Archive for the 'Research' Category

Uber may reduce alcohol-related auto accidents

5 April 2017

My graduate student advisee, Jessica Peck, has written an excellent paper that examines the impact of Uber on alcohol-related auto accidents in New York City using both differences-in-differences and synthetic control methods.

The Economist has a very nice Daily Chart about her work.

Congrats, Jessica!


Quarter of birth is not a good instrument! Hallo?

17 August 2009

Arrrgh! After 15 years, it’s hard to believe that John Bound’s and my critique of Angrist and Krueger’s quarter of birth instrument (and the whole weak instrument literature that followed, plus a bunch of papers that specifically adress the AK 1991 QJE paper) has yet to make a dent. How can the Economist claim that

In an influential early example of this sort of study, Joshua Angrist of the Massachusetts Institute of Technology (MIT) and Alan Krueger of Princeton University used America’s education laws to create an instrumental variable based on years of schooling. These laws mean that children born earlier in the year are older when they start school than those born later in the year, which means they have received less schooling by the time they reach the legal leaving-age. Since a child’s birth date is unrelated to intrinsic ability, it is a good instrument for teasing out schooling’s true effect on wages.

WRONG!!! Quarter of birth meets neither the relevance (at least in AK’s preferred specification) nor the exclusion restriction assumption required for IV. See Kasey Buckle’s and Dan Hungerman’s excellent paper that, one hopes, is the death knell for any paper that claims quarter of birth is unrelated to outcomes except through the compulsory schooling/school age starting law mechanism. There’s also a very good paper by Rashmi Barua and Kevin Lang that shows that the compulsory schooling/school age starting law mechanism doesn’t meet the monotonicity requirements for a LATE interpretation of the IV estimate of the effects of school entry on outcomes.

Despite all this, Angrist and Pischke in their otherwise excellent Mostly Harmless Econometrics once again belabor the AK results and present them as if they are sensible estimates of the returns to education.

Why is this? How can a result which has for many reasons and by many authors been shown to have problems persist in being held up as a shining example of the usefulness of the IV technique? Part of the answer is that more than any other IV story (and every good IV paper has a story), AK’s story is really good. Incredibly clever. Easy to see in graphs. Believable. So, we (even I) want to think that AK’s instrument is sensible and good. Because if the AK story doesn’t hold, then lots of other IV stories are probably invalid. And with them goes the whole natural experiment movement.

(I posted a slightly different version of this to The Economist website.)

QJPS Style Guide — Required Reading!

26 June 2009

I received today the best style guide for a journal that I have ever received… from politcal scientists, no less.

Here are a few excerpts (and my comments) from the style guide of the Quarterly Journal of Political Science:

Acknowledgments. Please be brief. Avoid hyperbole and effusive expressions of debts of gratitude to family, friends, pets, etc.

HA! I wish they’d also added that one shouldn’t include that craven caveat “All remaining errors are the responsibility of the author,” or somesuch stupid thing. Even worse, in my opinion, is the cryptic “The usual caveats apply.” Duh! Of course the errors are the responsibility of the author. If I give someone comments, they can take them or not… but it’s their name that’s on the byline, not mine.

Put another way — If you’re an economist (or political scientist), I’ll be happy to share authorship on any paper if you’re willing to let me take part of the blame if you’re found to have screwed up. Unless you’re Steve Levitt.

Reporting of Data

  • Please use three significant digits in tables and text.

I agree. This is usually sufficient.

  • We prefer the reporting of standard errors, because QJPS readers can divide. We
    frown upon stars and daggers not only because they are unsightly but also because
    QJPS readers have inalienable rights to choose their own critical values.

YES! I’ve been saying this for years (something I got from Gary Solon in grad school, I think). Ironically, the paper I submitted to the QJPS actually had asterisks in it… but only because we thought political scientists liked them. Now I know that the opposite is true, at least among the serious empirical political scientists.

  • Because the alternative hypothesis β ≠ 0 is not as interesting as β ≥ 0, p-values should be one-tailed in most instances.

Here I disagree. It depends on what’s appropriate to the underlying theory being tested. A priori, I’m not sure you can say that one is more interesting than the other.

One correction, though. If the null is β = 0, then the strong inequality should be used in the alternative, not the weak inequality. What if the coefficient were identically zero, would you reject or not reject the null?

  • In text and tables, avoid abbreviations in the names of variables. Modern typesetters can accommodate upper and lower case fonts and words longer than eight characters, so full descriptive words or phrases are preferred.

    •  Opaque: The coefficient for DINVTSHR is positive.

    •  Clear: The coefficient for Democratic incumbent’s vote share is positive

Again, YES! I hate it when people use variable names in their tables and in the text of their papers. You have to spend all your time reading the paper flipping back between the table that defines the variables (if there even is one) and the table you’re interested in. In my experience, the Industrial and Labor Relations Review is especially egregious in this regard.

So, hats off to Keith Krehbiel and Nolan McCarty, the chief editors of the QJPS. Everyone doing empirical work should adopt these guidelines, whether they’re in political science, economics, demography, public policy, or other fields.