, "You just heard the president say that young people ought to be able to take money out of Social Security and put it in their own accounts. Now, my fellow Americans, that's an invitation to disaster." Kerry went on to state, "I will not privatize it. I will not cut the benefits. And we're going to be fiscally responsible. And we will take care of Social Security."

While the issue did not receive as much attention as drugs from Canada or tax breaks for overseas profits in the debates, the candidates' issue positions were widely discussed in the press. Two Los Angeles Times, five New York Times, and two Washington Post articles discussed both candidates' positions on Social Security contributions in the stock market. In addition, four Los Angeles Times, nine New York Times, and seven Washington Post articles noted Bush's position on Social Security in the stock market but did not mention Kerry's. (7) For example, in an October 19 New York Times piece, David E. Rosenbaum and David M. Halbringer explained that the candidates differed on Social Security, noting that "since his 2000 campaign, Mr. Bush has advocated allowing workers to put some of their Social Security tax money into personal retirement accounts that could be invested in the private markets" (p. A20).

Again there is a heuristic-based explanation for those who misattributed this position. Traditionally it has been Democrats, not Republicans, who have favored the creation and preservation of Social Security. Who would support "investing" some of individuals' Social Security contributions in the stock market? Perhaps some were misled by the words "investing" and "Social Security" and failed to hear the heuristic implications of the words "stock market."

Overall, however, the fact that these issues were prominently featured both in news and debates probably explains why in 2004 even on these issues, the numbers voicing the correct answer was high. In 2000, when asked where Bush and Gore stood on allowing workers to invest some of their Social Security contributions in the stock market, 63.0 percent of citizens correctly stated that Bush favored the position, while 60.1 percent correctly stated that Gore opposed the position. (8)

Did Debate Watching and Newspaper Reading Help Inform Citizens?

At the outset of this essay, we observed that if the media do not cover an issue, then citizens will not be able to learn how, if at all, the candidates differ on it. In 2004, the presidential debates and the press did cover those issues on which respondents were least informed. While the media did convey candidate information, not all citizens were necessarily exposed to that coverage. Had citizens paid attention to the presidential debates and newspaper coverage, would they have learned the candidates' issue positions? To address this question, we used logistic regressions to examine the relationships of debate viewing and newspaper reading to issue knowledge. Our dependent variables were the three knowledge items that gave citizens the most trouble. We controlled for education and party identification in our models. Education is often a strong predictor of political knowledge. Party identification was held constant because we assumed that identification with a major party would increase the likelihood that one would use party as a heuristic.

As shown in Table 5, watching at least one presidential or vice presidential debate and newspaper reading predicted knowing the answers to the three most challenging knowledge questions. Even when we controlled for education and party identification, watching a debate significantly increased a citizen's odds of knowing that Kerry was the candidate who favored reimporting drugs from Canada by 105.6 percent, that Kerry was the candidate who favored eliminating tax breaks for overseas profits of American corporations and using the money to cut taxes for businesses that create jobs in the United States by 41.5 percent, and that Bush was the candidate who favored allowing workers to invest some of their Social Security contributions in the stock market by 136.5 percent. This finding is consistent with that of Johnston, Hagen, and Jamieson (2004) that accurate placement of Gore on Social Security increased after the first and third presidential debates of 2000. The effect of newspaper reading on learning was statistically significant but much smaller than watching the debates. For each additional day that a respondent said she read the newspaper, the odds of knowing the answers to the issue questions increased by 2.3 percent, 2.7 percent, and 8.5 percent, respectively. From this analysis, we conclude that the confusion possibly created by misapplied party heuristics and strategic candidate ambiguity may be overcome by exposure to clarifying information in debates and print news.

Conclusion

Compared to the 2000 presidential election, citizens were much more informed about the issue positions taken by the major party presidential candidates in 2004. On average, citizens answered our candidate knowledge items correctly 71.0 percent of the time, surpassing the 2000 average by 14 percent. Information appears to have been consequential, as more informed voters tended to prefer Kerry over Bush after taking six demographic variables and party identification into account. Citizens' lack of knowledge benefited Bush more than Kerry. On five of seven issues, citizens' perceptions of agreement with Kerry were lower than their actual levels of agreement. On the three issues on which citizens performed the worst, the candidates' issue positions were articulated in at least one of the presidential debates. The issue positions were also covered in the Los Angeles Times, New York Times, and Washington Post. Moreover, citizens who watched at least one debate and/or read a newspaper were much more likely to know the answers to these issue knowledge items.

The notion that campaign communication helps voters make informed choices and informed choices produce predictable outcomes assumes that the major party candidates both run effective campaigns (Gelman and King 1993, 449; Holbrook 1996, 55). We suggest here that in 2004 as in 2000, the Democrats failed to persuade some voters that their views on some major issues were more aligned with the Democratic than the Republican ticket. We have offered a number of hypotheses to explain this disparity on the issues of reimporting drugs from Canada, eliminating tax breaks for overseas profits of American corporations and giving tax breaks at home, and investing some Social Security contributions in the stock market. On the issue of drugs from Canada, Bush took a nuanced position that perhaps made it more difficult for citizens to distinguish between the candidates. On the issues of tax breaks for overseas profits and Social Security in the stock market, the candidates took positions that may for some have operated against party stereotypes. These environmental factors may have interacted with citizens' limited cognitive processing capabilities. And for the 2004 election, these environmental constraints favored the Republican candidate.

AUTHORS' NOTE: We thank Miriam White, research and project assistant at the Annenberg Public Policy Center at the University of Pennsylvania, for her contributions to the content analysis presented in this article.

References

Allison, Paul D. 1991. Logistic regression using the SAS system: Theory and application. Cary, NC: SAS Institute.

Brody, Richard A., and Benjamin I. Page. 1972. Comment: The assessment of policy voting. American Political Science Review 66(2): 450-58.

Campbell, James E. 1983. Ambiguity in the issue positions of presidential candidates: A causal analysis. American Journal of Political Science 27(2): 284-93.

Candidates' proposals. 2004. Washington Post, October 22, p. A06.

Cappella, Joseph N., and Kathleen Hall Jamieson. 1997. Spiral of cynicism: The press and the public good. New York: Oxford University Press.

Commission on Presidential Debates. 2004, September 30. The first Bush-Kerry presidential debate. Retrieved September 17, 2005 from http://www.debates.org/pages/trans2004a.html.

Commission on Presidential Debates. 2004, October 5. The Cheney-Edwards vice presidential debate. Retrieved September 17, 2005 from http://www.debates.org/pages/trans2004b.html.

Commission on Presidential Debates. 2004, October 8. The second Bush-Kerry presidential debate. Retrieved September 17, 2005 from http://www.debates.org/pages/trans2004c.html.

Commission on Presidential Debates. 2004, October 13. The third Bush-Kerry presidential debate. Retrieved September 17, 2005 from http://www.debates.org/pages/trans2004d.html.

Delli Carpini, Michael X., and Scott Keeter. 1996. What Americans know about politics and why it matters. New Haven, CT: Yale University Press.

Gelman, Andrew, and Gary King. 1993. Why are American election polls so variable when votes are so predictable? British Journal of Political Science 23(4): 409-51.

Holbrook, Thomas M. 1996. Do campaigns matter? Thousand Oaks, CA: Sage.

Johnston, Richard, Michael G. Hagen, and Kathleen Hall Jamieson. 2004. The 2000 presidential election and the foundations of party politics. Cambridge, UK: Cambridge University Press.

Kenski, Kate. 2003. Testing political knowledge: Should knowledge questions use two response categories or four? International Journal of Public Opinion Research 15(2): 192-200.

Kenski, Kate, and Natalie Jomini Stroud. 2005. Who watches presidential debates? A comparative analysis of presidential debate viewing in 2000 and 2004. American Behavioral Scientist 49(2): 21328.

Kurtz, Howard. 2004. Kerry ad says flu vaccine shortage is typical of Bush's policy blunders. Washington Post, October 17, p. A06.

Lupia, Arthur, and Matthew D. McCubbins. 1998. The democratic dilemma: Can citizens learn what they need to know? New York: Cambridge University Press.

Milbank, Dana, and Lois Romano. 2004. Candidates spar on Iraq, terrorism war. Washington Post, October 19, p. A08.

Miller, George A. 1956. The magical number seven, plus or minus two: Some limits on our capacity of processing information. Psychological Review 63(2): 81-97.

Page, Benjamin I. 1976. The theory of political ambiguity. American Political Science Review 70(3): 74252.

Page, Benjamin I., and Richard A. Brody. 1972. Policy voting and the electoral process: The Vietnam War issue. American Political Science Review 66(3): 979-95.

Page, Benjamin I., and Calvin C. Jones. 1979. Reciprocal effects of policy preferences, party loyalties and the vote. American Political Science Review 73(4): 1071-89.

Pear, Robert. 2004. A.M.A. says government should negotiate on drugs. New York Times, October 17, Section 1, p. 18.

Rosenbaum, David E., and David M. Halbringer. 2004. Kerry goes beyond some of Bush positions. New York Times, October 19, p. A20.

Rutenberg, Jim. 2004. A list of topics aimed squarely at the middle class. New York Times, October 28, p. A25.

Shepsle, Kenneth A. 1972. The strategy of ambiguity: Uncertainty and electoral competition. American Political Science Review 66(2): 555-68.

Waldman, Paul, and Kathleen Hall Jamieson. 2003. Rhetorical convergence and issue knowledge in the 2000 presidential election. Presidential Studies Quarterly 33(1): 145-63.

KATE KENSKI

University of Arizona

KATHLEEN HALL JAMIESON

University of Pennsylvania

(1.) The 2000 NAES post-election knowledge battery used a separated candidate format rather than a combined format, unlike the 2004 survey, in which the combined format was used. The combined format presents a specific issue position to the respondents and then asks them to identify whether candidate A, candidate B, both, or neither favored that position. The separated format presents a candidate to the respondents and then asks them whether that candidate favored or opposed a given issue position. Given that the combined format is "a more demanding test of the voter" (Kenski 2003, 198), the comparatively high knowledge levels exhibited in 2004 are especially noteworthy.

(2.) The perceived agreement percentages were compared with the actual agreement percentages using proportion tests. Given the large sample size on which these analyses were based (n = 5,507), the perceived and actual agreement levels were statistically significant (p < .001) in all but two instances for Bush: support for a constitutional amendment banning homosexual marriages (p = .07) and abortion (n.s.). Considering that the 1.2 percent discrepancy between perceived and actual agreement with Bush on homosexual marriage almost reached conventional levels of statistical significance, it is important to keep in mind that statistical significance does not always connote substantive significance.

(3.) All perceived and actual agreement differences were statistically significant for Kerry (p < .001).

(4.) When one is working with a binary outcome and wants to control for a variety of predictors, logistic regression is often used for the analysis. The logistic regression model is one that predicts the odds of an event occurring rather than the probability of the event occurring. The "odds of an event is the ratio of the expected number of times that an event will occur to the expected number of times it will not occur" (Allison 1991, 11).

(5.) Allison (1991) uses the following formula to interpret the logit models in terms of probabilities: [beta][p.sub.i](1-[p.sub.i]), where p refers to the overall proportion of cases for which the event (i.e., voting for Bush) occurred and [beta] is the coefficient from the variable of interest (i.e., political knowledge).

(6.) Initially, articles were gathered for each topic using the same set of search terms for each newspaper. For reimporting drugs from Canada, the search terms were: drug AND Canada AND (Bush OR Kerry). For eliminating overseas tax breaks, the search terms were: (overseas OR foreign) AND (tax breaks OR tax cuts OR tax credits OR incentives OR loophole OR profit) AND (Bush OR Kerry). For Social Security in the stock market, the search terms were: Social Security AND (Bush OR Kerry). Articles were then analyzed to determine how many of them clearly articulated both candidates' issue positions, only Bush's positions, or only Kerry's positions (both, Bush only, and Kerry only were coded as mutually exclusive categories).

(7.) In addition, one Los Angeles Times and five Washington Post articles referred to Bush and private accounts. It is unclear whether the average reader would necessarily have understood that private accounts meant investment or retirement accounts as opposed to personal checking accounts. Readers with background knowledge could perhaps have inferred the answer to the Social Security knowledge question having read articles that mentioned private accounts.

(8.) As noted previously, in 2000, respondents were asked about each candidate separately.

Kate Kenski teaches political communication at the University of Arizona and was a member of the 2004 National Annenberg Election Survey team at the Annenberg Public Policy Center of the University of Pennsylvania.

Kathleen Hall Jamieson is the Elizabeth Ware Packard Professor of Communication at the Annenberg School for Communication of the University of Pennsylvania and Walter and Leonore Annenberg Director of its Annenberg Public Policy Center.

TABLE 1

Knowledge of Candidate Issue Positions in 2004 (%)

 Don't
 Know/
 Bush Kerry Both Neither Refused

Which candidate favored an 77.3 6.9 6.3 5.5 4.1
 amendment to the
 Constitution saying that NO
 state can allow two men to
 marry each other or two women
 to marry each other?
Who favored laws making it more 77.5 6.8 3.6 6.7 5.3
 difficult for a woman
 to get an abortion?
Which candidate wanted to make 8.0 78.8 4.0 3.7 5.6
 additional stem cell
 lines from human embryos
 available for federally funded
 research on diseases like
 Parkinson's?
Who favored making the recent 75.3 9.7 4.2 4.6 6.2
 tax cuts permanent?
Who favored changing the recently 13.6 62.9 5.7 7.9 9.9
 passed Medicare
 prescription drug law to allow
 reimporting drugs from Canada?
Who favored eliminating tax 18.2 58.2 7.1 5.9 10.7
 breaks for overseas profits of
 American corporations and using
 the money to cut taxes
 for businesses that create
 jobs in the United States?
Who favored allowing workers to 66.9 13.7 4.9 6.1 8.5
 invest some of their
 Social Security contributions
 in the stock market?

Note: Row percentages are given. Correct answers are in italics.
The percentages were based on weighted data from a sample of 5,507
citizens.

Source: NAES General Election Panel (second wave) November 4 to
December 28, 2004.

TABLE 2

Knowledge of Candidate Issue Positions in 2000 (%)

 Don't
What about (Insert Candidate Name)? Know/
Do You Think He Favored or Opposed ... Favors Opposes Refused

... giving patients the right to sue Bush 26.5 44.4 29.0
 their health maintenance Gore 53.1 20.5 26.4
 organization or HMO?
... making it harder for a woman to Bush 68.6 19.2 12.2
 get an abortion? Gore 21.7 67.6 10.7
... requiring a license for a person Bush 44.7 40.7 14.6
 to buy a handgun? Gore 71.0 16.6 12.4
... using government funds to make Bush 36.8 42.1 21.2
 sure that every child in the U.S. Gore 76.5 8.5 15.1
 is covered by health insurance?
... a ban on all soft money campaign Bush 32.8 38.7 28.5
 contributions? Gore 37.8 34.6 27.7
... the death penalty for some crimes? Bush 84.1 6.7 9.1
 Gore 38.1 41.9 20.0
... allowing homosexuals to serve Bush 14.3 60.7 25.0
 openly in the United States Gore 57.1 19.7 23.3
 military?
... using government money to help Bush 62.4 21.4 16.2
 some parents send their children to Gore 30.0 52.8 17.3
 private schools?
... allowing workers to invest some Bush 71.0 14.7 14.2
 of their Social Security Gore 17.5 65.4 17.2
 contributions in the stock market?
... selling some of the oil reserve Bush 27.3 48.6 24.1
 to increase the winter heating oil Gore 59.5 17.9 22.5
 supply?

Note: Row percentages are given. Correct answers are in italics.
The percentages were based on weighted data from random samples
of either 3,156 or 3,220 citizens.

Source: NAES General Election Panel (second wave) November 11
to December 7, 2000.

TABLE 3
Self-Placement on Issue Positions in 2004 (%)

 Neither Favor
 Favor Nor Oppose Oppose

An amendment to the Constitution saying 41.2 7.7 51.2
 that NO state can allow two men to
 marry each other or two women to marry
 each other
Laws making it more difficult for a woman 40.5 5.3 54.2
 to get an abortion
Make additional stem cell lines from human 64.7 5.3 30.1
 embryos available for federally funded
 research on diseases like Parkinson's
Making the recent tax cuts permanent 59.4 7.1 33.5
Changing the recently passed Medicare 78.3 5.7 16.0
 prescription drug law to allow
 reimporting drugs from Canada
Eliminating tax breaks for overseas 74.2 6.6 19.1
 profits of American corporations and
 using the money to cut taxes for
 businesses that create jobs in the
 United States
Allowing workers to invest some of their 55.7 6.1 38.2
 Social Security contributions in the
 stock market

Note: Row percentages are given. The percentages were based on
weighted data from samples with a minimum of 5507 citizens.
"Don't know" and refused as neutral.

Source: NAES General Election Panel (second wave)
December 28, 2004.

TABLE 4

Logistic Regression Model Predicting the Two-Party Vote

 B S.E. Sig. Exp(B)

Intercept 4.568 .433 .000 96.351
Gender (female = 1, male = 0) -.080 .106 .448 .923
Age (in years) -.007 .003 .035 .993
Hispanic (yes = 1, no = 0) .424 .195 .030 1.527
White (yes = 1, no = 0) 1.133 .155 .000 3.104
Education (in years) -.019 .025 .441 .981
Household income (in thousands) .002 .001 .155 1.002
Democrat (yes = 1, no = 0) -2.344 .129 .000 .096
Republican (yes = 1, no = 0) 2.023 .139 .000 7.564
Ideology (very conservative = -1.386 .071 .000 .250
 1 to very liberal = 5)
Political knowledge (0 to 7) -.184 .034 .000 .832

Percentage predicted correctly 88.0
Cox & Snell R-square .546
Nagelkerke R-square .728

Note: Data are weighted.

Source: NAES General Election Panel (second wave) November 4 to
December 28, 2004.

TABLE 5

Logistic Regression Models Predicting Correct Answers to Candidate
Issue Knowledge Questions

Who favored changing the recently passed Medicare prescription
drug law to allow reimporting drugs from Canada?

 B S.E. Sig. Exp(B)

Intercept -2.643 .191 .000 .071
Education (in years) .190 .013 .000 1.209
Democrat (yes = 1, no = 0) .229 .077 .003 1.257
Republican (yes = 1, no = 0) -.353 .074 .000 .703
Watched at least one debate .721 .081 .000 2.056
 (yes = 1, no = 0)
Newspaper reading (0 to 7 days) .023 .011 .029 1.023
Percentage predicted correctly 66.9
Cox & Snell R-square .078
Nagelkerke R-square .107

Who favored eliminating tax breaks for overseas profits of American
corporations and using the money to cut taxes for businesses that
create jobs in the United States?

 B S.E. Sig. Exp(B)

Intercept -2.087 .184 .000 .124
Education (in years) .159 .013 .000 1.173
Democrat (yes = 1, no = 0) .261 .074 .000 1.298
Republican (yes = 1, no = 0) -.527 .072 .000 .590
Watched at least one debate .347 .081 .000 1.415
 (yes = 1, no = 0)
Newspaper reading (0 to 7 days) .026 .010 .010 1.027
Percentage predicted correctly 63.2
Cox & Snell R-square .066
Nagelkerke R-square .089

Who favored allowing workers to insert Social Security contributions
the stock market?

 B S.E. Sig. Exp(B)

Intercept -2.384 .194 .000 .092
Education (in years) .157 .013 .000 1.170
Democrat (yes = 1, no = 0) -.056 .077 .470 .946
Republican (yes = 1, no = 0) .081 .077 .290 1.085
Watched at least one debate .861 .081 .000 2.365
 (yes = 1, no = 0)
Newspaper reading (0 to 7 days) .081 .011 .000 1.085
Percentage predicted correctly 69.1
Cox & Snell R-square .072
Nagelkerke R-square .101

Note: Data are weighted.

Source: NAES General Election Panel (second wave) November 4
to December 28, 2004.