October 2008
Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31  


PPBW Sponsors


Recent Entries

Recent Comments


« November 2005 | Main | January 2006 »

22 December 2005

Happy holidays

With the arrival of the semester's end, the Political Behavior Blog will be much quieter over the holidays. Look for more regular posts after the new year. PPBW begins again on February 3.

Posted by Barry Burden at 8:16 PM

21 December 2005

Looking for variance

The old Michigan triad of partisanship, issues, and candidate evaluations as an explanation for vote choices is proving less useful in recent days. The main reason is that party identification and the vote are practically one and the same. In the 2000 and 2004 NES data, better than 90% of partisans voted for the presidential candidate of their party. In 2004 only 40 respondents (7% of partisans) voted against their stated party identification. The last two presidential contests look strikingly like what the Michigan model would call "normal" elections.

For reference I checked out the 1960 NES since it too was a close election with high turnout. Roughly 95% of Republicans voted for Nixon but just 84% of Democrats voted for Kennedy. A larger base of Democrats back then obviously helped put Kennedy over the top. Over the past 45 years Democratic loyalty has almost caught up to Republican loyalty as the Democratic camp has become smaller and presumably more homogeneous.

Extremely high correlations between party identification and vote choice leave little room for other variables to influence the vote. There are at least three consequences of this for researchers. First, we ought to back the research question up a step (or two) and ask what determines party identification. Second, we ought to consider -- and test -- whether issues and candidate evaluations are already in party identification. Third, attention should shift to studying independents rather than partisanship since they broke almost evenly in 2000 and 2004.

Posted by Barry Burden at 10:59 AM

18 December 2005

Krosnick book conference

Eric M. Mindich Encounters with Authors Symposium takes place at the Institute January 19-21. This year the program features Jon Krosnick's new book, The Handbook of Questionnaire Design: Insights from Social and Cognitive Psychology. The conference features three days of lectures and discussion and is sponsored by the Institute's Program on Survey Research. The discussants being brought in are first rate. Be sure to follow the RSVP instructions on the web site if you're interested in participating.

Posted by Barry Burden at 4:56 PM

16 December 2005

Hillygus on ideology

One of the recurring themes of this semester’s PPBW has been the ideological nature of both the public and elites in American politics. We’ve discussed whether red states are ideologically pitted against blue states and how much mass publics echo the ideological polarization of their elected officials. Today’s workshop finally put ideology center-stage with Sunshine Hillygus’ new paper on "The Structure and Meaning of Political Ideology".

The paper uses an item response model to estimate a latent measure of ideology from a long series of issue questions in the 2000 NES. One of the key findings of the paper, I think, is that this issue-based latent measure differs from respondent’s own ideological self-placement on the traditional liberal-conservative scale – and that the new measure has a significant impact on things like presidential vote choice.

The audience certainly made plenty of suggestions for where to take the project. I guess that's the danger of coming up with a new measure for an important concept - everyone wants to resolve all current debates using it. Suggestions included seeing whether the dimensions of ideology had remained the same over time, whether this measure of ideology had become more correlated over time with party ID or vote choice, and whether mass ideology had changed in response to elite ideology or was in fact distinct from it. People also wanted to know about those respondents who seem cross-pressured between ideological dimensions, or between new issues and theirprevious ideological affiliations.

One comment really struck me, though. Ideology seems to be of increasing interest to pundits and political scientists alike. Yet we seem to have little conceptual guidance on what the term means – to some extent we don’t quite know what we’re looking for even as we develop different measures of it. Respondents to surveys are clearly thinking of something other than just issue preferences when placing themselves on the liberal-conservative scale. But what is that "something else"? What should they be thinking of? And should political scientists be re-evaluating what we mean when we say “ideological?? I’ve already tried to persuade one political theorist to write a dissertation with the working title "The Concept of Ideology", but to no avail. Hopefully this kind of empirical work will persuade someone to fill these gaps in the way we think about what is a substantively important concept.

Posted by Phil Jones at 5:59 PM | Comments (2)

Not-So-Happy Holidays

Religion and politics became ever more entangled this month as the religious right attacked political leaders for anti-Christmas sentiments. President Bush felt the heat as well; he was widely criticized for sending "Holiday Cards" rather than "Christmas Cards" to thousands of supporters this season. The President and First Lady explained that the cards were an attempt to express good cheer to people of all faiths; the religious right saw the cards as further evidence of Bush's bowing to leftist secularism. As Bill O'Reilly, host of the O'Reilly Factor, says, "This Christmas madness will not stop until traditional Americans hold the anti-Christmas forces accountable. If you do that, Christmas will return to the marketplace and to the public square. If you do nothing, the Christmas tradition will diminish to be replaced by the winter holiday tradition."

This story is interesting not for its ridiculousness but for the fact that politicians and businesses are responding to the pressure. Many retailers are putting "Christmas" back into their advertisements and some communities are reaffirming that their tree really is a "Christmas" tree. Has the Christian right gone too far? Have we crossed the line set out by the First Amendment more than 200 years ago? According to Justice Sandra Day O'Connor in Lynch v. Donnely (1984), "The Establishment Clause prohibits government from making adherence to a religion relevant in any way to a person's standing in the political community." Are Christians, by demanding that governments and government officials celebrate their holiday, also demanding special standing in the political community?

The quotation from Bill O'Reilly can be found at http://www.foxnews.com/story/0,2933,177145,00.html.

Posted by Traci Burch at 4:43 PM

14 December 2005

Attitudes toward Flag Burning

Hillary Clinton recently came out in support of legislation that would outlaw flag burning in some circumstances, but says she still opposes a constitutional amendment. The most recent poll I could find in a quick search was a ABC News/Washington Post poll from 1998, in which 79% of respondents thought that flag burning should be illegal, but only 51% supported a constitutional amendment. Needless to say, how the issue is framed dramatically alters the level of support. I would be interested to see how constant support has been over time. I expect that there would have been even greater support for outlawing flag burning immediately following 9/11, but I wonder if attitudes have changed at all as a result of the situation in Iraq. Alternatively, this may be one of the issues where public opinion is fairly stable, as not a whole lot of information is needed to answer the question.

Posted by Ian Yohai at 11:49 AM | Comments (1)

12 December 2005

PPBW: Hillygus on Ideology

Our final PPBW session of this semester is this Friday, December 16. Sunshine Hillygus will present "The Structure and Meaning of Political Ideology." The paper introduces a new means for operationalizing ideology using item response models on issue questions from the NES. Just what is ideology? Here's the abstract:

Given the focus of media and scholarly attention on the increasingly ideological nature of American politics, it is somewhat surprising that so little is known about the structure and meaning of political ideology in the electorate. In this paper, we estimate a Bayesian issue-based latent measure of ideology estimated using an item response model (IRT). We compare our latent measure of ideology with ideological self-placement, party identification, and vote choice in the 2000 presidential election. The latent measure offers a number of methodological and conceptual advantages, and offers more more complete picture of the belief systems of the American public.

Posted by Barry Burden at 2:00 PM

Shapiro on foreign policy polarization today

Not to be last minute or anything, but there's a talk today at the Kennedy School by Robert Shapiro of Columbia. The paper is titled "Partisan Conflict, Public Opinion, and U.S. Foreign Policy." Here's the abstract:

Since the 1970s American politics has become increasing polarized along partisan and ideological lines. This polarization has been widely observed and debated in the area of domestic economic and social policymaking, and since September 11th and since the war and U.S. occupation of Iraq, there are signs of this in the conflict among political leaders concerning American foreign policy. If partisan and, especially, ideological conflict were to be a persistent characteristic of foreign policy debates, this would be major change in the nature of American politics, in which such conflict thus far has not extended beyond domestic politics. This paper examines the surveys conducted from 1998 to 2004 by the Chicago Council on Foreign Relations to see the extent to which, if any, the American public’s as well as leaders’ opinions toward foreign policy issues have become more polarized then in the past along Democratic-Republican partisan and ideological lines. It finds that elite opinions have indeed become more polarized, and there are indications that this may be occurring for the mass public as well, especially in ideological terms. Such increasing divisions in leaders’ opinions on foreign policy issues, if they persist, and the extent to which similar divisions have become more pervasive among the public, would indicate that American partisan politics and the nature of public opinion has changed in profound ways over the last fifty years.

Posted by Barry Burden at 10:27 AM

6 December 2005

Bounds on Presidential Approval

One of the things that polarization does is limit the size of the political playing field. Having hard cores of Democrats and Republicans at opposite ends of the ideological continuum effectively limits the action to the center. As long as the bases remain loyal, swings in all kinds of political indicators will be dampened. I suspect that this is one overlooked factor limiting congressional turnover. The last four cycles have seen remarkably little turnover (the lowest of the postwar era aside from the mid-1980s). More immediately, polarization also keeps presidential approval from dipping below about 35% as of late. As Charles Franklin pointed out in a response to my question, the strong support of the Republican base has kept Bush affloat. As as long as the conservative base remains in tact -- still an open question -- we shouldn't expect the president's ratings to drop any further.

Posted by Barry Burden at 10:00 AM | Comments (6)

5 December 2005

Lawless talk: cancelled due to snow

The final CAPS talk of the semester is by Jennifer Lawless of Brown University. The title is "It Takes a Candidate: Why Women Don't Run for Office." (Jen might know since she happens to be running for Congress at the moment.)

Posted by Barry Burden at 5:35 PM

3 December 2005

The Mixed Electoral System: Are rumors of its death greatly exaggerated?

In the last decade, the electoral system of choice for dictatorships turned democratic and for established democracies undergoing reform has been one that combines majoritarian and proportional institutions in a single election. Since 1990, "mixed electoral systems" have spread to dozens of countries and sub-national units on every populated continent. While the dynamics that led to their adoption differed considerably, mixed systems were invariably heralded by their proponents as election rules that would combine the most beneficial consequences of both majoritarian and proportional institutions while tempering their well-known shortcomings. A similar optimism also characterizes prominent scholarly work on the subject. Shugart and Wattenberg (2001) unambiguously state that the mixture of majoritarian and proportional institutions offers "the best of both worlds," in part because of its anticipated moderating effects on national party systems.

Whether or not mixed systems have fulfilled their promise - to combine the best of the majoritarian and proportional worlds - has been the subject of a lively debate in recent years (to engage in shameless self-promotion, the book I co-wrote on the subject - "Mixed Electoral Systems: Contamination and Its Consequences" - will be published by Palgrave next week). Nonetheless, while mixed systems may have been considered an optimal solution to the challenges faced by institutional designers throughout the 1990s, political elites have frequently had second thoughts about their introduction. Mixed systems have been abandoned in Azerbaijan, Bulgaria, Croatia, Kyrgyzstan, Macedonia and Ukraine. Russia may soon follow. In Italy, Silvio Berlusconi's Forza Italia is a Senate vote away from replacing the mixed system with a particularly disproportional kind of proportional representation. It may well be that these changes are purely “outcome-contingent? - or exclusively based on the intent of manipulating the rules to ensure more favorable outcomes. However, if mixed systems are really the best of both worlds, it is curious that their reforms have generally been so smooth and their defenders so few. Everything is not for the best – so it appears – in this best of possible worlds.

Posted by Federico Ferrara at 3:45 PM | Comments (1)

2 December 2005

Polls on polls

I came across an interesting article the other day which shows the results of a 1985 Roper Organization opinion poll which asked the public for their views on…opinion polls (full cite: Roper, Burns W. 1986. “Evaluating Polls With Poll Data?. Public Opinion Quarterly 50: 10-16). What do the people who take part in surveys think of them? A couple of the results stand out for their sheer humor value:

When asked about the type of person who takes part in a survey, a full 7% thought that pollsters interview “mostly unusual, nontypical, or even oddball types of people?. Unfortunately, the obvious follow-up question wasn’t asked, and we’re left to guess whether this 7% were projecting inside information about their own abnormality on to the rest of the sample.

When asked about the honesty of survey respondents, 16% believed that “only some? or “very few? people tell the truth to interviewers. Again, are these people projecting from their own dishonesty? I don’t know how we can begin to evaluate – let alone trust – the answers to this question if they were!

Beyond these questions, respondents were also asked how many times they had been interviewed in a poll prior to this one. The breakdown of the results was:

Never before - 41%
Once - 17%
Twice - 16%
3 to 5 times - 16%
6 or more times - 9%
Don’t Know - 1%

These numbers were far higher than I had expected, and raised a question in my mind.We know that taking part in a survey has an effect on respondents' attitudes: they become more interested in politics, and seem to learn more about current affairs after talking to an NES interviewer for an hour, for example. The worry here is that by virtue of taking part in a randomly sampled poll, a person is no longer representative of the public at large. We also know that there has been an explosion in the sheer number of opinion polls conducted since this Roper survey in 1985. The worry here is that many more people have previously taken polls than was the case in previous decades. At what point do we need to start controlling for a respondent’s prior survey participation when using polls to understand public attitudes?

Posted by Phil Jones at 5:46 PM

PPBW - Hopkins Paper

In the December 2 Political Psychology and Behavior Workshop, Daniel Hopkins of Harvard University presented his paper, "The Diversity Discount: How Increasing Ethnic and Racial Diversity Dampens Support for Tax Increases." This paper merges the literatures from political economy, sociology, and comparative politics to theorize about the effects of racial and ethnic heterogeneity on expenditures on public goods. This paper makes two contributions. First, Hopkins explores four mechanisms through which diversity may suppress the provision of public goods, one of which thinks seriously about how elites set the public agenda. Second, Hopkins argues that changes, rather than the level, of diversity affect the provision of public goods.

To examine these claims, Hopkins analyzes instances in which towns hold and approve debt override and exclusion votes to fund public goods. The findings indicate that in Massachusetts, all other factors being equal, a locality with increases in diversity over the period of the study was more likely to hold votes on debt exclusions. If these findings hold, Hopkins's research implies that if communities are to maintain adequate levels of provision in the face of demographic change, then policymakers should think of ways to help communities manage public expenditures in periods of increasing diversity. These implications are more optimistic than those provided to us by the existing literature, from which one can conclude that heterogeneous localities will not be able to provide the public goods its citizens need.

The audience responded positively to the paper. However, the audience urged greater complexity in the theoretical understanding of the ways in which diversity matters. First, the analysis might benefit from disaggregating different types of public goods--for instance, diversity may have opposite effects on expenditures for corrections than one would expect from this model. Second, the relationship between elites and voters remains a bit of a "black box" in this paper and should be explored further. One audience member also asked Hopkins to include the type of town or city government in the model.

Posted by Traci Burch at 4:10 PM