This document describes the HUDSON INSTITUTE FILEs that were created and used in The Affordable Housing Stock: A Longitudinal Analysis, 1985-2013. The data source is the American Housing Survey (AHS), sponsored by the U.S. Department of Housing and Urban Development (HUD) and conducted by the Census Bureau. The document identifies the AHS variables that were the basis of the analysis, and defines the variables that were added in the course of the analysis.
The AHS sample was drawn in 1985; 66 percent of the units come from this original sample. Residents of these units were re-interviewed every two years through 2013. In addition, new units were added to the sample in each survey, to present units that had entered the housing stock through new construction or by other means. The remainder of the sample comes from these additions between 1985 and 2013, the end year of our analysis and the last year and the last year of data collection for this sample. (A new sample was drawn for 2015.)
The first HUDSON INSTITUTE FILE, hudson_institute_file_1, contains 95,444 observations and 2,164 variables. This file merges all 15 AHS national surveys based on the 1985 survey and extracts key AHS variables. New variables were created for use in the merging and cleaning of the individual AHS files for each survey. We also have appended relevant variables from HUD’s Housing Affordability Data System (HADS) and the Census Bureau’s Where Did They Go (WDTG) file, which contains historic information on all units that are part of the national sample. Finally, we eliminated units that were part of supplemental samples used in some but not all of the AHS surveys.
The second HUDSON INSTITUTE FILE, hudson_institute_file_2, contains 65,540 observations and 3,728 variables. The second file modifies the first by eliminating cases that were absent from too many AHS surveys, cases with too much missing information on affordability, and cases that were defective in other respects. It also standardizes descriptive variables, based on 2013 values, or the most recent values if 2013 data are missing. Further, the file includes a new bedroom variable that reduces survey-to-survey variation in the bedroom count, creates a variable identifying assisted project-based housing, and creates various special variables for our analysis.
Hudson_institute_file_1 is designed for research who want to conduct different analyses than we have, but want to avoid the work of merging the files and extracting the data. Hudson_institute_file_2 is available to researchers who want to conduct longitudinal analyses of housing policy issues and take advantage of some of the special features developed in the course of the MacArthur study.
We want to express our gratitude to the John D. and Catherine T. MacArthur Foundation for suggesting this project and supporting the work.