Harmonized Data for the FSRDC Catherine A. Fitch Minnesota - - PowerPoint PPT Presentation

Harmonized Data for the FSRDC

Catherine A. Fitch

Minnesota Population Center & IPUMS University of Minnesota

Overview

I. What is IPUMS?

• II. IPUMS in the FSRDC
• III. Metadata and the FSRDC

What is IPUMS?

IPUMS provides census and survey data from around the world integrated across time and

• space. IPUMS integration and documentation

makes it easy to study change, conduct comparative research, merge information across data types, and analyze individuals within family and community context.

http://ipums.org

1991: Eight Public Use Census Samples All Incompatible!

Relationship Variable (part): 1900 Public Use Sample

72 categories

0 P03 REL RELATIONSHIP TO HEAD COLS 9-11 100 HEAD OF HOUSEHOLD 21336 21.243 108 PARTNER / COHEAD 173 .172 120 WIFE OF HEAD 16665 16.592 128 WIFE OF PARTNER/COHEAD 1 .001 129 SECOND OR THIRD WIFE OF HEAD 3 .003 130 CHILD OF HEAD 46174 45.973 131 STEP-CHILD OF HEAD 755 .752 132 ADOPTED CHILD OF HEAD 103 .103 133 SON/DAUGHTER-IN-LAW 466 .464 136 FOSTER CHILD / FOUNDLING 23 .023 140 HUSBAND / NOT HEAD 17 .017 200 RELATIVE - UNSPECIFIED 23 .023 210 PARENT OF HEAD 920 .916 211 STEP-PARENT OF HEAD 24 .024 213 PARENT-IN-LAW OF HEAD 568 .566 220 BROTHER/SISTER OF HEAD 1325 1.319 221 STEP/HALF BROTHER/SISTER 12 .012 223 BROTHER/SISTER-IN-LAW 688 .685 230 NIECE/NEPHEW 822 .818 232 ADOPTED NIECE/NEPHEW 1 .001 233 NIECE/NEPHEW-IN-LAW 4 .004 237 GRAND NIECE/NEPHEW 15 .015 240 COUSIN 108 .108 243 COUSIN-IN-LAW 1 .001 249 SECOND COUSIN 5 .005 250 AUNT/UNCLE OF HEAD 99 .099 253 AUNT/UNCLE-IN-LAW 2 .002 260 GRANDPARENT OF HEAD 27 .027 261 STEP-GRANDPARENT 1 .001 263 GRAND-PARENT-IN-LAW 2 .002 270 GRANDCHILD OF HEAD 1541 1.534 271 STEP-GRANDCHILD 33 .033

Relationship Variable: 1940 Public Use Sample

23 categories

Relationship Variables: 1960 Public Use Sample

12 categories, excluding redundancies

Relationship Variables: 1980 Public Use Sample

20 unique categories

1991 IPUMS proposal: An integrated database for

1880, 1900, 1910, 1940, 1950, 1960, 1970, 1980, 1990

Harmonized codes Consistent record layout Integrated documentation No loss of information

.

Variable Harmonization

Home Ownership

2012 ACS 1 = Owned with mortgage

• r loan

2 = Owned free and clear 3 = Rented 4 = Occupied without payment of rent B = N/A

Variable Harmonization

Home Ownership

2012 ACS 1 = Owned with mortgage

• r loan

2 = Owned free and clear 3 = Rented 4 = Occupied without payment of rent B = N/A 1960 1% 0 = Owned or being bought 2 = Rented for cash rent 3 = No cash rent 4 = N/A

Variable Harmonization

Home Ownership

2012 ACS 1 = Owned with mortgage

• r loan

2 = Owned free and clear 3 = Rented 4 = Occupied without payment of rent B = N/A 1960 1% 0 = Owned or being bought 2 = Rented for cash rent 3 = No cash rent 4 = N/A 1900 5% 1 = Owned 2 = Rented 9 = Missing/blank

Translation Table

Input

1 = Owned with mortgage or loan 0 = Owned or being bought 1 = Owned 2 = Owned free and clear 2 = Rented for cash rent 2 = Rented 3 = Rented 3 = No cash rent 9 = Missing/blank 4 = Occupied without payment of rent 4 = N/A B = N/A

Translation Table

Input

2012 ACS 1960 1% 1900 5%

1 = Owned with mortgage or loan 0 = Owned or being bought 1 = Owned 2 = Owned free and clear 2 = Rented for cash rent 2 = Rented 3 = Rented 3 = No cash rent 9 = Missing/blank 4 = Occupied without payment of rent 4 = N/A B = N/A

Harmonized

Code Label

Translation Table

Input

2012 ACS 1960 1% 1900 5%

B = N/A 4 = N/A 9 = Missing/blank 0 = Owned or being bought 1 = Owned 2 = Owned free and clear 1 = Owned with mortgage or loan 2 = Rented 4 = Occupied without payment of rent 3 = No cash rent 3 = Rented 2 = Rented for cash rent

Harmonized

Code Label

Translation Table

Input

2012 ACS 1960 1% 1900 5%

00 N/A B = N/A 4 = N/A 9 = Missing/blank 0 = Owned or being bought 1 = Owned 2 = Owned free and clear 1 = Owned with mortgage or loan 2 = Rented 4 = Occupied without payment of rent 3 = No cash rent 3 = Rented 2 = Rented for cash rent

Harmonized

Code Label

Translation Table

Input

2012 ACS 1960 1% 1900 5%

00 N/A B = N/A 4 = N/A 9 = Missing/blank 10 Owned or being bought 0 = Owned or being bought 1 = Owned 12 Owned free and clear 2 = Owned free and clear 13 Owned with mortgage

• r loan

1 = Owned with mortgage or loan 2 = Rented 4 = Occupied without payment of rent 3 = No cash rent 3 = Rented 2 = Rented for cash rent

Harmonized

Code Label

Translation Table

Input

2012 ACS 1960 1% 1900 5%

00 N/A B = N/A 4 = N/A 9 = Missing/blank 10 Owned or being bought 0 = Owned or being bought 1 = Owned 12 Owned free and clear 2 = Owned free and clear 13 Owned with mortgage

• r loan

1 = Owned with mortgage or loan 2 = Rented 4 = Occupied without payment of rent 3 = No cash rent 3 = Rented 2 = Rented for cash rent

Harmonized

Code Label

Translation Table

Input

2012 ACS 1960 1% 1900 5%

00 N/A B = N/A 4 = N/A 9 = Missing/blank 10 Owned or being bought 0 = Owned or being bought 1 = Owned 12 Owned free and clear 2 = Owned free and clear 13 Owned with mortgage

• r loan

1 = Owned with mortgage or loan 20 Rented 2 = Rented 21 No cash rent 4 = Occupied without payment of rent 3 = No cash rent 22 With cash rent 3 = Rented 2 = Rented for cash rent

Harmonized

Code Label

Translation Table

Input

2012 ACS 1960 1% 1900 5%

00 N/A B = N/A 4 = N/A 9 = Missing/blank 10 Owned or being bought 0 = Owned or being bought 1 = Owned 12 Owned free and clear 2 = Owned free and clear 13 Owned with mortgage

• r loan

1 = Owned with mortgage or loan 20 Rented 2 = Rented 21 No cash rent 4 = Occupied without payment of rent 3 = No cash rent 22 With cash rent 3 = Rented 2 = Rented for cash rent

Harmonized

Code Label

Translation Table

Input

2012 ACS 1960 1% 1900 5%

00 N/A 10 Owned or being bought 12 Owned free and clear 13 Owned with mortgage

• r loan

20 Rented 21 No cash rent 22 With cash rent

Harmonized

Code Label

Translation Table

N/A 1 Owned or being bought 1 Owned free and clear 1 Owned with mortgage

• r loan

2 Rented 2 No cash rent 2 With cash rent

Harmonized

Code Label

Translation Table

N/A 1 Owned or being bought 2 Rented

Harmonized

Code Label

Translation Table

Additional Harmonization and Data Enhancements

• Geographic Areas
• Consistent industrial and occupation coding

schemes

• Other complex variables
• Constructed family interrelationship

variables

Integrating Documentation

• Sample Descriptions
• Variable Descriptions

– Availability by Sample – Universes – Comparability – Allocation and Imputation Flags – Questions and Instructions to Respondents – Instructions to Enumerators

IPUMS USA

• U.S. decennial censuses (1850-2010)

– Complete-count data: 1850 - 1940

• American Community Survey (2000-2016)
• IPUMS format data in the FSRDC

– Available now: census data, 1960 – 2000 – Underway: ACS, 2000 - forward

Why IPUMS in the FSRDC

• More data

– Complete long-form decennial census data – More ACS cases

Why IPUMS in the FSRDC

• More data
• Better geographic detail

Geography

• Geographic in recent public use samples:

– State – Some Metropolitan Areas – Public Use Microdata Areas (PUMAs)

• Geographic in FSRDC data:

– Census block and tract – Consistent census tracts (IPUMS variable)

Why IPUMS in the FSRDC

• More data
• Better geographic detail
• Additional detail on key variables
• IPUMS harmonization and constructed

variables

Federal Statistical Research Data Centers

30 locations and growing

Metadata and the FSRDC

• Metadata drives IPUMS
• Public metadata made FSRDC work easier

Public documentation

Variable-level metadata

Metadata

• Metadata drives IPUMS
• Public metadata made FSRDC work easier
• Public metadata for IPUMS in the FSRDC will

become a tool for other researchers

Census Portal

• Facilitates project planning and proposal

preparation

• Encourages other harmonization projects
• Saves researcher time preparing outside the

FSRDC

Questions

• Metadata production is hard work and can be

labor intensive. Who is going to do it?

• Data are not always static. How do we keep

metadata up to date?

Questions? fitch@umn.edu