Thursday, May 23, 2013

Jason Richwine: About That Dissertation





Posted by Nicholas Stix

May 20, 2013 12:00 P.M.

About That Dissertation

On being the media's villain of the week

By Jason Richwine

National Review Online

On Tuesday, May 7, I had one of my most productive days as an employee of the Heritage Foundation. Our big report on the fiscal cost of amnesty had just been released, and I packed in 18 radio interviews to promote it.

I expected more of the same on Wednesday. Instead, I found myself unplugging my office phone to avoid pesky reporters, trying in vain to do any real work, and watching helplessly as a public-relations crisis sprang up around me. Two days later I would resign.

I'm telling this story not because I want or expect pity for my personal situation. Rather, it's important for people to understand how hostile the political class can be toward scientific facts that make them uncomfortable. That discomfort is what caused a mainstream policy analyst to be rebranded overnight as a bigoted extremist.

Although my Ph.D. dissertation was about immigration, I was hired by the Heritage Foundation in 2010 to be a jack-of-all-trades quantitative analyst. I worked a little bit on immigration during my time at Heritage, but I developed a specialty in public finance — fair-value accounting for student loans, public-pension reform, teacher compensation, etc.

My frequent co-author Andrew Biggs and I have gotten some press for demonstrating over and over that generous pensions push public-sector compensation above fair-market levels. A teachers' union in Texas even put us on its "Top Ten Most Wanted" list. But even as we attracted this attention, I could still see people's eyes glaze over when I told them it was based on accumulated benefit obligations using fair-value discount rates.

Given all my wonkery, it felt especially strange to be suddenly characterized as an extremist. That happened on Wednesday morning, when the media first reported on my 2009 Harvard dissertation. Entitled "IQ and Immigration Policy," the dissertation obviously deals with some sensitive topics. Media reports grabbed short quotes from the text and presented them as shocking. Some bad words started getting tossed around: eugenics, racism, pseudoscience, and, of course, extremism.

So what is actually in the dissertation? The dissertation shows that recent immigrants score lower than U.S.-born whites on many different types of IQ tests. Using statistical analysis, it suggests that the test-score differential is due primarily to a real cognitive gap rather than to culture or language bias. It analyzes how this cognitive gap could affect socioeconomic assimilation, and it concludes by exploring how IQ selection might be incorporated, as one factor among many, into immigration policy.

I got into all of this because I found the science of mental ability to be fascinating. I wanted to learn more and think about what lessons it might hold for public policy. Doctoral students are told to pick a topic they're sincerely interested in, since they'll be stuck with whatever choice they make for three years or more.

I was not so naïve as to think my topic wouldn't generate controversy. But individual quotes from my dissertation are much more understandable when placed in their full context. For example, this sentence on page 66 has been widely circulated: "No one knows whether Hispanics will ever reach IQ parity with whites, but the prediction that new Hispanic immigrants will have low-IQ children and grandchildren is difficult to argue against."

I don't think someone reading my full dissertation would find this statement objectionable, for two reasons. First, as Chapter 1 makes clear, the simple existence of ethnic differences in IQ is scientifically uncontroversial. (Skeptical readers should consult the American Psychological Association for confirmation.) Such differences are revealed by tabulations of test scores and calculations of arithmetic means. Their existence is no more debatable than the widely publicized ethnic differences in SAT scores. What the differences mean and what causes them are the interesting issues, which I discuss at length.

Second, the prediction that IQ differences will persist over generations does not rely on assumptions of genetic transmission, but rather on observational data from past immigrant waves. The IQ differences have been persistent — for whatever reason — and nothing is happening to the education or socialization of the current generation of Hispanics that gives reason to expect a break with past experience. Therefore, it is literally "difficult to argue against" continued differences in the next generation — unless hope trumps experience, but I doubt my dissertation committee would have found that argument compelling.

Why did I discuss differences between Hispanics and non-Hispanic whites at all? Because the largest portion of the post-1965 immigration wave has come from Latin America. Studies of Hispanic IQ are naturally useful in estimating overall immigrant IQ and its intergenerational transmission.

That last point bears elaborating: There is absolutely no racial or ethnic agenda in my dissertation. Nothing in it suggests that any groups are "inferior" to any others, nor is there any call to base immigration policy on ethnicity. In fact, I argue for individual IQ selection as a way to identify bright people who do not have access to a university education in their home countries.

I realize that IQ selection rubs some people the wrong way, but it can hardly be called "extremist." Canada and Australia intentionally favor highly educated immigrants. My proposal is based on the same principle they use (pick skilled immigrants), but it offers a much better chance for disadvantaged people to be selected.

If the dissertation were taken seriously, its real contribution would be to open a forthright debate about the assimilation challenge posed by the post-1965 immigration wave. Because regardless of what one believes IQ scores really measure, or what determines them, they are undeniably predictive of a wide variety of socioeconomic outcomes that people care about.

We're still waiting for that assimilation debate to start. I am not aware of a single major news outlet that acted as if my results merited real discussion. The reporters scanned the text for damning pull-quotes, giddily pasted them into stories about "extremism" on the right, and presented my statements as self-evidently wrong. Liberal bloggers piled on with ignorant condemnations. Even some conservative supporters of the Schumer-Rubio amnesty eagerly joined the hatefest. At no time did the critics seem to wonder whether what I was saying might be true.

The reason for that is simple. The media were never interested in me or in the substance of my dissertation. They wanted only to use my work to embarrass the Heritage Foundation and, by extension, all opponents of amnesty. It's a familiar formula for "gotcha" journalism: Uncover an "extremist" associated with a mainstream organization, then demand to know how the organization could possibly associate itself with him. Keep turning up the pressure, hour after hour, with "shocking" new revelations.

To see how the furor over my dissertation is so inextricably linked to today's heated debate over immigration, consider that no less a mainstream-media institution than the New York Times reported on some of my dissertation's ideas in 2009. The newspaper's Idea of the Day blog discussed my proposal for IQ selection in neutral terms. No moral panic ensued. What's different now is that immigration reform is at stake, and the whole conversation is hopelessly politicized.

I don't apologize for any of my writing, but I deeply regret that it was used to hurt my friends and colleagues at Heritage. Seeing them struggle on account of me was the most painful aspect of the whole ordeal. I remember a particularly difficult moment when a Heritage spokesman went on Univision to defend the Heritage report. He explained, accurately, that I was just the number cruncher for the study. Here's the question he was given by the host:

So you're telling me that you used the numbers from a man who has written that Hispanics have a low IQ and will have a low IQ for generations. So what makes you think, unless you agree with that premise, what makes you think that his numbers are sufficiently good in order for, for them to be included in your study?

How can anyone respond to a question as absurd as that one?

Claims that my dissertation influenced the Heritage fiscal analysis are completely false. Anyone who reads the Heritage study will discover that the basic framework — adding up government benefits received by immigrants and comparing that sum to the total taxes they pay — was developed by the National Academy of Sciences in 1997. Robert Rector adapted that framework for his 2007 fiscal-cost study, and he chose the same framework again in 2013, when I helped him run the numbers.

In my judgment, the initial criticisms of the Heritage study were not enough to sink it, so the media latched on to my dissertation as a convenient distraction. Better to shoot the messengers than to deal seriously with what they are saying.

Some students at Harvard are now using the same strategy to denounce my dissertation findings. An open letter signed by 23 ethnic student groups contains this gem: "Even if such claims had merit, the Kennedy School cannot ethically stand by this dissertation whose end result can only be furthering discrimination under the guise of academic discourse." It would be difficult to find a more explicit embrace of censorship.

A student petition is currently circulating that calls on the Harvard administration to reject all scholarship based on "doctrines" that the signers don't like. The petition, which at last count had nearly 1,000 signatures, isn't just shameful, it's worrisome. Many of these students will come to positions of national leadership, yet they openly oppose intellectual freedom. Going forward, I wonder what other thoughts they will seek to ban.

The furor will soon pass. Mercifully, the media are starting to forget about me. But a certain amount of long-term damage to political discourse has been done. Every researcher who writes on public policy over the next few years will have a fresh and vivid memory of how easy it is to get in trouble with the media's thought police, and how easy it is to become an instant pariah. Researchers will feel even more compelled to suppress unpopular evidence and arguments that should be part of an open discussion. This is certainly not the way science should be conducted, and it's not the way our politics should be either.

— Jason Richwine was a senior policy analyst at the Heritage Foundation from March 2010 to May 2013.

1 comment:

jeigheff said...

There was an ad campaign here in Texas a couple years ago. The ads (which appeared on buses and billboards)featured photos of smiling Hispanic girl babies (no boy babies) with headlines like "Rocket Scientist", etc. I think there were three variants. (I work at a printing company which printed some of these images as small posters for a local ad agency.)

On one hand, I can't begrudge parents for hoping that their children will do well when they grow up. But it's interesting that some people want you to believe that only Hispanic females will amount to anything, which is what these ads implied.

There can be little doubt that the people who came up with these ads probably don't want to hear what people like Jason Richwine have to say.