Experimental Poverty Measures, 1998: Public-Use Dataset Notes

These are notes for analysts using the Public-Use SAS dataset or
ASCII dataset associated with the US Census Bureau 1998 update to the
P60-205 report on Experimental Poverty Measures, 1990-1997. 

There are five files available at the Census ftp site: 
1. povpu98.ssd01  2. povpu98.sd2  3. povpu98.txt  4. pov98in.sas
and 5. pov98in.lst
These files can be accessed via ftp at following census ftp site:  
ftp.census.gov
Use anonymous as the user name, and your email address as the 
password. Within the ftp site the files are located in the folder: 
/pub/housing/povmeas/

The SAS dataset, povpu98.ssd01 was created using SAS in UNIX. Also 
available is povpu98.sd2, a PC-SAS version of the dataset. Contained 
in the SAS datasets are variables used to construct most of the 
marginal and experimental poverty measures contained in the P60-207 
report, but now updated to 1998. Also included are variables 
necessary to replicate the cross-tabulations between marginal and 
experimental poverty measures and population characteristics 
displayed in that report. For details on the construction  of these 
measures and their component elements, please refer to the P60 
report itself and Appendix C of the report.

All variables in the public use SAS datasets have variable labels, 
and, where appropriate, value labels. Household, family, and person-
level id variables are also contained in the dataset to allow 
analysts to re-merge the file with the 1999 March CPS Public-Use 
file from which the dataset was created.

Povpu98.txt is a space-delimited ASCII version of the file. The SAS 
file pov98in.sas, also provided, reads in the ASCII file, and, for 
illustrative purposes, also displays the final SAS datasteps used to 
create the experimental poverty measures already contained in the
datasets. These steps are shown to help analysts replicate the 
experimental poverty measures and to provide guidance for those who 
wish to appropriately recombine various elements (i.e., thresholds 
and income definitions) to view alternative poverty measures.

It is important to note that many of the poverty rates generated using 
this public-use SAS dataset differ slightly from those contained in 
the 1998 update to the P60 report. This is due to the fact that some 
public-use variables (such as income and taxes) are top-coded, and 
some values for the geographic variables used to construct geographic 
adjustments are suppressed. For these reasons, poverty rates produced 
using the public use dataset are also available at this site in the 
file pov98in.lst. Users can check their numbers with those in this 
output file. Also due to these changes, data users wishing to recreate 
standardized experimental poverty measures will need to change the 
appropriate adjustment factors so that the experimental measures once 
again match the official measure. Finally, users should note that the 
only difference between the variables in the SAS data files and the 
ASCII file is that the weighting variable, marsupwt, is in 1000s in 
the SAS data files.