We thank Peter Egger, Eric Strobl, participants at the Fall 2004 Mid-West International Economics Group meeting and participants at the DIW/GEP Workshop on FDI and International Outsourcing. Any errors or omissions are the responsibilities of the authors. The views expressed herein are those of the author(s) and do not necessarily reflect the views of the National Bureau of Economic Research.

-We introduc e a simple allocation-of-tim e model to explain the high school athletic participatio n choice and the implication s of this choice for educationa l and labor market outcomes. Four differen t factors that could explain athletic participatio n are identi ed in the context of this model. A variety of tests of the model are provided using two data sets: the National Longitudinal Survey of Youth and the National Longitudinal Study of the High School Class of 1972. We nd some evidence that athletic participatio n directly affects wages and educationa l attainment. However, much of the effect of athletic participatio n on wages and educationa l attainment appears to re ect difference s across individual s in ability or value of leisure.Sports and other forms of vigorous physical activity provide educational experience which cannot be duplicated in the classroom. They are an uncompromising laboratory in which we must think and act quickly and efficiently under pressure and then force us to meet our own inadequacies face-to-face and to do something about them, as nothing else does. . . . Sports resemble life in capsule form and the participant quickly learns that his performance depends upon the development of strength, stamina, self-discipline and a sure and steady judgment.-Supreme Court Justice Byron White

We reconsider the effect of very low birth weight classification on infant mortality. We demonstrate that the estimates are highly sensitive to the exclusion of observations in the immediate vicinity of the 1,500-g threshold, weakening the confidence in the results originally reported in Almond, Doyle, Kowalski, and Williams (2010).

This study uses Monte Carlo simulations to demonstrate that regression-discontinuity designs arrive at biased estimates when attributes related to outcomes predict heaping in the running variable. After showing that our usual diagnostics are poorly suited to identifying this type of problem, we provide alternatives. We also demonstrate how the magnitude and direction of the bias varies with bandwidth choice and the location of the data heaps relative to the treatment threshold. Finally, we discuss approaches to correcting for this type of problem before considering these issues in several non-simulated environments.