Experimental Poverty Measures, 2004: Public-Use Dataset Notes These notes are for analysts who use the public-use file that contains alternative poverty estimates for calendar year 2004 and other variables related to poverty measurement. Corresponding data based on the Census Bureau's internal data files may be found at http://www.census.gov/hhes/www/povmeas/tables.html (accessed 6 July 2006). The estimates included in these files are an update of the estimates in the report P60-227 (Alternative Poverty Estimates in the United States: 2003 -- available at http://www.census.gov/hhes/www/poverty/altpovest/altpovestrpt.html) that were based on recommendations from a National Academy of Sciences (NAS) panel. Three files are available from the Census Bureau's poverty measurement 2004 dataset FTP site: 1. povpu04.sas7bdat 2. povpu04.sas 3. povpu04.lst The SAS dataset, povpu04.sas7bdat, was created using SAS version 8.2 on a UNIX platform. Contained in the SAS dataset are variables used to construct these experimental poverty measures. Also included are variables necessary to replicate the cross-tabulations between experimental poverty measures and selected population and geographic characteristics displayed in P60-227. For details about the construction of the measures and their component elements, please refer to the P60-227 report (referenced above) and to P60-205, Experimental Poverty Measures: 1990 to 1997 (available at http://www.census.gov/prod/99pubs/p60-205.pdf), especially Appendix C. All variables in the public-use SAS dataset 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 2005 Current Population Survey Annual Social and Economic Supplement (CPS ASEC) public-use file from which the datasets were created. The SAS program povpu04.sas reads in the SAS dataset, and, for illustrative purposes, also displays the final SAS data steps used to create the experimental poverty measures already contained in the dataset. (The recodes testpoor1 - testpoor13, created within the program, replicate poor1 - poor13 which are already on the file.) 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. INCOME VARIABLE AND TOPCODED VARIABLE CAVEATS: It is important to note that many of the poverty rates generated using these public-use SAS datasets differ slightly from those shown in Census Bureau publications. These differences occur because some public-use variables (such as the variables for total income, income by source, and taxes) are topcoded to protect respondents' confidentiality. To illustrate, the SAS output file povpu04.lst (available on the FTP site) shows two recode variables for official poverty -- poorpub and poor1. Each recode uses different CPS ASEC variables to construct the official poverty measure. The first, poorpub, is a recode of PERLIS, while poor1 was computed by dividing the CPS family income variable by the threshold variable (FTOTVAL / FPOVCUT). The two methods are conceptually identical; however, poorpub produces output consistent with Census Bureau reports while poor1 does not. The PERLIS variable (used to create poorpub) uses un-topcoded income, and protects respondents' confidentiality by grouping them into broad categories by their ratio of income to poverty, whereas poor1 uses the topcoded income variable, FTOTVAL. Therefore, when computing alternative resource definitions--which by necessity use topcoded variables as components--please bear these differences in mind. GEOGRAPHIC VARIABLE CAVEATS: Three issues with geographic variables warrant the user's attention: a change in sample design in the CPS ASEC public-use file meant that complete information on metropolitan/nonmetropolitan status was not available for every area; a change in geographic concepts prompted a new set of geographic variables; and last, the geographic-adjustment indices for poverty thresholds (geo2) were constructed with estimated metropolitan status information and with appropriate suppression of confidential data. 1. Change in sample design and availability of information on metropolitan status. The 2005 CPS ASEC did not provide metropolitan/nonmetropolitan status for every county because the CPS underwent a transition in sample design from a 1990 Census-based sampling frame to a Census 2000-based sampling frame. See the CPS ASEC public-use file technical documentation at http://www.census.gov/apsd/techdoc/cps/cpsmar05.pdf. Page 5-1 describes the differences between the 2004 CPS ASEC and 2005 CPS ASEC files. The variable used in previous years for metropolitan residence, HG_MSAS, was not available on the 2005 CPS ASEC file. 2. Change in geographic concepts and new geographic variables. In accordance with OMB Bulletin 03-04, the 2005 CPS ASEC file included revised geographic definitions. The CPS ASEC technical documentation (referenced above) includes descriptions of the new geographic variables. The Census Bureau webpage about geographic concepts is available at http://www.census.gov/geo/www/reference.html. OMB Bulletin 03-04 is available at http://www.whitehouse.gov/omb/bulletins/b03-04.html (accessed 6 July 2006). 3. Construction of geographic-adjustment indices for poverty thresholds. Since some areas in the 2005 CPS ASEC did not have metropolitan status provided in the file, their metropolitan status had to be estimated in order to construct the factors used to geographically-adjust the poverty thresholds (variable geo2 in the alternative poverty measure public-use file, povpu04.sas7bdat). A cross-tabulation of GTMETSTA and GTCBSASZ indicated that most of those areas whose metropolitan status was unidentified were likely to be from metro areas. As such, when the geographic adjustment indices were computed (variable geo2), areas with unidentified metropolitan/nonmetropolitan status were treated as if they were metropolitan areas. Metropolitan status was needed to determine which geographic index to assign--the index for "metro" or the index for "non-metro" within the respondent's state. To protect confidentiality, the index does not provide any greater specificity than "metro" or "non-metro." See P60-216, Experimental Poverty Measures: 1999, available at http://www.census.gov/prod/2001pubs/p60-216.pdf, for further information on the methods used to construct the geographic indices for the poverty thresholds. In light of the caveats listed above, the file povpu04.lst shows unweighted counts of the number of people in poverty under each definition, and their corresponding percentages. Users can check their output using these unweighted numbers.