Christopher R. Bollinger and Martin H. David, "I didn't tell, and I won't tell: Dynamic Response Error in the SIPP", Journal of Applied Econometrics, Vol. 20, No. 4, 2005, pp. 563-569. The data used in this study are confidential census data. These data were collected by Kent Marquis and Jeffrey Moore at the Census Bureau. Details on collection and matching can be found in Marquis K. and Moore, J. (1990), "Measurement Errors in SIPP Program Reports," proceedings of the Bureau of the Census 1990 Annual Research Conference, Washington, D.C.: Bureau of the Census, pp. 721-745. Application for permission to use these data can be made to Bureau of Census, Division of Housing and Household Economic Statistics. The data were provided in state files (flfood, pafood and wiaffs,). Additionally, the data set "scram-id" is necessary to merge the validation data back to the full SIPP public use panel. We provide SAS programs to read these data and construct the variables used in the analysis. First run reflfood.sas, repafood.sas, and rewiaffs.sas Rename those data sets flaffs, paaffs, and wiaffs2. Then run affs.sas. The program scramid.sas reads in the id match data. The program affsid.sas merges the public use IDs to the validation data. Finally, the program wave2.sas takes the data, constructs the household valued measures of participation, and merges the validation data to two extracts from the public use SIPP files (which are provided) containing demographic data on the households. The two extracts from the public use sipp are provided here in the files newmiss.txt and finale2.txt. Newmiss.txt provides data on the first wave and has 19878 observations. The variables are: Hhid: the public use household id number Hhern4: total household earnings during the 4th reference month of the first wave. Hhinc4: total household income during the 4th reference month of the first wave. Ms5w1sex: marital status of the head of household at the time of the interview in the first wave. Sex: gender of the head of household during wave1. Finale2.txt provides data on the second wave and has 19294 observations. The variables are: Hhid: the household id number Hhern4: total household earnings during the 4th reference month of the second wave. Headms: marital status of the head of HH during the interview month of the second wave. Headsex: gender of the head of the HH during the interview month of the second wave. The two files include all households at the time of the first and second interview month respectively. They differ in size due to attrition. The validation data are a smaller subset of these households. We provide the full sample extract here to protect the confidentiality of the validation data. The program w12fsout.sas writes out the variables of the analysis for reading into gauss programs. The gauss program w12joint.gau estimates the models. All programs are zipped in the file bd-progs.zip. The two data files are zipped in the file bd-data.zip. The program and data files are ASCII files in DOS format. Unix users should use "unzip -a".