Monthly Archives: May 2011

Does rejection cause physical pain?

Leave a reply

Whenever I read about a study claiming causation using an MRI, the skeptical part of my brain lights up. This piece in the New York Times fashion section is about one such study. One reason I don’t like these studies is they are almost laughably atheoretical–just throw up some MRI plots and talk about  this or that “area of the brain.” To wit:

New research suggests that the same areas in the brain that signify physical pain are activated at moments of intense social loss. “When we sat around and thought about the most difficult emotional experiences, we all agreed that it doesn’t get any worse than social rejection,” said the study’s lead author, Ethan F. Kross, an assistant professor of psychology at the University of Michigan.

Then I got to the part about how they conducted the study (which, by the way, the fashion writer gets bonus points for mentioning this at all).

The image of a bunch of social scientists inflicting pain on laboratory volunteers seems creepily Mengelian, but in this case the experiments involved were markedly less cruel. First off, the subjects weren’t socially rejected by the laboratory technicians — each of the 40 volunteers was recruited specifically because he or she felt intensely rejected as a result of a recent (unwanted) breakup.

Oh, ha ha, that’s a relief–the subjects weren’t “rejected” by the researchers.

Um, wait. Selecting on the dependent variable? That is, isn’t there a problem with making inferences about the population from a sample chosen explicitly for its outcome values?

Well, I thought, maybe that’s just how they do things in psychology, and I don’t really know much. Or, maybe this is just the first step in the broader project: identify the correlation in this subsample, then do some random trials.

Turns out, no. Indeed the study seemed to be motivated by a lack of findings from random trials:

Previous research had shown that while social rejection hurt, it did not activate parts of the brain associated with physical distress. But this team found that when the emotional pain was awful enough, those parts of the brain were affected as well, and in equal part. According to the authors, the emotional pain simulated in previous experiments (being told a stranger dislikes them, looking at rejection-themed paintings) wasn’t powerful enough to elicit a true-to-life response. “We were shocked because no prior research had demonstrated this same connection,” Dr. Kross said.

So the moral is, when randomization fails, cook the numbers by selecting a biased sample? Am I being too harsh?

Causes and consequences of job discrimination

Leave a reply

Two related pieces in the Globe caught my eye. The first is a letter to the editor which asserts, “Unfortunately, particularly in the case of new hires, [age discrimination] is virtually impossible to prove.” The second is an article (printed the same day as the letter, in fact) titled “Transgender bias in workplace costs millions, study says.” Interesting how the letters to the editor are more guarded about causal claims than the “objective” news articles!

I think both pieces have problems. Haven’t there been lots of studies where fake employee resumes are randomly sent to companies? If so, why is it “virtually impossible” to prove? And why wasn’t such a standard adopted in the case of the study on the consequences of “bias”?

What leads to success in negotiations?

Leave a reply

These professionals are ex-FBI agents, labor mediators, divorce counselors. They have learned the rules that help resolve unsolvable standoffs: Don’t lie to a man on a high ledge. Don’t box yourself in with sweeping threats. Don’t tell your adversary to “act like an adult.”

From a May 11 article in the Washington Post. It’s nice (and “good copy”) to think that politicians are just acting like babies, but it strikes me as a little naive. Politics is a pretty high stakes game, so it strains the imagination to think that the people who end up in Congress won’t be good at negotiation, whatever our impressions are or how they behave in front of the cameras. So I’m inclined to think that members of Congress know better than the extra-legislative experts quoted in this piece; but I really just don’t know. I don’t see any suggestion in this piece that negotiation strategies have been evaluated systematically (read: with randomization).

Do high taxes make rich people move away? (ii)

Leave a reply

Via tax.com’s facebook wall, this reuters story is misleadingly titled “Analysis: Americans try to outrun state, local tax hikes.” The first piece of deception is that there is no analysis in the story. The second is that there is no data. Please catch me if I’m just reading too fast, but the only data point seems to be the quote from the accountant that opens the story. The accountant says he’s getting more clients asking about changing their residency for tax purposes.

In contrast, Ezra Klein / The Wall Street Journal did a nice job summarizing a recent empirical study, which concludes the opposite. Look here.

More evidence that the news media has problems accumulating knowledge. Only a couple weeks ago there was coverage of a major empirical study on this question, and now it’s as if it never happened.

Analysis: Americans try to outrun state, local tax hikes

Do tax breaks for nonprofits benefit cities?

Leave a reply

Boston taxpayers benefit when nonprofit salaries get spent at local businesses, and sales taxes are collected when visitors to the city’s museums eat, sleep, and shop in the city. The US Conference of Mayors found that governments see a return on investment of more than $7 in taxes for every $1 invested in educational institutions.

From a letter to the editor by Ford W. Bell and Tim Delaney in today’s Boston Globe. Bell and Delaney write in response to an unsigned editorial endorsing Boston Mayor Thomas Menino’s plan to send mock tax bills to city nonprofits, urging them to contribute to city revenues voluntarily.

That sounds like an interesting experiment, the sending of the letters. But what caught my eye here is the claim that investments in educational institutions (presumably, higher ed institutions) lead to a “more than” 7 to 1 return on investment. Normally, I don’t actually look up studies myself to evaluate them–the point of this blog is mainly to critique how these issues get talked about in the media. But here I’m intrigued. So I’ll look up this study.

…OK, I spent my three minutes searching and can’t find it. So the methodology is going to have to remain a mystery. What we can speculate about is what these letter-writers get out of the publication of their letter. I would guess maybe <50% of those who read this letter, a tiny audience, would find it at all convincing. Factor in the hard economic times, and that number is less. Factor in how many people would retain this information, and it’s less. Factor in how many would use this information in making a decision in the real world, and the number is even less. The conclusion is that letters like this have a very low ROI.

Is politics unpredictable?

Leave a reply

Chris Cillizza and Aaron Blake seem to think so.

The death of Osama bin Laden proves, yet again, that politics is the most unpredictable of businesses.

President Obama’s announcement late Sunday night that bin Laden had been killed in a firefight in Pakistan came as a shock to a political world in which the war on terrorism had receded as a defining issue in recent years.

While it’s too soon to draw broad conclusions about what the death of bin Laden will mean in raw political terms, it’s not too soon to conclude that his demise will re-adjust the political world — at least in the short term — in a meaningful way.

[...]

Remember that moments matter in campaigns, particularly when they are as unpredictable and unexpected as this one. The vote for president is ultimately a vote for the person best equipped to represent and lead the country.

So, um, politics is unpredictable, but moments are what systematically “matter” in campaigns? Seems contradictory. If politics is unpredictable, then nothing really matters. Also, if that last sentence of theirs that I quote were true, then it would be evident in the data. But this is not what the data show. In truth, moments don’t seem to matter much. It’s not entirely mysterious why they don’t, either.

Another person who thinks politics is random is this letter-writer in the Globe the other day, writing about a related topic:

I AM compelled to write this letter because of our government’s seemingly irrational and random reaction to genocide and tyranny around the world. We commit military power in Libya, but sit idly by while hundreds, if not thousands, are brutally attacked or killed in Syria. We have allowed millions to die in Darfur, and even greater millions to perish in Congo, and America allows women and children to be raped, mutilated, and sold into sexual slavery all around the world.

It’s not hard to think of some explanations for the patterns of US intervention. I know less about this topic, but there are many people who study interventions systematically. But it’s probably a reasonable interpretation of events for readers, if they depend on people like Cillizza and Blake to interpret the world for them.

Is salt bad for your health if you’re healthy?

Leave a reply

The findings published yesterday in the Journal of the American Medical Association indicate that healthy people who eat the least amount of sodium don’t have any health advantage over those who eat the most. In fact, they had slightly higher death rates from heart disease.

From the Boston Globe today. Usual caveats (unstated in the article) apply. One question I have is how useful it is to know that healthy people don’t benefit from some intervention.

Did torture lead to the death of bin Laden?

Leave a reply

Before a day had passed, the torture debate had flared. The chairman of the House Homeland Security Committee, Representative Peter T. King, Republican of New York, told Fox News that the success of the hunt for Bin Laden was due to waterboarding. The next morning, Senator Dianne Feinstein, Democrat of California and chairwoman of the Senate Intelligence Committee, said just as flatly that “none of it came as a result of harsh interrogation practices.”

This from a New York Times article this morning on how the killing of bin Laden has revived the debate over the efficacy of torture. The evidence? Some prisoners who were tortured gave useful information; others didn’t.

But a closer look at prisoner interrogations suggests that the harsh techniques played a small role at most in identifying Bin Laden’s trusted courier and exposing his hide-out. One detainee who apparently was subjected to some tough treatment provided a crucial description of the courier, according to current and former officials briefed on the interrogations. But two prisoners who underwent some of the harshest treatment — including Khalid Shaikh Mohammed, who was waterboarded 183 times — repeatedly misled their interrogators about the courier’s identity.

I’m morally opposed to torture, so what follows is mostly just for argument’s sake. But if policy makers want to know whether torture “works”, presumably it shouldn’t be hard to conduct a random experiment to find out.

National security policy seems to raise a lot of interesting causal questions. One question is why the US is using torture. As the Times article mentions, there is a debate about whether torture is effective at extracting information from suspects. (Note the amorality of such a debate; yet such concerns seem to be outside of the domain of causal inference.) If the government were interested in useful information, they possess the means to put such concerns to rest, as I mentioned above: just do an experiment. The policy doesn’t seem as carefully considered as that, though. Instead, and these are just my impressions, the interrogation policies were assembled haphazardly, and were largely a function of machismo and the desire for revenge.