The Monkey Cage

See here.

Posted by Andrew Gelman at 09:51 PM | Permalink | Comments (0)

Time to toss every physics nerd stereotype you ever had:

Brian May, the guitarist and founding member of the legendary rock band Queen, earned his PhD in astronomy last year from Imperial College London. His PhD thesis A Survey of Radial Velocities in the Zodiacal Dust Cloud has just been co-published by Springer and Canopus Publishing Ltd.

Since this is a political science blog, this opens the question of who is the biggest celebrity with a Ph.D. in political science. My vote is for Robert Vaughn. He was one of the Magnificent Seven, the Man from U.N.C.L.E., and he has a Ph.D. in Communications and Political Science.

Posted by Phil Klinkner at 09:19 PM | Permalink | Comments (13)

I see that Political Theory has just published Alexander Wendt and Raymond Duvall’s piece, Sovereignty and the UFO (link to abstract; the article itself is behind a paywall) an article which, it is fair to say, has acquired a certain degree of notoriety. Wendt and Duvall make a complex argument, drawing on Derrida, Agamben etc, but their basic claim is pretty straightforward as I read it. First - there is some material basis to suggest that there is some objective phenomenon, perhaps involving extraterrestrials, perhaps not, behind UFOs. Second, the failure of states to further investigate UFOs, despite this evidence, suggests that there is some structural reason why UFOs simply do not compute for states. Third, the most important structural cause for this blindness likely lies in the nature of sovereignty - because sovereignty presupposes a kind of anthropocentrism, sovereign authorities can’t deal with the possibility of alien intelligences, and hence construct a ‘regime of truth’ in which the notion that UFOs might exist is ipso facto ridiculous. Not only that; a proper understanding of UFOs might lead to the collapse of our current notions of governmentality. In Wendt and Duvall’s words:

For a critical theory of anthropocentric rule, therefore, a science of UFOs ironically is required, and not just a science of individual cases after the fact, which can tell us only that some UFOs lack apparent conventional explanations. Rather, in this domain what is needed is paradoxically a systematic science, in which observations are actively sought in order to analyze patterns from which an intelligent presence might be inferred. That would require money, infrastructure, and a long-term commitment of the kind that to date has been possible only for epistemic authorities, or precisely those actors most resistant to taking UFOs seriously. Still, given the potential disjunction of interest between science and the state, it is possible here for science to play a key role for critical theory. Whether such a science would actually overcome UFO ignorance is unknowable today, but it is only through it that We might move beyond the essentially theological discourse of belief and denial to a truly critical posture.

Modern rule and its metaphysics are extraordinarily resilient, so the difficulties of such resistance cannot be overstated. Those who attempt it will have difficulty funding and publishing their work, and their reputations will suffer. UFO resistance might not be futile but it is certainly dangerous, because it is resistance to modern sovereignty itself. In this respect militant UFO agnosticism is akin to other forms of resistance to governmentality; however, whereas sovereignty has found ways of dealing with them, the UFO may reveal an Achilles heel. Like Achilles, the modern sovereign is a warrior whose function is to protect—in this case, from threats to the norm. Unlike conventional threats, however, the UFO threatens humans’ capacity to decide those threats, and so cannot be acknowledged without calling modern sovereignty itself into question.

… taking UFOs seriously would certainly embody the spirit of self-criticism that infuses liberal governmentality and academia in particular, and it would, thereby, foster critical theory. And indeed, if academics’ first responsibility is to tell the truth, then the truth is that after sixty years of modern UFOs, human beings still have no idea what they are, and are not even trying to find out. That should surprise and disturb us all, and cast doubt on the structure of rule that requires and sustains it.

I’m highly skeptical of claims that UFOs are interesting in any sense but the sociological, for reasons having to do with the Fermi paradox, and the weaknesses of evidence laid out in John Sladek’s excellent and entertaining The New Apocrypha (the relevant chapter is entitled ‘Will U Kindly F O”). But even apart from the question of whether or not there is something to UFO claims that is worthy of sustained scientific investigation, Wendt and Duvall’s argument is highly unconvincing. They claim that there is something about the possibility of alien subjectivities that fundamentally challenges the principles of sovereign rule and makes sovereign entities go into a kind of halting state (pun borrowed from Charlie Stross) when they try to think about them. Hence, the failure of states to investigate UFOs. But this doesn’t seem to me to hold up as a convincing explanation

First - the evidence is inadequate to the claims made. Even if we accept, for the purposes of argument, Wendt and Duvall’s contention that government sponsored reports such as the Condon report (which was prepared by thirty scientists and engineers for the government over a period of two years, based on twenty years of data gathered by the Airforce)¹ doesn’t serve as countervailing evidence because it was ‘politicized’ and ‘flawed,’ I would expect them to provide some serious evidence to support the claim that the purported outcome (failure of government to seriously investigate UFOs) is an outcome of the suggested causal mechanism (the inability of anthropocentric sovereign states to process evidence. We don’t get this; not only do we not get any sort of process tracing, we don’t even get a Foucauldian genealogy. Instead, we have an article which consists of (a) a claim about the nature of sovereignty, (b) a (contestable) claim about states’ failure to research UFOs, (c ) a series of mostly straw man counter-arguments, of mixed accuracy², that are duly knocked down, and (d) a re-iteration of the claim that in the absence of ‘reasonable’ counter explanations, the failure to investigate whether UFOs are indeed the product of alien intelligence must indeed be a result of anthropocentric sovereignty.

Not only is there no direct evidence adduced, but there’s no reason to think that governments’ failure to pay attention to UFOs is unique. There are many, many out-of-left-field claims out there that governments, mystifyingly, have failed to investigate, despite evidence that is at least as good as the evidence that Wendt and Duvall claim justifies further investigation of the UFOs-are-evidence-of-aliens-among-us hypothesis.³ For example astrology, which some have claimed has a statistical basis, and which would surely revolutionize predictive intelligence if properly developed. Transcendental Meditation - we have research published in The Journal of Conflict Resolution, no less, which points to the statistically observable effects of application of the Maharishi Technology of the Unified Field to the Israel-Lebanon conflict. Curiously, states have failed (to date) to act on this finding in order to solve the problem of war. Prayer, for healing and otherwise; now there’s one that the 2000-2008 US administration should have been investigating, one would have thought, given its ideological predilections and the technology’s obvious battlefield implications. And going a bit further into the wildlands, what about Scientology? If the US military had had a few Operating Thetans on tap to send out against the enemy, the Iraq war would have gone rather better than it did.

And so on. On Wendt and Duvall’s argument as I understand it, none of these potentially exciting fields of discovery challenge the anthropocentric notion of sovereignty. Yet none to date have been investigated - something that seems difficult to explain using their argument. If only one could come up with a single mechanism to explain this seeming regularity …

Even more troubling, there is evidence (not actually discussed in the piece, although the program is mentioned in passing) that seems to directly undermine Wendt and Duvall’s basic claim. In their argument, “one of the possibilities that we must countenance if we accept that the UFO is truly unidentified is that its occupants are ETs—and that threatens both the physical and ontological security of modern rule.” But if this is the underlying problem, we can point to a research area which is equally problematic for the notion of anthropocentric sovereignty, but that has been sponsored by the state. I’m referring, of course, to the ongoing SETI (Search for Extraterrestrial Intelligence) program, which uses radio telescopes and similar to search (so far unsuccessfully) for evidence of intelligent life elsewhere in the universe. This ought to be just as unthinkable for the modern state as UFO research, if an anthropocentric universal order is a necessary precondition for modern sovereignty. Yet it has been funded by the US over a period of many years. I think that this is substantial disconfirming evidence for their argument.

I’m much more familiar with Wendt’s work than Duvall’s; I do know both are very smart guys. Nor do I think that articles which investigate ‘weird’ topics should be dismissed out of hand - scientific progress often requires open filters. But I do think that there are real problems with the set up and evidence of this article, regardless of the topic - if you want to make a case that states simply cannot deal with certain problems, you need good evidence to support your claim, and I’m not seeing it. I am much more impressed with a superficially similar piece, Patrick Jackson and James Heilman’s chapter on “Outside Context Problems,” published in Donald Hassler and Clyde Wilcox’s New Boundaries in Political Science Fiction volume⁴ which convincingly argues (a) that Iain Bank’s spiffy series of ‘Culture’ novels is all about the difficulties that liberal civilizations have in encountering alien Others, and (b) that political scientists and theorists ought to pay attention to these books. Something along these lines, making less sweeping claims and using the evidence as a kind of thought experiment, would, I think, have made for a much better piece.

¹ summary of Sladek, op cit. As Sladek observes (p.39): “If nothing else, the Condon report should have convinced ufologists that saucers were being taken seriously. It didn’t. Scientists who worked on it, Dr. Edward U. Condon in particular, were accused of being ‘anti-UFO’ and their ‘professional bias’ was often noted by amateurs who themselves showed little inclination to accept anything less than visitors from space. … Donald Keyhoe sent an angry telegram to President Johnson urging him to terminate the project. Richard Shaver found Condon a ‘pedant’ for not getting on with the real business of science, i.e. fighting the evil, telepathic dero. (Shaver referred to the dero by the Velikovskyan phrase, ‘vermin from space’).”

² For example, Wendt and Duvall suggest that NASA’s defunct ‘Breakthrough Propulsion Physics’ program provides evidence that ideas about FTL travel are ‘speculative’ but ‘scientifically sound.’ I’ve been an enthusiastic amateur consumer of ideas such as the Alcubierre drive etc for years, but my understanding is that none of these ideas seem at all practically feasible for reasons to do with fundamental physical constraints, massive energy requirements etc - they serve less as blueprints for feasible stardrives than a form of entertainment for bored physicists. More to the point, while most of NASA’s BPL web pages seem to have disappeared with the passing of the ages, my very strong recollection is that the money available through this project was peanuts.

³ Wendt and Duvall do suggest that the willingness of states to investigate ESP, together with their unwillingness to investigate UFOs suggests that it really is about notions of sovereignty. But there are many other hypotheses which could explain this more simply. Most obviously, it is rather easier to study ESP in an experimental setting with controls and the like than it is to study UFOs, which have a disconcerting habit of disappearing will-o-the-wisp-like when you might like to get your hands on them.

⁴ Which, I should say, also has a piece by me.

Posted by Henry at 11:20 AM | Permalink | Comments (4)

The confluence of recent polls and yesterday’s report (below) on where the campaigns are spending their media money makes for some interesting insights into the presidential race.

To me, the most interesting state is Pennsylvania. McCain and the RNC have spent the most money there ($6 million dollars since June 3), but the state just hasn’t become competitive. The new Quinnipiac poll shows Obama leading by 7 among likely voters. Unless things begin to move quickly in his direction, I would guess that McCain needs to think about writing off the Keystone state. Of course, doing so will only make his route to 270 EVs all the more difficult since NH and MI seem to be the only Kerry states he has a shot at and he’s running even with or behind Obama in several states that Bush won in 2004 (CO, VA, IN, and FL).

Update: More bad news for McCain in PA. The latest Stategic Vision poll has him down by 9 among LVs.

Posted by Phil Klinkner at 08:59 AM | Permalink | Comments (0)

Since the Doha round of trade talks has been on life support for some time, it’s no huge surprise that the talks have collapsed with plenty of recriminations to go around. Much fuss is being made over one more sign that the United States has less influence than it did ten years ago, and that China has chosen to flex its muscles internationally in yet another venue.

What hasn’t received attention is a broader phenomenon of the Bush years: the rollback of Bill Clinton’s efforts to make economics co-equal with traditional national security in the conduct of American foreign policy. As Charlene Barshefsky, Clinton’s second term US Trade Representative argued to me and Derek Chollet for our book, America Between the Wars: From 11/9 to 9/11, “Clinton’s contribution was to elevate trade as a strategic component, not a strategic stepchild of a vibrant international policy.” And it was more than just trade – Clinton’s creation of the National Economic Council and his elevation of the Treasury Department in general foreign policy decisionmaking stemmed from the belief that the State Department, NSC and Department of Defense could not on their own ensure the realization of American interests in a globalizing world. Bush’s first two Treasury Secretaries were almost non-existent in foreign policymaking, and the episodic attention to Doha over the years was yet one more price paid for the dominance of Iraq in this administration.

Since it’s a process issue, the role of the US Treasury Secretary and USTR in the conduct of foreign policy decisionmaking won’t get much – if any – attention during the campaign. But the next president should, as Barshefsky puts it, view trade (and economics more generally) as a strategic component of American foreign policy. That, of course, will also bring politics front and center, but that will be the subject of another post.

Posted by James Goldgeier at 02:35 PM | Permalink | Comments (1)

We’re pleased to announce that James Goldgeier, Professor of Political Science and International Affairs at GW, has been captured and will be held in captivity here in the cage until he escapes in a few weeks.

After stints on the faculty at Cornell and as a visiting research fellow at Stanford, Jim joined the GW faculty in 1994. Since then, we’ve seen too little of him, but in a good way, as he has served as a Council on Foreign Relations International Affairs Fellow at the State Department and on the National Security Council staff, as a Fellow at the Council on Foreign Relations, as the Henry Kissinger scholar in foreign policy and international relations at the Library of Congress, and as a Public Policy Scholar at the Wilson Center; he is now a Senior Fellow at the Council on Foreign Relations.

Jim is the author or co-author of several books — most notably, perhaps, Not Whether But When: The U.S. Decision to Enlarge NATO, and the hot-off-the-presses America Between the Wars: From 11/9 to 9/11 (with Derek Chollet).

During his period of captivity, Jim will be weighing on on topics related to American foreign policy, international security, transatlantic relations, or whatever else captures his fancy. (We’re hoping that he’ll ignore his life-long fixation on the Baltimore Orioles baseball, a topic far too dismal to be well-suited for this cheery blog.)

Posted by Lee Sigelman at 10:20 AM | Permalink | Comments (0)

In 1991-92, political science theory, campaign politics and global trends converged, and Bill Clinton’s foreign policy team stressed the importance of democracy when discussing America’s attitudes toward other countries. It was a way to differentiate the Arkansas governor from incumbent George H.W. Bush, who was charged with being too soft on China and too cozy with Gorbachev. It’s rare that theory gets so ingrained in policy conversations. (Other examples include balance of power, which had more play in the 1970s than the 1990s, and bureaucratic politics, which, although dismissed by many academics, is taken as a given by most of official Washington.) But Democratic Peace theory was all the rage in the Clinton years.

Since a signature foreign policy theme of Senator John McCain’s is the creation of a League of Democracies, and since that idea is one that has been supported by top Obama advisers such as Anthony Lake, it’s worth thinking back on the theory/politics/global trends convergence in 1992 regarding democracy promotion compared with the theory, politics and global trends behind creating a league of democracies.

In 1991-92, democracy was flourishing – communism had collapsed, and states throughout Central and Eastern Europe were rushing to join the West. It was a no-brainer to accept that regime change mattered – former enemies now wanted to be allies. Russia looked like it was on a democratic track. Citizens in places like Mongolia and Namibia were braving long distances to polling places (and extreme weather when they arrived and stood in line for hours), showing that an enthusiasm for democracy wasn’t limited to Western culture (where people likely wouldn’t brave either the long distances or the extreme weather).

Not only was democracy flourishing, so was Democratic Peace theory. So it made sense that promoting democracy would become part of the policy mainstream. But there’s another reason democracy promotion became central to the policy ideas articulated by the Clinton campaign and then the administration that followed. As Derek Chollet and I argue in our book, America Between the Wars: From 11/9 to 9/11, Clinton’s advisers believed an emphasis on democracy promotion could be a political winner. The goal was to bring back to the Democratic fold the neo-conservatives (yes, those neo-conservatives) who had left the party in the 1970s and 1980s and had supported Republicans for president in 1980, ’84 and ’88. The neo-conservatives were unhappy that George H.W. Bush wasn’t Ronald Reagan, and a number of prominent members of that community ended up endorsing Clinton. So, with theory to buttress the idea, a political rationale for pushing it, and a seeming rush by former authoritarian regimes to join the democratic fold, Clinton in his foreign policy speeches during the campaign – a campaign that admittedly wasn’t about foreign policy – made sure to highlight the need to do more than George H.W. Bush to help fledgling democracies. Clinton’s 1994 National Security Strategy (much less noticed than George W. Bush’s 2002 call for a “balance of power in favor of freedom) emphasized a strategy of democratic engagement and enlargement.

Well, here we are in 2008. McCain says we should have a league of democracies. So do leading Democratic foreign policy experts such as Ivo Daalder and James Lindsay. “Democracies,” they wrote in the American Interest awhile back, “share the most important value of all—a common dedication to ensuring the life, liberty and happiness of free peoples. And democracies constitute the world’s most capable states in terms of military potential, economic capability and political weight. A Concert that brings the established democracies together into a single institution will be best able to meet the many challenges that beset the new age of global politics.”

Daalder and Lindsay make a powerful case. And thanks to McCain’s support for the idea, the most recent issues of key policy journals (e.g., Foreign Affairs, Foreign Policy and World Affairs) all have at least one article asking whether or not the idea makes sense. Having argued in Foreign Affairs two years ago with Daalder that we need to include non-European democracies in NATO, I’m sympathetic to the notion of a more general effort to organize the democracies.

Those who supported the war in Kosovo know that Clinton could never have gotten U.N. Security Council authorization for that effort and thus went through NATO, which at the time the war started had just grown to an organization of nineteen democratic countries. More recently, seeing Russia and China veto the U.N. effort to sanction Mugabe reminds us once again of the problem that the major autocracies can pose.

But South Africa also voted against sanctions against the rulers of Zimbabwe. And that raises a theoretical question: would “a common dedication to ensuring the life, liberty and happiness of free peoples” lead to a convergence of foreign policy preferences among states in league? Much academic literature has been devoted to studying whether established democracies don’t go to war with one another because of their political systems. We don’t have the same body of literature that can speak to the notion of the foreign policy actions that a league of democracies might take.

Nor do we have either the global trends or the political imperatives. The United States has lost both legitimacy and relative power since the early 1990s; Russia isn’t democratic, and it doesn’t look like democracy is sweeping the Middle East. It’s not even clear that any major democracies would answer McCain’s call to form a league – and there’s not going to be a league if no one wants to join. But there is also a lack of political payoff. McCain has had the support of the neo-conservatives ever since he backed the Kosovo war in 1999, and they certainly like hearing his charge for the league. But how much does that really get him? Clinton needed to find a few issues that he could use to attack an experienced foreign policy president. In the early 1990s, IR theory, global trends and political calculations by the New Democrats all pointed in one direction when it came to discussions about democracy. That doesn’t seem to be the case in 2008.

Posted by James Goldgeier at 09:37 AM | Permalink | Comments (2)

Why don’t governments redistribute more wealth to the less fortunate who, by most definitions of “fortunate,” out-number the wealthy? A standard explanation is Marx’s: the poor are distracted from their economic self-interest — e.g., a vote for left-wing parties who tend to favor redistribution — by things such as religion (the “opium of the people,” etc.). Is this true?

According to a recent article by Anna de la O and Jonathan Rodden, the short answer is a qualified “yes”:

…the impressive relationship between church attendance and voting against the parties of the left is driven disproportionately by the poor.

That’s the “yes.” The qualification is this:

However, we also discover that these relationships are primarily driven by the large presence in our sample of countries in continental Europe that use proportional representation.

In other words, the “opium effect” (my words, not theirs) emerges most strongly in countries with proportional representation. Why? De la O and Rodden write:

Conflicted voters in majoritarian countries with two-party systems must often choose between their moral and economic preferences when voting, whereas proportional representation reduces the barriers to entry for hybrid political parties that take leftist positions on one issue dimension and rightist positions on the other.

Another interesting tidbit:

Although preferences on the moral values dimension do push the religious—especially the religious poor—strongly to the right, they also create a strong push to the left among the secular rich—in fact, a stronger push than that of the secular poor.

And another one:

In spite of all the talk about a culture war, this study shows that economic preferences are far better predictors of vote choice in the United States than moral values preferences. Yet the opposite seems to be true in several relatively religious European countries with multiparty systems.

The article is here. Recommended.

[For an analysis of the American case, see Larry Bartels’ critique of Thomas Frank’s What’s the Matter with Kansas? and Andy et al’s new book. See also Andy’s post on a related project by John Huber and Piero Stanig.]

Posted by John Sides at 08:56 AM | Permalink | Comments (6)

Seth’s statement here and his follow-up comment reminded me of a big difference between psychologists, who like to do experiments, and political scientists, who like to analyze existing data. I’m always impressed that when a psychologist wants to learn something, he or she will typically run an experiment. Whenever I’ve tried to run an experiment, it’s been a mess. Which makes sense: I’ve been analyzing data continuously for the past 25 years but I have very little experience with data collection (and that is mostly helping others with their data collection projects).

Yes, I realize that some political scientists do run experiments, but I’m not usually so impressed by them, certainly not in comparison to psychological experiments. Political science is a different sort of field, it doesn’t lend itself to experimentation so much. Small n and all that. (But Seth might disagree, since he’s a psychologist who does small-n nonrandomized experiments.) And, yes, I realize that political science data gets collected: certainly the people who run the National Elections Study, the Pew survey, Polimetrix, etc., do data collection. But I think I’m typical of most empirical political scientists in that I spend much much more time analyzing than collecting data.

Posted by Andrew Gelman at 10:45 PM | Permalink | Comments (4)

I’ve always wanted to be president but I’ve been so busy that I’ve just never gotten around to it. Now I learn that millions of people are planning to vote for me anyway. Please join them. Click here for the full story.

And now I’ve got to begin practicing saying “When I’m president”:

[Tip of the chapeau to Zsuzsa Csergo]

Posted by Lee Sigelman at 07:01 AM | Permalink | Comments (1)

Tom Edsall has a good overview of the election predictions offered by various political scientists. The consensus? A big win for Obama, unless he loses.

Posted by Phil Klinkner at 12:10 PM | Permalink | Comments (0)

Posted by John Sides at 10:07 AM | Permalink | Comments (1)

Time magazine’s The Page has this description of Cindy McCain’s op-ed on her visit to Rwanda:

The potential first lady pens a WSJ column on her recent trip to the small west African nation, which is recovering from genocide.

But Rwanda isn’t in west Africa. It’s in east Africa.

Posted by Phil Klinkner at 09:07 AM | Permalink | Comments (0)

Here’s some wacky data on social class and views of corporations. In addition to the mysterious popularity of Citibank and Microsoft, here are some other interesting findings:

80% of Democrats and Independents were concerned about business collecting personal information and 65% were concerned about government. Among Republicans, 60% were concerned about business collecting the information and only 40% concerned about government. Also, people with higher income and higher education have “less concern about government data collection, while lower income is associated with higher concern. Income and education did not affect opinions about businesses collecting data.” The bit about higher status people trusting the government more makes sense and is consistent with other survey results I’ve seen, but I’m surprised that there isn’t a similar pattern regarding concern about businesses.

P.S. I’m not saying that people should hate Citibank, Microsoft, etc.; I’m just surprised that the average American views them so positively. Some of the differences by social class and party ID are interesting too.

Posted by Andrew Gelman at 05:14 PM | Permalink | Comments (0)

Two newspapers, two sides to the story. See here.

Posted by Andrew Gelman at 04:51 PM | Permalink | Comments (0)

Early in Ethan’s life, my wife sensed that he wasn’t getting as much sleep as some people recommend, and as a flyer on the bulletin board of our pediatrician’s office suggested. (That’s the foundation of good parenting: flyers.) I wasn’t sure she was right, so we decided to count how much he slept. A friend (and fellow nerd) had mentioned a website that they used to keep track of their daughter’s activities. I couldn’t remember it, but a little Googling led me to Trixie Tracker.

Trixie Tracker is a site invented by a stay-at-home dad. It allows you to keep track of basically everything your kid does — sleeping, breastfeeding, bottles, medicine, poop. The part I liked best was the graphs. Although the website will dump your data into a file, allowing you to manipulate them yourself, I just relied on the canned graphs that the website produces.

Below is an example. Each chart covers roughly a month of Ethan’s life up until a month ago, when we stopped keeping track. The charts capture the probability that Ethan was asleep during each 10-minute increment within the average 24-hour period from that month. The darker the area, the higher the probability that Ethan was asleep.

Ethan went from a typical newborn with an irregular schedule, to more consistent sleep at night with some lighter areas indicating feedings, to uninterrupted sleep for 11-12 hours every night with, after a few weeks or so, consistent naps. This is not because of any herculean Ferber-esque efforts on our part. We’re just very, very, very, very lucky.

The upshot? My wife was right. And Trixie Tracker is cool.

I welcome humor at my nerdy expense in the comments.

Posted by John Sides at 09:58 AM | Permalink | Comments (4)

Magneto does a video-reply to Lee’s new series of posts …

(taken from this)

Posted by Henry at 09:52 PM | Permalink | Comments (0)

I don’t really know what to make of these survey results on which candidate is a riskier or safer choice, but it’s interesting.

But my favorite recent question wording example remains this one reported in the New York Times from a CBS News poll in 2006:

Should U.S. troops stay in Iraq as long as it takes to make sure Iraq has a stable democracy, even if it takes a long time, or leave as soon as possible, even if Iraq is not completely stable?

What interested me about this particular survey item was not so much whether it is “slanted” but rather that it contained such a big implicit assumption (that if the troops stay long enough, that Iraq will have a stable democracy), and it seemed that the survey writers, the people at CBS, and the people at the New York Times all just took the assumption for granted. At least, I didn’t see any comments on it.

Posted by Andrew Gelman at 09:11 PM | Permalink | Comments (0)

The Washingtonpost.com, Quinnipiac, and the Wall Street Journal have a new poll, which is advertised as showing this: “McCain Makes Significant Gains in Four Key Battleground States.” (Here is virtually the same headline at the Wall Street Journal and Quinnipiac. I would bet both the Washingtonpost.com and the WSJ were basically transcribing Quinnipiac’s press release.) At least one prominent left-wing blog, Talking Points Memo, calls it a “reality check.”

This analysis is wrong. For two reasons:

1) In 2 of these 4 key states, there has been no meaningful change, given the inherent sampling error in polls. In both Michigan and Wisconsin, McCain’s share is unchanged; Obama’s is down 2 points. See here.

2) Repeat after me: ONE POLL DOES NOT MAKE A TREND. Look at Pollster.com’s poll aggregations for another of these four states: Minnesota.

Only in Colorado does this poll’s numbers appear to conform to a trend. See again Pollster.com.

This kind of analysis may generate clicks for the Washingtonpost.com — which, it must be noted, is largely separate from the Washington Post and its polling outfit — and for the WSJ.com, but this analysis is terrible.

Posted by John Sides at 01:56 PM | Permalink | Comments (2)

A few days ago I was reading an article in a psychology journal — I’ll post an item about it sometime soon — and when I came to the main explanatory variable in the analysis, I was awe-struck by its sheer weirdness. This wasn’t the first time I’d had that reaction — shaking my head and asking myself “How do people come up with stuff like this?” — to something I was reading in a social science journal. I’m not referring here just to some idea that comes from outside the prevailing paradigm, seems counterintuitive, etc. I’m talking about stuff that seems to come from outer space.

It is in this context that I am pleased to announce the opening of a major new initiative of the vast “Monkey Cage” empire, which we have dubbed the Truly Strange Research Hall of Fame. To get things off with appropriate fanfare (Ta Da!), let me introduce the TSRHoF’s inaugural inductee:

Middlemist, R.D., E.S. Knowles, and C.F. Matter. 1976. Personal space invasions in the lavatory: suggestive evidence for arousal. Journal of Personality and Social Psychology 33 (May): 541-546.

ABSTRACT: The hypothesis that personal space invasions produce arousal was investigated in a field experiment. A men’s lavatory provided a setting where norms for privacy were salient, where personal space invasions could occur in the case of men urinating, where the opportunity for compensatory responses to invasion were minimal, and where proximity-induced arousal could be measured. Research on micturation indicates that social stressors inhibit relaxation of the external urethral sphincter, which would delay the onset of micturation, and that they increase intravesical pressure, which would shorten the duration of micturation once begun. Sixty lavatory users were randomly assigned to one of three levels of interpersonal distance and their micturation times were recorded. In a three-urinal lavatory, a confederate stood immediately adjacent to a subject, one urinal removed, or was absent. Paralleling the results of a correlational pilot study, close interpersonal distances increased the delay of onset and decreased the persistence of micturation. These findings provide objective evidence that personal space invasions produce physiological changes associated with arousal.

The TSRHoF is hereby open for business and and eager to accept nominations for new honorees. All that’s required is a full citation. As Self-Appointed Curator of the TSRHoF, I’ll take care of the rest.

Posted by Lee Sigelman at 01:13 PM | Permalink | Comments (6)

The traffic accident that John mentioned makes me think of some more general issues of retrospective and prospective decision making.

It often seems to happen that punishments for reckless driving are much less severe than the effect of the crime itself. (Even being “slightly injured” in a car crash has gotta be a personal loss of much more than $50, not even counting hospital costs.) This is particularly striking given that not every offender is caught, so you might think that punishments would be higher for their deterrent value.

Why are the punishments so low? One reason is that many of the legislators who write the laws and judges who decide sentencing are themselves dangerous drivers at times, and I suspect that it can be easier for them to identify with the criminals than the victims. (If Gary Larsen were writing the laws, there’d probably be a death penalty for running over a dog.)

But I think there’s something deeper going on, having to do with retrospective and prospective decision analysis. In the driving example, it goes like this.

1. Suppose somebody (e.g., Dick Cheney) is driving dangerously but nobody is hurt, or not seriously. Then the response is that no serious harm was done—it’s just one of those things—so no point in having a big punishment.

2. Suppose somebody (e.g., Ted Kennedy or Laura Bush) is driving dangerously and seriously injures or kills someone. Then the response is that it’s a terrible tragedy but very bad luck, so what is gained by seriously punishing the driver.

The issue is that deaths and serious injuries are also rare—even if you drive recklessly, it’s extremely unlikely that you’ll kill someone in any given outing. So you’re stuck between punishing the almosts and might-have-beens or really laying down the hammer on the serious cases. No option seems quite right. Although I guess in this case the pedestrian will do all right because he’ll probably sue the driver for a couple of million dollars.

Posted by Andrew Gelman at 10:44 AM | Permalink | Comments (1)

What do you say when you strike a pedestrian with your car, the pedestrian goes up onto and then off of the hood, you continue driving for a block until a bicyclist cuts you off and forces you to stop, and then the police cite you for failing to yield to a pedestrian, who, in this case may be hurt?

If you’re Robert Novak, you say “I didn’t know I hit him…I feel terrible…[But] he’s not dead, that’s the main thing.”

Posted by John Sides at 09:10 AM | Permalink | Comments (2)

A new book, The American Voter Revisited, argues the following:

The American Voter Revisited re-creates the outstanding 1960 classic The American Voter—which was based on the presidential elections of 1952 and 1956—following the same format, theory, and mode of analysis as the original. In this new volume, the authors test the ideas and methods of the original against presidential election surveys from 2000 and 2004. Surprisingly, the contemporary American voter is found to behave politically much like voters of the 1950s.

Its authors are Michael Lewis-Beck, William Jacoby, Helmut Norpoth, and Herb Weisberg.

The book is profiled today in this Washington Post piece. The article discusses debates among political scientists about the quality of citizens’ political decisions. It includes this piquant quote from Sam Popkin, who tends to disagree with the book’s pessimism:

“If I say to you, ‘What did the guy you didn’t marry say to you in bed?’ ” and you can’t remember, “does that mean you didn’t enjoy it?”

Who said political science wasn’t sexy?

Posted by John Sides at 08:56 AM | Permalink | Comments (3)

With a presumptive black presidential nominee, there will be plenty of speculation about the “Wilder Effect”: when (presumably white) survey respondents tell interviewers that they support a black candidate but really intend to support the opposing white candidate. The effect is named for Douglas Wilder, whose margin of victory in the 1989 Virginia gubernatorial race was closer than the polls suggested. In this election, Hillary Clinton’s surprise victory in the New Hampshire primary raised the possibility that the Wilder Effect may apply to Obama. I discounted this possibility here with some quickie data analysis.

Now comes political scientist Dan Hopkins with a more definitive answer:

The 2008 election has renewed interest in the Wilder effect, the gap between the share of survey respondents expressing support for a candidate and the candidate’s vote share. Using new data from 133 gubernatorial and Senate elections from 1989 to 2006, this paper presents the first large-sample test of the Wilder effect. It demonstrates a significant Wilder effect only through the early 1990s, when Wilder himself was Governor of Virginia. Although the same mechanisms could impact female candidates, this paper finds no such effect at any point in time. It also shows how polls’ over-estimation of front-runners’ support can exaggerate estimates of the Wilder effect. Together, these results accord with theories emphasizing how short-term changes in the political context influence the role of race in statewide elections. The Wilder effect is the product of racial attitudes in specific political contexts, not a more general response to under-represented groups.

Below I reprint part of one of his graphs. It shows the gap between the candidate’s expected performance in the polls and their actual performance on Election Day. The blue trend line shows that while in the early 1990s black candidates did better in polls than at the polls, this “gap” disappeared soon thereafter.

Hopkins investigates the 2008 polls and actually finds the opposite of a Wilder Effect: on average, Obama does a little better than the polls predicted.

The mean polling-performance gap was 1.4 percentage points for Senator Clinton, and the reverse for Senator Obama. This estimated mean is not at all sensitive to particular polls or states: if we remove the observations for five randomly chosen states at a time, we still observe that Senator Obama’s election-day performance was better than his polling on average in every one of the 10,000 simulations. This is yet more evidence that the Wilder effect, strong in the early 1990s, is strong no longer.

There is much more in the paper. Find it here.

Posted by John Sides at 01:34 PM | Permalink | Comments (6)

Jerome Holtzman (pictured here with Don Zimmer — a great photo), the “dean” of baseball writers, inventor of the “save” statistic, and member of the Baseball Hall of Fame, has passed away. No crying in the press box.

Here is the New York Times obituary.

[Cubbies’ hat tip to Paul Wahlbeck]

Posted by Lee Sigelman at 10:28 AM | Permalink | Comments (2)

The idea that how a survey question is asked affects – sometimes decisively – the way it is answered is nothing new. (The best source on issues of question-framing is still Schuman and Presser’s Questions and Answers in Attitude Surveys.) The very same thing holds in doing research. Here’s a good case in point.

Wal-Mart and Toys R Us announced this spring that they will stop selling plastic baby bottles, food containers and other products that contain a chemical that can leach into foods and beverages. Even low doses of the chemical (bisphenol A, or BPA) are linked to prostate and mammary-gland changes in laboratory animals that were exposed as fetuses and infants. The big retailers are responding to the fears of parents, and Congress is considering measures to ban the chemical.

But is there enough evidence of harmful health effects on humans? One of the eyebrow-raising statistics about the BPA studies is the stark divergence in results, depending on who funded them. More than 90 percent of the 100-plus government-funded studies performed by independent scientists found health effects from low doses of BPA, while none of the fewer than two dozen chemical-industry-funded studies did.

That’s from an article by David Michaels, an epidemiologist who teaches environmental health policy at, ahem, the George Washington University School of Public Health, in the Washington Post (July 15, 2008).

So why does this “funding effect” consistently emerge? According to Michaels, “Within the scientific community, there is little debate about the existence of the funding effect, but the mechanism through which it plays out has been a surprise.”

The simplest explanation, of course, is that companies hire researchers to produce shoddy studies that whitewash their products. But as Michaels notes, “Such scientific malpractice does happen, but close examination of the manufacturers’ studies showed that their quality was usually at least as good as, and often better than, studies that were not funded by drug companies.

Rather, it turns out that industry researchers rely on various “tricks of the trade,” including “testing your drug against a treatment that either does not work or does not work very well; testing your drug against too low or too high a dose of the comparison drug because this will make your drug appear more effective or less toxic; publishing the results of a single trial many times in different forms to make it appear that multiple studies reached the same conclusions; and publishing only those studies, or even parts of studies, that are favorable to your drug, and burying the rest.”

In short, the problem lies more in the framing of the research question and the presentation of the results than it does in the conduct of the research per se. “As long as sponsors of a study have a stake in the conclusions,” Michaels argues, “these conclusions are inevitably suspect, no matter how distinguished the scientist.” And the solution, he continues, lies in de-linking sponsorship and research.

One model is the Health Effects Institute, a research group set up by the Environmental Protection Agency and manufacturers. HEI has an independent governing structure … [and] conducts studies paid for by corporations, but its researchers are sufficiently insulated from the sponsors that their results are credible.

Posted by Lee Sigelman at 06:55 AM | Permalink | Comments (1)

At our Red State, Poor State, Rich State, Poor State blog, we answer some questions about these graphs on voter and congressmember ideologies that we posted here a couple days ago.

Posted by Andrew Gelman at 10:16 PM | Permalink | Comments (0)

These pictures are based on data from the post-election wave of the 2006 Cooperative Congressional Election Study. Respondents took the survey via the internet and typed in their answers to these questions: “What is the most important problem facing the country?” and “What have been the most interesting stories in the news this past week?” The former is at top and the latter below it.

The most important problem question reveals nothing surprising. Iraq was mentioned most frequently. The most interesting news stories question generated lots of mentions of the election results, the Democratic takeover, and also Rumsfeld’s resignation of Nov. 8.

Fun, eh? Make these pictures yourself at Wordle. (Hat tip to Henry.)

Posted by John Sides at 09:55 AM | Permalink | Comments (0)

Noah Kaplan, an assistant professor of political science at the University of Houston puts forth this interesting argument:

Traditionally, election forecasters focus on two key “structural” factors: the popularity of the current president and the state of the national economy (these factors are considered structural to the extent that they are not a function of the candidates or the campaign). Based on quantitative analyses of post-WWII presidential election results, the percent of the popular vote received by the challenger increases the worse the state of the economy and the less popular the sitting president. Given the relatively poor state of the economy (and the public’s dismal perceptions of the economy) and President Bush’s record low job approval ratings, it would be reasonable to expect that the Republican Presidential candidate would be running significantly behind in the polls. However, many recent national polls place McCain within 3% of Obama. Does McCain really have a shot at winning the election?

A Time Magazine poll conducted between June 19th-June 25th asked two questions that shed light on this question:

1. Which candidate – Barack Obama or John McCain – would best be described by each of the following statements [note: order of candidates and statements randomized]:

e. Best able to handle the economy

2. On another topic, if Republican John McCain is elected, do you expect that he would mainly continue the policies of President Bush, or is he mainly independent of Bush?

1. Mainly continue Bush policies
2. Mainly independent of Bush
3. (VOL) Both/neither/No answer/Don’t know

Of the 805 likely voters polled, 525 respondents articulated a candidate preference and said that there was “No Chance” that the respondent might change his/her mind about voting for his/her preferred candidate. In other words, the remaining 280 likely voters (35%) can be categorized as swing voters.

Surprisingly, despite the relatively poor state of the current economy under the Bush administration, voters see little difference between McCain and Obama relative abilities to handle the economy:

The second question helps us understand why almost as many likely voters think McCain is more capable of managing the economy than Obama as vice-versa despite the current economic downturn. A plurality of likely voters think McCain is “mainly independent of Bush.” Even better for McCain, a smaller fraction of swing voters than all likely voters tie him to Bush.

As you might expect, the correlation between the two questions for all likely voters is moderately strong at 0.51 (though its only 0.26 among swing voters): respondents who see McCain as independent of Bush are more likely to say that McCain would do a better than Obama of handling the economy than are respondents who expect that McCain would mainly continue the policies of President Bush.

The above breakdown suggests that McCain is less burdened by Bush’s economic record than might be anticipated; Obama is going to have a difficult time “tarring” McCain with the economic failures of the Bush administration. To the extent that the election is about the economy, Obama needs to be seen as the superior manager. But the public will not assume this based upon the current administration’s failures alone. Obama needs to be proactive and emphasize an economic program that is seen as a credible alternative to that offered by McCain and the Republicans. Otherwise, McCain has a real chance of winning the swing voters.

Posted by David Park at 07:25 AM | Permalink | Comments (3)

[Hat tip to Bob Sigelman]

Posted by Lee Sigelman at 11:31 AM | Permalink | Comments (1)

The chart summarizes the starting salaries for lawyers who graduated from law school in 2006. One reason the bi-modal structure is so jarring is that it demonstrates that measures of central tendency, such as average or median, are not necessarily reliable guides for law students’ future earning power. In conventional labor markets, that disconnect is rare.

In 2006, 44% of graduates received entry-level salaries in the $40K to $60K range; the second mode, over on the right side of the graph, was at $145K. This makes for a huge salary gap, with relatively few entry-level lawyers making salaries between the two extremes.

That’s from a post by Willilam Henderson over at the Empirical Legal Studies blog (July 18, 2008), here.

What makes the 2006 distribution even more striking is that it’s such a departure from the distribution just 15 years earlier:

What’s going on here? What happened between 1991 and 2006? Henderson sees this trend as an unintended consequence of the operation of the “Cravath system.” That’s the strategy of

builid[ing] human capital by hiring the best students from the best schools and providing them with the best training. At the end of a multi-year tournament, the best associates are promoted to partner. …In recent years, however, the surge in demand for corporate legal services has outstripped the supply of one key input - elite law school graduates. The ensuing salary wars have significantly increased the cost structure of large corporate law firms and undercut clients’ willingness to pay for associate training. These trends are unsustainable. More significantly, clients are unhappy and searching for ways to control costs.

For a paper by Henderson analyzing this situation and offering some ideas about how to avoid it, click here or follow the link in his post.

Posted by Lee Sigelman at 09:45 AM | Permalink | Comments (0)

John and Eric’s op-ed in the LAT got picked up in a frontpage diary at Kos by BarbinMD as an example of how the mainstream media disparage the netroots. The offending bit in BarbinMD’s eyes was the following.

To determine just how polarized blog readers are, we constructed a measure of political ideology by drawing on blog readers’ attitudes toward stem cell research, abortion, the Iraq war, the minimum wage and capital gains tax cuts. Using this measure, we then arrayed respondents from left to right. Here’s what we found.

Readers of liberal blogs were clustered at the far left ….

(The op-ed goes on to say that “readers of conservative blogs were bunched at the far right. There was little, if any, overlap between them on these issues.”)

BarbinMD responds

What does “the far left” mean? Here’s the attitudes of Americans as a whole on these issues:

and then goes on to list a series of opinion polls suggesting that a majority of Americans are against the Iraq war, favor stem cell research and so on. She concludes:

We reflect the majority opinion of this country on pretty much every issue, yet the media continues to pretend that we’re the far left, the lunatic fringe. They’re still unwilling to admit the obvious…we are the mainstream.

I didn’t write the op-ed, although I did co-author the research that the op-ed is based on. But BarbinMD’s interpretation of what the op-ed (and the research) is saying is mistaken. The reasons why are much easier to understand if you look at a graph of the data (unfortunately, the LAT, like most other newspapers, doesn’t usually print graphs of any complexity).

The first thing to note is that when John and Eric talk about readers of liberal blogs being ‘clustered on the far left,’ they are not claiming that these readers are part of the ‘far left’ in the sense that they are political extremists or members of the ‘lunatic fringe.’ They are saying that readers of liberal blogs are clumped together at the far left end of the scale (the specific measure) that they are using. This isn’t a pejorative description or a suggestion that these readers are crazy. It’s a claim about the data.

The simplest way to understand the data is as follows. With regard to this particular set of issues, readers of both liberal and conservative blogs are far more ideologically consistent than non-blog readers. That is to say that readers of liberal blogs are much more likely than others to hold left-leaning positions on all or most of the issues we have measures for. The surveyed readers of liberal blogs were much more likely to oppose the Iraq war and support stem cell research and be in favor of abortion rights and so on, all at the same time. Similarly, readers of conservative blogs are much more likely to favor the Iraq war, and to be opposed to stem cell research and abortion rights and so