and from the American Community Survey
* File size [download size, uncompressed size]
Public Use Microdata Sample from Decennial Census 2000
Disc product support page updated 08/26/2003
ASCII text data file processing
The data files do not contain field delimiters or a header record (a first record with fieldnames). Each data file contains two record types ; housing unit and person. The first character of each record denotes its' type. Each housing unit record immediately preceeds all associated person records (zero in the case of a vacant housing unit).
The housing unit and person records contain 266 and 314 characters of data respectively. Since the housing unit records are not padded to equal the lengths of the person records, the data files are not fixed width.
All of the geographic identifier fields (region, division, state, PUMA etc.) are contained in the housing unit records only.
SERIALNO (Housing/Group Quarters (GQ) Unit Serial Number) is the only link between the housing unit and person records.
File Documentation [4.32 MB]
ASCII text data file directory
1% [634 KB], 5% [391 KB] record layouts (Excel format)
Contact your local state data center (SDC) for SAS™ and SPSS™ code.
Public Use Microdata Sample from Decennial Census 1990
ASCII text data files
1990 PUMS 5% (A sample) 1990 PUMS 1% (B sample)
Documentation and code lists
Download [13.2 KB, 60.2 KB] or view record layout 60.2 KB
Download A (5%) sample Public Use Microdata Area (PUMA) code listings [354 KB, 3.22 MB]
Download B (1%) sample Public Use Microdata Area (PUMA) code listings [376 KB, 3.41 MB]
Download other code listings [38.6 KB, 192 KB]
Download SAS program, Usernote
Revised Texas 5% Data file [451 KB, 3.07 MB] This file contains the 13,853 records missing from the end of PUMSAXTX.TXT on the February 1995 re-release of the 1990 5% PUMS. These records are included on the subsequent release in December 1995.
Notes about the ASCII files
Each data file contains two record types ; housing unit and person. The first character of each record is either an H or a P. Each housing unit record is immediately followed by all associated person records (zero in the case of a vacant housing unit).
All of the geographic identifier fields are contained in the housing unit records only.
The data files do not contain field delimiters or a header record (a first record with field names). A header record, even if included, could apply to only one record type.
Records are delimited by a carriage return character plus a line feed character.
Strategy for using PUMS files
DOS program for extracting a PUMA from the state file (CD-ROM data file users only)
Usage: PUMA2DSK StAbbr Sample Puma Output_path
PUMA2DSK CA A 00500 C:\
PUMA2DSK IL B 00100 C:\PUMS\
Output filename: aappppps.TXT
aa = State abbreviation
The input file is read from the current drive:\path\ and the index file is read from \TOOLS\PUMAS\ of the current drive. The output file is named as follows:
ppppp = PUMA code
s = sample
aa = State abbreviation
Public Use Microdata Sample from Decennial Census 1980
ASCII text data files (1980)
Technical documentation for the 1980 Public Use Microdata Sample and earlier releases exist only in hard copy (paper) format. The technical documentation for the 1980 PUMS files is available from the Census Bureau.
Public Use Microdata Sample from Decennial Census 1970 (and earlier)
Data from 1970 and earlier PUMS files is available from the National Archives and Records Administration. Copies of the technical documentation for these files should be available for browsing from many university libraries.