As the DAPE project was being conducted, it became clear that having advance data from Census 2000 sample files was critical to the success of DAPE. However, it also became clear that the formal development, clearance, and release of these data would occur too late to have an impact on DAPE. As a consequence, POP worked with DSCMO and DSSD to develop an advance set of sample data including key variables that would be needed by DAPE analysts.
Development of this special file involved several important steps: 1) obtaining coded place of birth and migration data, 2) creating edit and imputation specifications that would be a scaled-down version of the full edit but follow the same basic philosophy, and 3) a simplified version of the formal weighting scheme developed by DSSD for full sample processing.
Coded responses, reflecting the results of automated coding (99.4% of all place of birth codes and 96% of all migration codes) were obtained from GEO. DSCMO obtained these codes from GEO and merged them into the sample census unedited file (SCUF) to produce a pseudo sample census information file (SCIF). The SCIF was then put through the scaled-down version of the edits to produce a sample census edited file (SCEF). DSSD, using its scaled-down version of the Census 2000 weighting specs, produced weights that were included in each sample record on the SCEF. These procedures were followed to produce records for the housing unit and group quarters populations.
As with Census 2000, a SCIF and SCEF (in ASCII format) was produced for each state. The Data Management Team (DMT) converted each of these files into SAS data sets and merged them into a national file. This file is stored on POPDEC2K in the directory /dec2k/scratch/scef.
Documentation - all documentation for this Census 2000 DAPE file is located in: X:\SHARED\DAPE\Background-Resources
Record layout (SCIF) ALIEN-SCIF.doc
Record layout (SCEF) ALIEN-SCEF.doc
Edit procedures DAPE SPECS REVISED.doc
Weighting procedures Dainvest4.doc