Todd E. Elder, John H. Goddeeris, and Steven J. Haider, "Isolating the Roles of Individual Covariates in Reweighting Estimation", Journal of Applied Econometrics, Vol. 30, No. 7, 2015, pp. 1169-1191. The data used in the article are from the 1 percent PUMS file of the 2000 Census data distributed by IPUMS USA (Ruggles et al. (2010)). We analyze the black-white wage gap among males aged 25-59 employed in the civilian labor force. Given the large sample sizes available, we use a 10% random sample and exclude observations with missing data on any of the five covariates (education, potential experience, marital status, and occupation). We also exclude observations with hourly wages below $1 and above $3000. Our final analysis sample consists of 213,908 observations for whites and 23,945 observations for blacks. We have used the following IPUMS variables, as described on the IPUMS website (https://usa.ipums.org/) incwage annual wage and salary income uhrswork usual hours of work wkswork1 weeks worked last year educd educational attainment age respondent age marst categorical marital status occ 3-digit numeric variable representing the person's primary occupation ind 3-digit numeric variable representing the industry of the person's occ hispand classifies persons of Hispanic origin raced classifies person's race The comma-delimited ASCII file "egh-data.txt" includes each of these variables on a single line per observation, with commas separating each variable value. There are 237853 lines. The Stata file "egh-code.do" includes Stata commands for recoding the raw Ruggles et al. (2010) variables into the variables that we use in the paper. Both of these files are ASCII files in DOS format. They are zipped in the file egh-files.zip. Unix/Linux users should use "unzip -a". Please address any questions to: Todd Elder Dept. of Economics Michigan State University 110 Marshall-Adams Hall East Lansing, MI 48911 USA telder [AT] msu.edu