The Forum. Manuscript Voting vs. Thinking: Unified Partisan Voting Does Not Imply Unified Partisan Beliefs

Transcription

1 The Forum Manuscript 1516 Voting vs. Thinking: Unified Partisan Voting Does Not Imply Unified Partisan Beliefs Jeremy C. Pope, Brigham Young University 2012 De Gruyter. All rights reserved.

2 Voting vs. Thinking: Unified Partisan Voting Does Not Imply Unified Partisan Beliefs Jeremy C. Pope Abstract The Myth of the Independent Voter crystallized views that were becoming accepted in political science and began an effort to convince the press that self-described independent voters were not truly independent, but that these leaners were really closet partisans. Despite the success of that book, its argument has come with a cost. Most especially, a misrepresentation of the argument can play into the myth of a deeply polarized electorate. When scrutinized closely, it becomes clear that Americans often vote in a highly partisan way, yet they are far less likely to think in a clearly partisan fashion. All partisans, but especially weak partisans and independent leaners, disagree frequently with their respective parties. Indeed, just as we no longer see independents as a united block, we should avoid making claims that partisans are united, because even though these partisans often vote alike, they do not always think alike. KEYWORDS: independent voter, unified partisan voting Author Notes: Jeremy C. Pope is an Associate Professor of Political Science at Brigham Young University and a Research Fellow with the Center for the Study of Elections and Democracy. His interests include elections, public opinion, legislatures, political parties and the American founding.

3 Pope: Voting vs. Thinking 1 Political scientists sometimes do not reflect enough on just how successful the discipline has been at educating the public and the press. There is obviously much work to do as journalists and commentators continue to perpetuate myths that do not reflect reality (Noel 2010). Nevertheless there are successes, like reminding people of the efficacy of negative ads (Geer 2006), the fact that money cannot buy an election (Ansolabehere, Figueiredo and Snyder 2003), or that election fundamentals are crucial for studying the campaign (Nyhan and Sides 2011). Despite these successes, there are occasions where strengths can become weaknesses, and one element of our collective success exhibits that very pattern. Introducing the Problem In the 1980s, political discussion was awash with claims about the rise of independents. Political scientists knew that the claim of independence was at best, exaggerated and at worst fraudulent: [Leaners] are never neutral and the extent of their [partisan] affect almost invariably resembles that of weak partisans (Keith, Magleby, Nelson, Orr, Westlye and Wolfinger 1992, 70, emphasis in original). In other words, leaners are not true independents but closet partisans. This finding has become so widely accepted in political science that it is difficult to overstate its importance. When the media and political scientists speak about partisans now, they often explicitly include leaners in their accounting. (Though there are exceptions; see Magleby, Nelson, and Westlye 2012). Despite some dissenters, few if any of the efforts of political scientists to correct media misperceptions have been as successful. Success has come, however, with a hidden cost. Media accounts often overstate the argument that is central to The Myth of the Independent Voter: that there are few true independents because most people vote according to an underlying partisanship. The emphasis on that term vote is important, because the claim was never that partisans are divided into two camps that consistently think differently. In chapter seven, the authors of The Myth of the Independent Voter went looking for issues that showed why the number of independents had grown in size, but they conclude by saying they [acknowledged] a failure to locate the issues that led to the rise of independents (168). That was because, with a few exceptions noted below, they found that on many public policy questions, there was little evidence for a correlation between issue attitudes and partisanship. People in the different categories may have been voting for different candidates (and parties), but each category had a diverse range of opinion on many questions. Put differently, the evidence for partisan voting does not imply two united camps of partisans, marching in lockstep, always thinking alike, always taking the same position, always in agreement with their party.

4 2 Submission to The Forum The irony is that it is often activists within the parties who peddle the claim that there are two camps of united voters: Democrats voting and thinking as a block and Republicans voting and thinking in opposition. Markos Moulitsas, the proprietor of Daily Kos, wrote earlier this year that [d]espite repeated efforts by Beltway hacks to appeal to a mythical and nonexistent bipartisan middle, it s clear there is zero appetite for such constructs from the American public (Moulitsas (2012). The right has its own version of this argument. In a piece titled Two Tickets: Two Americas, Stanley Kurtz, writing in National Review, argues that we need to understand that our political divisions are real and growing. They are rooted not in top-down political rhetoric but in profound and lasting social and cultural differences (Kurtz 2012). The meme makes its way into all kinds of places. For activists to offer the image that the public is divided into two united partisan camps is, perhaps, a sensible strategy. They can try to convince moderates or the uninterested that the side they lean away from is clearly deranged and dangerous for the country. Certainly at an elite policy level, the activists may have a point about unified extremism (see, e.g., Mann and Ornstein 2012), but there is a crucial difference between voting for a party and thinking like a party or at least always agreeing with one s chosen party. Political scientists need to be clear about this difference or we lead the media astray. This is not to say that all media accounts on this right now are wrong. For instance, Ron Brownstein suggests that America has reverted to being a nation (Brownstein 2012), but he goes on to argue that the party leaders should seek compromises that reflect the nation s diversity of opinion. Whatever one s normative opinion about compromise, Brownstein presents an accurate portrayal of the public. It is diverse and without really consistent policy views. It is this crucial difference between voting alike and thinking alike that matters in the way we portray the electorate. How can it be true that relatively few citizens are true independents and simultaneously be true that the electorate is not deeply polarized into two camps? The answer is actually pretty simple: the parties, especially among the electorate, are not nearly as unified many people seem to think. Descriptions of them as unified blocks are overstated. Alan Abramowitz obviously disagrees and writes that Americans who identify themselves as independents but who indicate that they lean toward one of the two major parties generally think and behave more like partisans than like true independents (Abramowitz 2012; see Abramowitz 2011 for a similar argument). A perusal of Pew s latest report on public opinion Partisan Polarization Surges in Bush, Obama Years documents a growing partisan divide (p. 2). The key piece of evidence is that on an average of selected values questions, the average

5 Pope: Voting vs. Thinking 3 difference between Democrats and Republicans has moved from 10 points in 1987 to 18 points in When one reads deeper, though, it becomes clear that there is a great deal of diversity within those two parties. To take just one example, it is true that more Republicans take a positive view of Wall Street than Democrats, but the difference is hardly a chasm: 69 percent to 53 percent (p. 64). Majorities of both camps view Wall Street as a positive force despite the meltdowns, bank bailouts, and other financial arguments. 2 A similar investigation of most of the questions reveals that deep chasms in partisan opinion tend to be more connected to candidates and elections; less often are they simply discrete policy questions. The claim that independents can be a new force in American political life is out there (Killian 2012) and needs to be corrected as well, but in doing so we should not be arguing that [independents] who lean toward the Republicans think and vote just like regular Republicans or that they think and vote just like regular Democrats (Teixeira 2012). This ignores too much diversity and difference in thinking within the parties. There is nothing wrong with the idea that leaners are closet partisans when they vote. Just as there is nothing wrong with the idea that leaners often look very much like the weak partisans in their attitudes. The problem is that the parties in the electorate do not have nearly the policy coherence necessary to think of them as unified camps. A return to The Myth of the Independent Voter helps tell the story. Data on Partisanship and Policy This is not the place for an exhaustive review of party positions on policy, but it is useful to return to The Myth of the Independent Voter to see how they tackled the question of the policy content of partisan views. In a chapter on issues and dealignment, the authors suggest that they have been unable to locate any particular issue that is motivating the rise of independents. In general, they looked for issues to explain the rise of independents in the public (Vietnam, women s roles, and busing, among others). With respect to these issues, they conclude that the range of opinion among Democrats, Independents, and Republicans is rather narrow (Keith et al. 1992, 157). When they turned to New Deal issues like government provided jobs and health insurance, they found somewhat wider 1 The scale here is theoretically 100 points, meaning that there has clearly been some movement but hardly into two unified camps. Another possible interpretation of the data is that the public is better sorted into parties now than twenty-five years ago, but hardly characterized by the unity necessary to think of the polity as being in two camps. 2 On a similar note, more than two-thirds of both camps believe that Wall Street only cares about making money for itself.

6 4 Submission to The Forum ranges of opinion. With respect to these scales, they always found a monotonic relationship between policy attitudes and partisanship. Careful not to make any strident claims about these differences, they merely noted that economic issues were different from other issues. These areas where they did find the greatest partisan differences thus seem like the best place to look for unified political party blocks. To that end, I take up three issues merely as an example and a brief replication of the work in The Myth of the Independent Voter. 3 They used the ANES seven-point scales where voters are asked: Some people feel that the government in Washington should see to it that every person has a job and a good standard of living. Suppose that these people are at one end of this scale [at which point a card illustrating the scale is shown to the respondent] at point number 1. Others think the government should just let each person get ahead on his own. Suppose that these people are at the other end at point number 7. And of course, some other people have opinions somewhere in between. Where would you place yourself on this scale, or haven t you thought much about this? Three issues that were relevant then, and clearly remain relevant today, are government health insurance, government services and spending, and the guaranteed jobs and income scales. Each of those scales has a similar wording to the scale described here (see the ANES for specific question wording). This list limits us to economics and government spending. The issues seem clearly salient in 2012 and have been salient for the past several years (and so have been asked often). And, just to emphasize the point, these are the issues where we see the largest differences between the parties. Tables 1 through 3 are similar to those presented in The Myth of the Independent Voter. They present the mean placement on that seven-point scale for each level of partisanship. As a check on the overall range of the distribution, the final column of each table includes the standard deviation for the distribution. 4 Scrutinizing Tables 1 through 3 reveals a few patterns. First of all, it is clearly still true that there is a relationship between partisanship and policy. Monotonically, Democrats place themselves on the left of each scale and Republicans place themselves on the right. It also appears still to be true that leaners look a great deal like weak partisans. Their policy views are usually closer to the party s weak partisans than they are to the pure independents. This confirms the basic findings argued in The Myth of the Independent Voter. 3 Though, for reasons laid out in the text, The Myth of the Independent Voter focused on white respondents, whereas here I will include all adult respondents. 4 Respondents who indicated that they did not know how to answer survey questions were placed at a 4 on the scale.

7 Pope: Voting vs. Thinking 5 Table 1. Mean Self-placement on the Government Health Insurance Scale Health Insurance Year SD WD LD I LR WR SR σ Note: the far right column indicates the standard deviation for the overall scale for each year. Table 2. Mean Self-placement on the Guaranteed Jobs and Income Scale Jobs Year SD WD LD I LR WR SR σ Note: the far right column indicates the standard deviation for the overall scale for each year.

8 6 Submission to The Forum Table 3. Mean Self-placement on the Government Services and Spending Scale Services / Spending Year SD WD LD I LR WR SR σ Note: the far right column indicates the standard deviation for the overall scale for each year. As to the question of the range of opinion across the groups, there is some evidence that it has grown larger. In Table 1, the difference between strong Democrats and strong Republicans is 1.3 units on the scale in By 2000, that difference had grown by a full point to 2.3 units. In Table 2, the growth is even larger, as the difference between the two camps grows from 1.1 points to 2.4 points (more than doubling). Finally, in Table 3 (which has a shorter time-period under consideration) growth still exists, though it is a much smaller amount. The difference there shifts from 1.4 points to 1.8 points. All of this is consistent with a partisan sorting thesis discussed elsewhere (Fiorina, Abrams, and Pope 2006 and Levendusky 2009). The phenomenon is also referred to as partisan polarization (Bishop 2009 and Abramowitz 2010). I prefer the term sorting on the grounds that describing it as polarization could lead one to think that the underlying distribution of opinion has spread out a great deal. As a check on this claim, each table includes a measure of the standard deviation of the overall distribution for each year. This measure, in the case of both health insurance and guaranteed jobs and income, is actually smaller than it was in the 1970s. On the question of government services and spending, the distribution has spread out since 1984: by one-tenth of a point on the scale. Whatever term one uses for sorting or partisan polarization, the overall distribution has not grown much larger. The remaining question is whether the groups represented by the categories have moved so far apart as to make it impossible for them to see eye to eye on policy. Do they usually think alike by consistently taking similar positions? This question taps deeper issues about the nature of deliberation and compromise (Gutman and Thompson 2012). However, one way of approaching it

9 Pope: Voting vs. Thinking 7 is just to ask exactly how consistent are these partisans? We know that only the small group of pure independents in the middle of the spectrum are switching their votes between parties. We further know that the leaners tend to vote heavily for their party, though that can vary by election. For instance in 2008 (relying on the ANES), of those who classified themselves as leaning Democrats, 93 percent voted for Barack Obama. The equivalent calculation for Republicans was smaller, but still quite high, at 75 percent. These numbers can be affected by election-specific conditions like those that favored Obama in So, as to voting, the consistency is unquestionably high, but what about the thinking? Do voters consistently think alike as well? This raises a thicket of definitional issues relating to standards of consistency. How much is necessary to define an individual or a group as consistent? That is beyond the scope of a single short comment on the issue. However, we can look at the scales to see how often respondents are consistent across the scales, and the best way to do this is not by looking at agreement but at disagreement. A sports analogy may be useful here. When one wants to learn how good a team is, the best way to do so is not to simply look at their overall record. A better way is to look at the limits of their record. If a team is quite good, we ask which teams beat them? This gives a sense of the limits of that team s strengths. (We would do the reverse for a really bad team: did they beat anybody?) This is similar to what exams measure for students: which questions did a student miss on the exam? That tells a teacher what he or she has taught poorly or where there is a gap in student knowledge. The corollary for voters is probing the limits of their agreement, or finding where they disagree with their party. To operationalize disagreement on these scales, I will define it as placing oneself on the opposite side of the consensus view of the party. (Merely placing oneself in the center of the scale or offering a don t know response will not count as disagreement). Overwhelmingly, respondents place the Democratic Party on one side of each of those scales and the Republican Party on the other. Voters seem to know where to place the parties: they know that Democrats are on the left and Republicans are on the right. Given that fact, how likely are voters in each category to express an opinion in disagreement with the parties on at least one of the scales? The percentage of voters, by partisan identification, who are in opposition to their party on at least one of those scales is calculated and displayed in Table 4.

10 8 Submission to The Forum Table 4. Democratic and Republican Party Disagreement Republican Disagreement Democratic Disagreement SD WD LD I LR WR SR SD WD LD I LR WR SR Cell entries indicate the percentage within each category (Strong Democrat to Strong Republican) who placed themselves on the opposite side from which most voters placed the political party, for at least one scale. Placement in the middle of the scale was not counted as disagreement. In each panel the columns for the party members are bolded for emphasis. The typical strong Republican placed himself or herself in opposition to his or her chosen party on at least one of the scales about a third of the time looking across all years. 5 The number goes up for the weak Republicans and the leaning Republicans, to just under half of the time. For Strong Democrats, about forty percent of the time we find them disagreeing with their party on one of those scales. Among the weak and leaning Democrats, the statistic is often above half. The pattern of sorting does reveal itself here as well. There appears to be something of a trend toward agreement, though it is strongest among the strong Republicans and Democrats. Among weak and leaning Republicans, the hint of a trend is actually in the opposite direction: slightly greater disagreement in recent years. 5 The percentage obviously does vary, for instance 1996 appears to be a rather consistent year for that group.

11 Pope: Voting vs. Thinking 9 Bear in mind that this table, displaying the percent who express a policy opinion at odds with their party, considers only a few measures. 6 By looking at the place where The Myth of the Independent Voter found the greatest disagreement across parties (economic issues), we have minimized the chances of finding disagreement. Adding social-issue measures to the scale created here (or even expanding the scope of economic measures) would not make the percentages go down they would only cause the percentages to rise. Clearly, citizens appear quite comfortable expressing policy opinions that do not fit their party norms. Furthermore, it should be noted that the real division (when we look at these measures) is not between independents and partisans. It is between Strong Partisans and everybody else. On both sides of the aisle, the leaners and the weak partisans are often closer to the pure independents (on average) than they are to the partisans. This is especially true among Democrats. The Myth of the Independent Voter taught us that we should not think of independents as a unified block of voters. They are too diverse for that. Similarly we should not think of the parties (including the leaners, as we probably should for many conceptualizations) as unified blocks. They may, on average, be different from the opposite party, but they are rife with internal policy disagreement as well. One objection to this entire line of reasoning might be that these questions are insufficiently partisan. Some Democrats even elite Democrats might, in good faith, prefer that government spending and services go down a bit, or they might be uncomfortable with the specific language about jobs and income. (A similar story could be told about Republicans). Certainly that is possible, but it begs the question of unity. If one really can be consistent with the Democratic or Republican Party and place oneself at any point on that scale, then there really is not any standard of policy coherence to discuss in the first place. Summary and Discussion The preceding is obviously far from exhaustive, but I hope it reminds readers of a few points. First, though there is some evidence that citizens have become better sorted into political parties, there is little evidence that partisanship is as good an indicator of policy opinion. Citizens often and easily take stands at odds with their political party. As late as 2008, about a third of Strong Democrats (Republicans) placed themselves in the conservative (liberal) portion on the government health insurance, government services and spending, and the guaranteed jobs and income scales. The pattern for weak partisans and leaners is even more striking. Certainly, 6 In an earlier version I looked at adding abortion. There was relatively less disagreement with the party among Democrats, although among Republicans disagreement with the party was much higher and would only make these percentages larger.

12 10 Submission to The Forum partisanship and ideology have become more correlated, but that is far from saying that a voter s policy views can be clearly predicted by partisanship. Second, citizens are clearly quite capable of being policy moderates at least in the sense of holding a diverse set of views about the world and simultaneously voting in a very partisan way. This makes interpreting the public as supportive of broad political agendas questionable. The key lesson from The Myth of the Independent Voter was not that the citizens are dividing into two ideological camps. It was that persons who labeled themselves as independents were very likely to be voting with a political party on a consistent basis. There is no strong reason to believe that most citizen attitudes are similarly unified like those of the party elites. There is just too much diversity of opinion out there in the public (see also Levendusky and Pope 2011). Strong partisans think the most like elites, but even there, better than a third of the category expresses incongruent views on one of the three items. Far from being an indictment of their irrationality or inattention to politics, this may well be due to the fact that the political system does not invite principled consistency. Byron Shafer s invitation to write something for this issue began with the prompt The Changing American Electorate Fact or Fiction? That prompt suggests that we might examine the polity by focusing on the views of the citizens and then measure the types and amount of change in that arena. Without any disagreement with that particular goal, let me suggest that we spend more time thinking about the way that partisan elites have changed. In a provocative blog post from February of 2012, Ezra Klein reminded readers that the descriptions left and right often are taken to mean descriptions of where the two parties stand at any given moment rather than descriptions of the philosophies, ideologies, or ideas that animate, or should animate, political debates (Klein, 2012). Because political leaders are hardly monuments to intellectual consistency, they will often, in the pursuit of victory, take the position(s) necessary to get elected. Klein s entire piece is worth rereading, but a sampling of his points is instructive. Between the late 1990s and about 2007, supporting an individual mandate in health insurance was much more likely to be a position of the right rather than the left. 7 By 2010, this was a position held vigorously by the left and opposed by the right. During the George W. Bush administration, the left often worried about the expansion and abuse of executive authority, but since the election of President Barack Obama, such worries have dimmed significantly. 8 Another example Klein 7 Remember that Candidate Barack Obama campaigned against the individual mandate during the 2012 primary fight against Hillary Clinton. 8 Republicans on the other hand now seem much more interested in the question of executive power. According to Pew s American Values Survey the percentage of Republicans that are concerned the government is collecting too much information about people rose from 39 percent

13 Pope: Voting vs. Thinking 11 gives is the filibuster. As it has grown as a problem in the Senate, both sides have routinely switched positions on the question. This is not to claim that the parties are truly centrist or even interchangeable. They are just inconsistent, always evolving given the times. As I was drafting this short comment, I was also watching the Republican, and then the Democratic, National Party Conventions. It was impossible to miss Paul Ryan waxing eloquently about saving Medicare by maintaining spending levels and avoiding cuts to the program proposed by President Obama. The Republican delegates cheered. They cheered saving Medicare by increasing government funding over what had been proposed by President Obama. (!) Though it was not as consequential on policy, I could not help but also notice Democrats cheering as former President Bill Clinton approvingly quoted another former president who liked to say There you go again. Apparently, Ronald Reagan has become much more popular among Democrats than we might have expected given his career. 9 Of course there are clear reasons why the parties have taken each of these stands. All is explainable. Democrats probably do have significant disagreements with Ronald Reagan that might, in a different context, be more obvious. Those disagreements just do not matter as much as they once did. However, if the spectacle of elite Republicans cheering the restoration of funding to Medicare does not convince us that it is a pipe dream to look for true policy consistency among partisans, then I do not know what facts could possibly persuade us. The key change in each of the positions cited by Klein (and noticed by many others) was not new information about the consequences of policy. It was the partisan context. Parties saw an advantage in taking a slightly altered position. And this list only considers recent policy changes, hardly touching other issues that have seen major party changes on matters like race, trade, and the military. Party elites, existing within the constitutional restrictions and living under the mandate to win, simply have to do what is necessary to win elections and put their views into power. Partisan inconsistency is, literally, as old as the Republic. Thomas Jefferson and his Republican Party came into power arguing that the government had only very limited powers and should stay out of the business of in 2007 to 72 percent in Among Democrats the number fell only slightly (Stable URL: Democrats have similar shifts, depending upon the question. Between January of 2007 and August of 2011 the percentage of Democrats who felt that the government was doing very or fairly well in reducing the threat of terrorism rose from 36 percent to 85 percent (Stable URL: 9 It is also worth noting that this was a politician speaking out against de-regulation, a former president who had supported the repeal of Glass-Steagall, a policy widely associated with the financial crisis (Kuttner 2007).

14 12 Submission to The Forum the people as much as possible. He proceeded to govern by doubling the size of the country (via the Louisiana Purchase) and then instituting an embargo on all trade with Great Britain and France. Limited government may have been a good principle, but when confronted with a particular reality and a particular partisan context, it seemed less valuable than other goals. 10 Public opinion data from the early 19th century is difficult to come by but would, no doubt, show a bit of inconsistency as well. Citizens may have never had the kind of clear ideological views that exist among the elites. Yet the truth is that elites rarely have quite the consistency that we imagine in the heat of the campaign or when we focus on the present. Government and party leaders should not be faulted for being pragmatic and willing to update their positions based on changing circumstance, even if they do it more for winning elections than for the public good. A willingness to rethink one s own positions in light of new evidence is a hallmark of intelligence. Clinging, bitterly or otherwise, to a position purely for consistency is no virtue. The fact that leaders update their positions and frame questions in novel and interesting ways means that we should be placing the blame for polarization squarely where it belongs: on the leaders of the parties. Tom Mann and Norm Ornstein note that things are even worse than they look (Mann and Ornstein 2012) in Washington. Leaders and, as they emphasize in their book, this is even more true among Republicans than it is among Democrats are choosing policies far outside of the mainstream (see also Bafumi and Herron 2010). Sometimes leaders are choosing policies solely to score political points with relatively little policy commitment. 11 The party leaders should take responsibility for that kind of unprincipled partisanship. Partisan sorting may have played some role in unifying the political parties or creating an environment where party leaders can construct extreme coalitions, but that hardly absolves the leaders of responsibility. Morris Fiorina and Samuel Abrams have described this pattern as a disconnect between voters and their leaders (Fiorina and Abrams 2009). In a system where leaders engage in such behavior, is it any wonder that citizen attitudes lack policy coherence? Citizens are asked by public opinion researchers to offer their opinions on many questions, and they are given multiple considerations and cues by the political system. In that environment, we should not expect that they will respond consistently or clearly. We should expect that signals will be muddy, complicated, and contingent. Citizens have many goals, 10 It should also be noted that the Federalists, while in power, had argued vigorously for a broad interpretation of government powers. When confronted with members of the opposition party using those powers they found new reasons to oppose such actions. 11 The opening anecdote in Mann and Ornstein about how Republicans simply cannot permit President Obama even a small budget victory is illustrative.

15 Pope: Voting vs. Thinking 13 and they probably share them across the aisle: stronger economic growth, a sufficient national defense, and economic opportunity, among others. As the parties jockey to deliver policies that both deliver those goals and help the party win, citizens can be excused for lacking the policy wisdom to answer with deeply principled consistency. So let us be cautious in interpreting even the limited changes in citizen opinion as very meaningful. In the American political system, there is enough policy change at the elite level to keep us from expecting too much clarity among citizens. Voters may well be partisan, even very partisan, but the policy content of that partisanship is decidedly limited: citizens may consistently vote alike, but they do not consistently think alike. References Abramowitz, Alan The Disappearing Center: Engaged Citizens, Polarization & American Democracy. New Haven, CT: Yale University Press. Abramowitz, Alan Setting the Record Straight: Correcting Myths About Independent Voters, Sabato s Crystal Ball. Stable URL: Abramowitz, Alan Are Independent Leaners Closet Partisans or True Independents? Sabato s Crystal Ball. Stable URL: Ansolabehere, Stephen, John M. de Figueiredo and James M. Snyder Why Is There So Little Money in Politics? NBER Working Paper. Bafumi, Joseph and Michael C. Herron Leapfrog Representation and Extremism: A Study of American Voters and Their Members in Congress, The American Political Science Review 104: Bishop, Bill The Big Sort: Why the Clustering of Like-Minded America is Tearing Us Apart. New York, NY: Houghton Mifflin. Brownstein, Ron and National Journal, April 5. Stable URL:

16 14 Submission to The Forum Fiorina, Morris P., Samuel J. Abrams and Jeremy C. Pope Culture War? The Myth of a Polarized America, 2d ed. New York, NY: Pearson. Fiorina, Morris P. and Samuel J. Abrams Disconnect: The Breakdown of Representation in American Politics. Norman, OK: The University of Oklahoma Press. Geer, John In Defense of Negativity: Attack Ads in Presidential Campaigns. Chicago, IL: University of Chicago Press. Gutman, Amy and Dennis Thompson The Spirit of Compromise: Why Governing Demands It and Campaigning Undermines It. Princeton, NJ: Princeton University Press. Keith Bruce E., David B. Magleby, Candice J. Nelson, Elizabeth Orr, Mark C. Westlye and Rayomd E. Wolfinger The Myth of the Independent Voter. Berkeley, CA: University of California Press. Killian, Linda The Swing Vote: The Untapped Power of Independents. New York, NY: Macmillan Press. Klein, Ezra What Left and Right Really Mean. The Washington Post, February 24. Stable URL: Kurtz, Stanley Two Tickets: Two Americas, National Review, August 11. Stable URL: Kuttner, Robert The Alarming Parallels Between 1929 and 2007, The American Prospect, October 2, Stable URL: Levendusky, Matthew S The Partisan Sort: How Liberals Became Democrats and Conservatives Became Republicans. Chicago, IL: University of Chicago Press. Levendusky, Matthew S. and Jeremy C. Pope Red States vs. Blue States: Going Beyond the Mean, Public Opinion Quarterly 75:

17 Pope: Voting vs. Thinking 15 Magleby, David B., Candice J. Nelson and Mark C. Westlye The Myth of the Independent Voter Revisited, in Paul M. Sniderman and Benjamin Highton, eds., Facing the Challenge of Democracy: Explorations in the Analysis of Public Opinion and Political Participation, Princeton, NJ: Princeton University Press. Mann, Thomas and Norman J. Ornstein It s Even Worse Than It Looks: How the American Constitutional System Collided With the New Politics of Extremism. New York, NY, Basic Books. Noel, Hans Ten Things Political Scientists Know that You Don t, The Forum. Stable URL: Nyhan, Brendan and John Sides How Political Science Can Help Journalism (and Still Let Journalists Be Journalists), The Forum. Stable URL: Moulitsas, Markos No middle ground found The Hill, May 29. Stable URL: Teixeira, Ruy The Great Illusion, The New Republic, March 7. Stable URL: