SURVEY ATTRITION AND SCHOOLING CHOICES

We use data from three cohorts of the National Longitudinal Surveys of Labor Market Experience and from the Panel Study of Income Dynamics t o study the effect of survey attrition on estimates of statistical mod els of schooling choices. We estimate regressions using data on people ,who always respond to the surveys (stayers) and on people who miss so me surveys (attritors) and test whether the same statistical model des cribes the behavior of stayers and attritors. In general (with a few e xceptions) we find that attrition either has no effect on the regressi on estimates or only affects the estimates of the intercept (and somet imes the coefficients of birth year dummies) and does not affect estim ates of family background slope coefficients.