Introduction to PUMS: Public Use Microdata Sample

1
Introduction to PUMSPublic Use Microdata Sample
Nanda Srinivasan, CTPP Staff Support

• Presentation is based on material
• a. borrowed from Chuck Purvis
• MTC, Oakland, California
• b. PUMS 2000 DVD obtained from Census Bureau
• c. TechTalk Website
• http//www.census.gov/support/PUMSdata.html, and
• d. Cam McIntoshs presentation at the Washington
  CTPP Workshop

2
Introduction Applications

• What is PUMS?
• PUMS Geographic Areas (PUMAs)
• Top-Coding Rounding Issues
• Software to Analyze PUMS

3
PUMS Basics

• Microdata means that this in the actual
  individual record from a person/household
• To ensure confidentiality, the geography of the
  person is hidden by using a large geographic
  reporting unit (PUMA).
• PUMA is Public Use Microdata AREA. A PUMA must
  have a population of at least 100,000 persons.

4
What is PUMS?

• Public Use Microdata Sample
• Individual, disaggregate census long form data
• Geo-coded ONLY at the PUMA level to protect
  confidentiality of respondent
• One-percent PUMS (National File)
• Five-percent PUMS (State File)
• Evolving program, with 100,000 population PUMAs
  starting in 1990
• 1990 5 PUMS for Bay Area Data for 108,000
  households and 292,000 persons

5
PUMS Basics

• 1 sample is a National file and all detail is
  included including all ancestry and age detail.
• 5 sample are state files and some variables
  have been top-coded, or aggregated. For example,
  travel time to work often has a top code or
  maximum value of about 160 minutes. Time leaving
  home is aggregated (varies from 10 minute to
  larger aggregates depending on TOD).
• About one-third of long form records are
  included in the 5 sample.

6
What are PUMAs?

• Public Use Microdata Areas
• 5-percent sample PUMAs (100,000 pop.)
• 1-percent sample super-PUMAs (400,000 pop.)
• Defined by State Data Centers working with
  Regional Data Centers (MTC/ABAG) and others (UC
  Berkeley)
• California PUMAs
• 62 super-PUMAs 233 PUMAs
• Bay Area PUMAs
• 12 super-PUMAs 54 PUMAs

7
(No Transcript)
8
(No Transcript)
9
California PUMAs (5) andSuper-PUMAs (1)
5-Percent PUMAs
1-Percent PUMAs
10
Bay Area PUMAs (5) and Super-PUMAs (1)
5-Percent PUMAs
1-Percent PUMAs
11
Top-Coding Rounding (I)

• Income Top-Coding, use statewide mean for highest
  income groups
• Dollar Amounts (income, housing costs)

No Income 0
1 - 7 4
8 - 999 Round to nearest 10
1,000 - 49,999 Round to nearest 100
50,000 Round to nearest 1,000
12
Top-Coding Rounding (II)

• Age, top-coded at 90, persons are assigned
  statewide mean for persons age 90
• Race, Ethnicity, Ancestry
• Nationwide minimum of 10,000 in group
• Departure time from home-to-work
• 5-minute rounding for 500-1059 AM
• 10-minute rounding for 300-459 1100-midnight
• 30-minute rounding for midnight 459
• Travel times from home-to-work are rounded to
  nearest 5 minutes

13
PUMS 2000 Features

• Occupation based on Standard Occupational
  Classification (SOC)
• Industry based on the North American Industry
  Classification System (NAICS)
• Place of Work will be PUMA-of-Work
• Super-PUMA to Super-PUMA commute flows in
  1-percent PUMS
• PUMA to PUMA commute flows in 5-percent PUMS
• 1990 PUMS had PUMA-of-Residence to County-of-Work

14
Software to Analyze PUMS
SAS www. sas.com 5000
SPSS www. spss.com 1150
Stata www. stata.com 1200
Minitab www. minitab.com 1200
S-Plus www. insightful.com
Systat www. systat.com 1300
Statistica www. statsoft.com 800
Shazam shazam.econ.ubc.ca 400
List of Commercial Free Statistical Software,
at www.statistics.com Also, check out
www.stattransfer.com for conversion utility.
15
Where to Get PUMS?

• Integrated PUMS Project (Univ. Minnesota)
• http//www.ipums.org/
• http//www.ipums.org/census2000/
• University of Virginias GeoStat Program
• http//fisher.lib.virginia.edu/pums/
• Census Bureau PUMS Program
• http//www.census.gov/population/www/cen2000/pums
  .html
• MTC / ABAG / State Data Center

16
How to get a PUMS DVD

• Get it FREE if you are a member of your State
  Data Center?
• BUY it for 70 from the Census Bureau.
  301-763-INFO (4636)

17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)