The goal of this project was to construct and implement several adjustment procedures on 1980 census data. Given the tight schedule of the project, it was arbitrarily decided to focus on adjustment of total population at the county level and to use simple methods of adjustment. These criteria led the group to the development of synthetic and regression methods of estimation and to the use of the particular data files to be described.

The project originated in mid-June, 1982 at which time the estimation methods were developed. However, it was not until mid-August that any of the data files were available for our use. One particular file, the PEP first-cut file, consists of the April CPS sample only and it was used in our work only because we were not certain when the final PEP file (April and August CPS samples) would be available.

This report consists of a section summarizing the adjustment methods used. It also contains four attachments. The summary section describes the adjustment methods and documents the data sources used. Briefly, method A1 utilizes demographic estimates of the population at the U.S. level with three different assumptions as to the size of the illegal persons segment. Methods B1 and B4 utilize the PEP estimates at the state and regional levels. Both methods B1 and B4 as well as method Al are synthetic estimators. Methods C1, C2, and C3 are variants of regression models. PEP state data are regressed on independent variables tabulated from the 1980 census as well as variables collected annually from the states.

Attachment I describes some estimation detail regarding methods B1 and B4 while Attachment II lists the regression variables available. Attachment III provides various summaries of the estimates by method and some summary statistics related to the variables used. Attachment IV consists of Minitab computer printouts that relate to methods Cl and C2.