Social Science Statistics Blog

« December 6, 2005 | Main | December 8, 2005 »

7 December 2005

Applied Statistics - Michael Hiscox and Nicholas Smyth

Today, the Applied Statistics Workshop will present a talk by Michael Hiscox and Nicholas Smyth of the Harvard Government Department. Professor Hiscox received his Ph.D from Harvard in 1997 and taught at the University of California at San Diego before returning to Harvard in 2001. His research interests focus on political economy and international trade, and his first book, International Trade and Political Conflict, won the Riker Prize for the best book in political economy in 2001. Nicholas Smyth is a senior in Harvard College concentrating in Government. He is an Undergraduate Scholar in the Institute for Quantitative Social Science. Hiscox and Smyth will present a paper entitled "Is There Consumer Demand for Improved Labor Standards? Evidence from Field Experiments in Social Labeling," based on joint research conducted this summer with the support of IQSS. The presentation will be at noon on Wednesday, December 7 in Room N354, CGIS North, 1737 Cambridge St. Lunch will be provided. The abstract of the paper follows on the jump:

A majority of surveyed consumers say they would be willing to pay extra for products made under good working conditions abroad rather than in sweatshops. But as yet there is no clear evidence that enough consumers would actually behave in this fashion, and pay a high enough premium, to make “social labeling� profitable for firms. Without clear evidence along these lines, firms and other actors (including independent groups that monitor and certify standards) may be unwilling to take a risk and invest in labeling. We provide new evidence on consumer behavior from experiments conducted in a major retail store in New York City. Sales rose dramatically for items labeled as being made under good labor standards, and demand for these products was very inelastic for price increases of up to 20% above baseline (unlabeled) levels. Estimated elasticities of demand for labeled towels, for example, ranged between -0.36 and -1.78. Given the observed demand for labor standards, it appears that many retailers could raise their profits by switching to labeled goods. If adopted by a large number of firms, this type of labeling strategy has the potential to markedly improve working conditions in developing nations without slowing trade, investment, and growth.

Posted by Mike Kellermann at 10:27 AM

Fun with R²

Mike Kellermann

This semester, I have been one of the TFs for Gov 2000 (the introductory statistics course for Ph.D. students in the Government Department). It the first time that I've been on the teaching staff for a course, and it has been quite an experience so far. We've spent the past month or so introducing the basic linear model. Along the way, Ryan Moore (the other TF) and I have had some fun sharing the best quotes that we've come across about everyone's favorite regression output, R²:

Nothing in the CR model requires that R² be high. Hence a high R² is not evidence in favor of the model, and a low R² is not evidence against it. Nevertheless, in empirical research reports, one often reads statements to the effect that "I have a high R², so my theory is good," or "My R² is higher than yours, so my theory is better than yours." (Arthur Goldberger, A Course in Econometrics, 1991)

Thus R² measures directly neither causal strength nor goodness of fit. It is instead a Mulligan Stew composed of each of them plus the variance of the independent variable. Its use is best restricted to description of the shape of the point cloud with causal strength measured by the slopes and goodness of fit captured by the standard error of the regression. (Chris Achen, Interpreting and Using Regression, 1982)

Q: But do you really want me to stop using R²? After all, my R² is higher than all of my friends and higher than those in all the articles in the last issue of APSR!
A: If your goal is to get a big R², then your goal is not the same as that for which regression analysis was designed. The purpose of regression analysis and all of parametric statistical analyses is to estimate interesting population parameters....
If the goal is just to get a big R², then even though that is unlikely to be relevant to any political science research question, here is some "advice": Include independent variables that are very similar to the dependent variable. The "best" choice is the dependent variable; your R² will be 1.0. (Gary King, "How not to lie with statistics,"AJPS, 1986).

So this is old news, right? Maybe not. Quite possibly the thing that has surprised me the most so far is just how much students want R² to tell them how good their model is. You could almost see the anguish in their faces as we read these quotes to them, particularly among those who have taken some statistics in the past. The question I want to throw out is, why is R² such an attractive number? Why do we want to believe it? Maybe our cognitive science colleagues have some insight....

Posted by Mike Kellermann at 5:00 AM