Mingliang Li, Dale J. Poirier and Justin L. Tobias , "Do Dropouts Suffer
from Dropping Out? Estimation and Prediction of Outcome Gains in Generalized
Selection Models", Journal of Applied Econometrics, Vol. 19, No. 2, 2004,
pp. 203-225.

The data are taken from "High School and Beyond", a data set collected by
the National Center for Education Statistics. The official website for this
data set is 

   http://nces.ed.gov/surveys/hsb/ 

The data used in the paper are stored in the ASCII file lptdata.data, which
is zipped in lptdata.zip. This file is in DOS format. Unix users should
use "unzip -a".

The data file is rectangular with 12459 lines. Each line represents an
observation. For each observation, we provide 12 variables. Variables are
separated by white spaces. The variables are listed in the following order:

(1)  follow-up year math test score
(2)  dropout dummy
(3)  base year math test score
(4)  female dummy
(5)  white dummy
(6)  father education
(7)  father education unknown dummy
(8)  mother education
(9)  mother education unknown dummy
(10) family income
(11) number of siblings
(12) employment growth rate from 1980 to 1982