1 of 46

Geography in Full-Count Data

David Van Riper

vanriper@umn.edu

IPUMS – University of Minnesota

2 of 46

WHAT’S SPECIAL ABOUT �FULL-COUNT GEOGRAPHY?

Exceptionally detailed

… but georeferencing may be difficult

3 of 46

Full count census geography

1950 enumeration districts in St Paul, MN

4 of 46

Even more detail is available!

5 of 46

Historic Rondo Neighborhood, St Paul, MN

6 of 46

Coded geography in full-count data

1850

1860

1870

1880

1900

1910

1920

1930

1940

1950

State (or territory)

X

X

X

X

X

X

X

X

X

X

County

X

X

X

X

X

X

X

X

X

X

City

X

X

X

X

X

X

X

X

X

X

Region & division

X

X

X

X

X

X

X

X

X

X

7 of 46

Coded cities

  • IPUMS numbering system, consistent across time
  • Harmonized by name, not by “footprint”
  • Inclusion criteria:
    • 1870–1950: >10,000 residents (approx.)
    • 1850–1860: Largest 100 cities

8 of 46

Coded geography in full-count data

1850

1860

1870

1880

1900

1910

1920

1930

1940

1950

State (or territory)

X

X

X

X

X

X

X

X

X

X

County

X

X

X

X

X

X

X

X

X

X

City

X

X

X

X

X

X

X

X

X

X

Metropolitan area

X

X

X

X

X

X

X

X

X

*

Urban-rural status

X

X

X

X

X

X

X

X

X

X

*Will be added

Region & division

X

X

X

X

X

X

X

X

X

X

Metropolitan status

X

X

X

X

X

X

X

X

X

*

Size of place

X

X

X

X

X

X

X

X

X

*

18 categories from < 1,000 to > 2 million

9 of 46

Coded geography in some full-count data

1850

1860

1870

1880

1900

1910

1920

1930

1940

1950

Enumeration district

.

.

.

X

X

X

X

X

X

*

*Will be added

Ward

X

.

.

.

.

.

X

.

X

.

Urbanized area (1980)

X

X

X

X

X

X

X

X

.

.

Metropol’n district (1940)

X

X

X

X

X

X

X

X

.

.

Geo coordinates (long/lat)

.

.

.

X

.

.

.

.

.

.

39 cities

10 of 46

Most geographic variables

Enumeration districts,�geographic coordinates

(XGPS, YGPS)

11 of 46

Unedited strings in restricted-access data

  • Address is sometimes missing, especially for apartment units
  • Address is also typically available only in cities or incorporated places

1850

1860

1870

1880

1900

1910

1920

1930

1940

1950

Minor civil division

X

X

X

X

X

X

X

X

X

X

Incorporated places

X

X

X

X

X

X

X

X

X

X

Street address

X

X

X

X

X

X

X

X

X

X

12 of 46

1950 Addresses

Street Name

Apt #

House #

13 of 46

14 of 46

Street Name

Apt #

House #

15 of 46

1940 Addresses

Street_name

House #

16 of 46

GEOREFERENCING?

  • How to map full-count data?
  • How to measure spatial relationships?
  • How to link data spatially?
  • Challenge: �Limitations of historical geospatial data

17 of 46

18 of 46

NHGIS

National Historical�Geographic Information System

  • Geographic U.S. census & survey data
  • Summary tables and GIS files
  • 1790 to the present

19 of 46

20 of 46

21 of 46

PLACE POINTS�1900 – 2015

22 of 46

OTHER SOURCES…

  • Enumeration district GIS data

23 of 46

24 of 46

25 of 46

26 of 46

Color is important, but many scans are monochrome ☹ ¯\_(ツ)_/¯

27 of 46

Color is important, but many scans are monochrome

28 of 46

Color is important, but many scans are monochrome ☹ ¯\_(ツ)_/¯

29 of 46

30 of 46

31 of 46

32 of 46

33 of 46

Urban Transition Historical GIS Project

  • 1930 & 1940 EDs & ward boundaries + street grids
    • 73 localities
    • Several in same metro (e.g., Minneapolis and St Paul)
  • 1900-1930 ED & ward boundaries
    • “Northen Cities” project (Allison Shertzer, Randall Walsh & John Logan)
    • 10 cities (NYC, Chi, Phi, Bos, Bal, Cin, Cle, StL, Pit, Det)
  • 1880 ED boundaries
    • 39 cities (26 largest + selected others)

34 of 46

The Mark Fossett Lab

  • Mapping the Origins of Urban Segregation
  • 1940 EDs + ED summary data
    • 18 metropolitan areas (single and multi-county)
  • Best practices user guide

35 of 46

Availability

36 of 46

Urban Transition and Fossett Lab

https://s4.ad.brown.edu/Projects/UTP2/citymaps.htm

https://thefossettlab-fresnostate.hub.arcgis.com

�

37 of 46

Considerations when choosing a source

  • Discrepancies in the number of EDs

38 of 46

Considerations when choosing a source

  • Discrepancies in the number of EDs

Shertzer et al.

Logan

39 of 46

Considerations when choosing a source

  • Demographic data without ED polygons

40 of 46

Considerations when choosing a source

  • Demographic data without ED polygons

ed_padded

total_pop

00770

269

ed_padded

totalpop

00760

482

00780

1025

00790

851

IPUMS full-count

Logan GIS

41 of 46

Considerations when choosing a source

  • Demographic data without ED polygons

42 of 46

Considerations when choosing a source

  • ED polygons without demographic data

43 of 46

Considerations when choosing a source

  • ED polygons without demographic data

Logan

44 of 46

Considerations when choosing a source

  • ED polygons without demographic data

ed_padded

total_pop

mean_age

male_share

white_share

02721

738

29.90921

0.4742547

1.00000000

02722

772

26.64637

0.4753886

1.00000000

02724

598

27.79599

0.4966555

1.00000000

02725

1049

29.36892

0.5147760

1.00000000

45 of 46

Urban Transition Historical GIS Project

  • Geographic reference files (GRF), 1910-1940
    • Microdata with geographic info
      • Ward & ED
      • Block in 1930
      • Cleaned house number & street name
      • HISTID for linking to IPUMS full-count data
    • 191 cities (> 30k population in any of these years)

46 of 46

Conclusion

  • Geographic detail in full-count census data is exceptionally detailed
  • Sub-county GIS data available for selected cities (1900-1940) and counties (1940), but relatively little for suburbs or rural areas
  • Mapping methodologies may lead to mismatches between GIS data and full-count census data