Search title image

Frequently Asked Questions: Census 2000 Transportation Planning Package (CTPP)

Last Updated: May 10, 2004

1. What journey-to-work data is available from the 1960 Census?
2. What journey-to-work data is available from the 1970 Census?
3. What journey-to-work data is available from the 1980 Census?
4. What journey-to-work data is available from the 1990 Census?
5. With the CTPP 2000, Part 3, where is the place-to-place data for Belvedere and Byron?
6. What does the term "remainder of county" include?
7. Who do I contact for more information on how to use the CAT software (CTPP Access Tool)?
8. Why don't the numbers add up?
9. What are some of the best ways to map commuter flow data?
10. In the CTPP 2000 superdistrict-to-superdistrict table, MTC is showing 41,229.8 commuters living-and-working in superdistrict #1 (northeast San Francisco). How can you have "0.8" commuters, and why isn't this an integer value?
11. Why is there no data on zone-to-zone (and superdistrict-to-superdistrict) bicycle and walk commuters from Census 2000?

What journey-to-work data is available from the 1960 Census?
Not very much. MTC has estimates of 1960 county-to-county, intra-regional total commuters in the Bay Area. We don't have information on county-to-county by means of transportation, or data at finer levels of geographies such as places or superdistricts. We don't have information on inter-regional commuting. The first data we have on inter-regional and place-level commuting is from the 1980 Census. Means of transportation to work included in the 1960 Census include: persons in autos or carpools, bus or streetcar, subway or elevated, railroad, taxicab, walk only, worked at home, and other means. The data on means of transportation is only available by county-of-residence and place-of-residence, manually copied from hard copy 1960 Census books.

What journey-to-work data is available from the 1970 Census?
MTC has the 1970 Census commuter matrices at the old 440 zone and old 550 zone levels. We've also aggregated these zone-to-zone commuter matrices to the MTC 34 superdistrict and nine county level. The superdistrict-to-superdistrict and county-to-county data are published on the MTC CTPP web page. The 440-to-440 and 550-to-550 zone data files can be provided on request, but we do not have GIS layers to support these old zone systems. Place-to-place commuter data is only available from the 1980-2000 Censuses. Means of transportation to work included in the 1970 Census include: vehicle driver, vehicle passenger, bus or streetcar, subway or elevated, railroad, taxicab, walk only, worked at home and other means.

What journey-to-work data is available from the 1980 Census?
We have the full set of 1980 Census Urban Transportation Planning Package (UTPP) data files, in ASCII flat file format. This is the first dataset to provide inter-regional county-to-county commuters by means of transportation. The place-to-place data in the 1980 UTPP is only for intra-regional places, not inter-regional place-to-place. The detailed zone-to-zone 1980 data is available on request. Means of transportation to work included in the 1980 Census include: car, truck, van, bus or streetcar, subway or elevated, railroad, taxicab, motorcycle, bicycle, walk only, worked at home and other means. Bicycle-to-work and motorcycle-to-work data was first collected in the 1980 Census. In addition, the 1980 Census is the first census to include data on: drive alone, 2-person carpool, 3-person carpool, 4-person carpool, 5-person carpool, 7-person carpool, and 7-or-more person carpool.

What journey-to-work data is available from the 1990 Census?
We have the full set of 1990 Census Transportation Planning Package (CTPP) data files, in ASCII flat file format. These 1990 CTPP files are available from the Bureau of Transportation Statistics (BTS) in both CD-ROM, and downloadable from the BTS TRANSTATS site. This is the first census where we have inter-regional place-to-place commute data. The means of transportation in the 1990 Census are similar to the 1980 Census, with the following changes: bus or trolleybus is separated from streetcar or trolleycar; ferryboat is a new mode; and carpooling is extended to: 7-to-9 person carpools and 10-or-more person carpools.

With the CTPP 2000, Part 3, where is the place-to-place data for Belvedere and Byron?
CTPP data was not produced for places of less than 2,500 total population. This is the same rule/threshold that was applied with the 1990 CTPP. This rule applies to both incorporated cities such as Belvedere in Marin County, as well as unincorporated census designated places (CDPs) such as Byron in east Contra Costa County. There are 40 places (out of 183) in the Bay Area with population less than 2,500, ranging from Elmira in Solano County (population 205) to the Mountain View CDP in Contra Costa County (population 2,468) (not to be confused with the City of Mountain View in Santa Clara County!) Belvedere, Ross and Colma are the only Bay Area incorporated communities that do not exceed the 2,500+ total population threshold. Hint: block group-to-block group commute pattern data may be suitable for persons interested in these fairly small towns and neighborhoods.

What does the term "remainder of county" include?
The term "remainder of county" is the total county, less all places of 2,500+ total population. Another synonymous term is "balance of county." It is not the same as unincorporated portions of the county, since there are unincorporated census designated places (e.g., Castro Valley) where we have place-to-place data.

Who do I contact for more information on how to use the CAT software (CTPP Access Tool)?
We recommend you contact either Mr. Shimon Israel of MTC [510-817-5839], or Mr. Nanda Srinivasan of the FHWA in DC [202-366-5021].

Why don't the numbers add up?
All of the data values in CTPP 2000 have been independently rounded. Total expanded census values that range from 1 to 7 are rounded to "4" and values above 7 are rounded to the nearest even multiple of five, e.g., 5, 10, 15, 20, etc. For more discussion of the rounding issue, please read the April 2004 edition of the CTPP 2000 Status Report (PDF format).

What are some of the best ways to map commuter flow data?
For analysts using ESRI ArcView and ArcGIS products we would recommend acquiring the free ArcView extension (*.avx) "Desire Line Maker" by Kevin O'Malley. Desire Line Maker is available from the ESRI ArcScripts library. (MTC will be developing some of these county-to-county, superdistrict-to-superdistrict, and zone-to-zone desire line "shp" files in the coming months.)

MTC will also be using our travel modeling network package (Cube/Voyager) to prepare "select link assignments" using our travel modeling networks and the CTPP-based zone-to-zone commuters by means of transportation. The resulting "select link trees" can then be mapped, and will show the origins and destinations of commuters using certain facilities, say, the Bay Area bridges.

In the CTPP 2000 superdistrict-to-superdistrict table, MTC is showing 41,229.8 commuters living-and-working in superdistrict #1 (northeast San Francisco). How can you have "0.8" commuters, and why isn't this an integer value?
The reason for the fractional number of commuters is due to the strange fact that CTPP 2000 does not have a table showing tract-to-tract commuters by means of transportation for the "all workers" universe, that is not subject to the 3 sample worker threshold. We can, however, derive workers by means of transportation to work by dividing the "aggregate travel time to work by mode" (Table 3-14), by the "average travel time to work by mode" (Table 3-8). We decided to round this derived value to the nearest tenth of a commuter to provide a clue to the data user that these particular variables were not rounded by the Census Bureau, but were rounded by the data analyst. These calculations have been checked (aggregate time / average time = commuters by means of transportation) and they appear quite reasonable.

We would recommend that data users round these values to the nearest integer (or hundreds, or thousands) when reporting the data to the public, the media, and policymakers. With the above example of intra-superdistrict #1, it may be useful to round the data off even more so, say, 41,200 commuters, or "over 41 thousand commuters" depending on your audience.

Why is there no data on zone-to-zone (and superdistrict-to-superdistrict) bicycle and walk commuters from Census 2000?
Data on detailed means of transportation for all workers (e.g., bicycle, walk, ferryboat, railroad, etc.) is from CTPP 2000 Table 3-6. This particular table is subject to the 3 sample minimum threshold, which has a serious data suppression impact at the zone-to-zone and tract-to-tract geographic levels. For example, the county-to-county commute file for intra-Bay Area shows 35,722 bicycle commuters, but the tract-to-tract commute file, intra-Bay Area, shows only 27,298 bicycle commuters, a data loss of 23.6 percent. Similarly, the county-to-county intra-regional walk commuters is 104,454, whereas the tract-to-tract intra-regional walk commuters is 87,318, a data loss of 16.4 percent. There is no obvious method to "impute" or make up the missing data.

Superdistrict-to-superdistrict commuter matrices are based on the tract-to-tract commuter flows, so any data losses at the tract-to-tract level are the same when aggregated to the MTC 34 superdistricts. Note that the data loss at the place-to-place level (intra-Bay Area) is fairly insignificant: 0.9 percent data loss for bicycle commuters and a 0.5 percent data loss in terms of walk-only commuters.

RETURN to MTC CTPP Page