To: CTPP-News
per Phil Salopek's request, I cobbled together some July 2002 threads from the State Data
Center Listserv. This is useful in showing the magnitude of differences when comparing
short form (SF1) and long form (SF3, CTPP) data.
Especially useful is the analysis produced by John Blodgett of the University of
Missouri. See link below.
My comments:
1. Compare CTPP to SF3, not SF1.
2. Given the rounding scheme implemented in the CTPP it will be difficult to match the geographic summary from a finer-grained geography, say, travel analysis zones or census tracts, to the data reported at a higher-level geography, say, county. That is, the "whole" is not necessarily the "sum of the parts."
3. Discussions on how folks are "unrounding" the CTPP data to match SF1 or SF3 or PUMS would be useful to share on this CTPP listserv.
Chuck Purvis, MTC
*********************************************************************************************************
I wonder if this a result of the Bureau's use of Counties as the primary
sampling unit to determine the weights for population and housing counts on
the sample data. In 1990, they used areas (counties, MCDs, places, and
census tracts) over a relatively small population threshold (I think 2,500).
Leonard M. Gaines, Ph.D.
Research Specialist
Empire State Development
e-mail: lgaines(a)empire.state.ny.us
Empire State Development & NY State Data Center Web Sites:
http://www.empire.state.ny.us
Voice: (518) 292-5300
Fax: (518) 292-5806
> -----Original Message-----
> From: Wu, Sen-Yuan [mailto:Sywu(a)DOL.STATE.NJ.US]
> Sent: Tuesday, July 30, 2002 7:02 AM
> To: FSCPE(a)LISTSERV.LOUISVILLE.EDU
> Subject: Re: Demographic Profile Errors
>
>
> In New Jersey, most discrepancies between SF1 and SF3 were found in CDPs.
> The differences between the 100% and sample population counts were as high
> as 38.1% in Diamond Beach CDP (218 vs. 135) and 31.4% in Vista Center CDP
> (541 vs. 711).
>
> Other than the CDPs, only 7 (out of 566) municipalities had 5% or more
> differences in population or housing unit counts. Pine Valley Borough had
> the largest discrepancies (20% in population, 66.7% in housing units).
> All
> except one are tiny municipalities with less than 600 residents.
>
> Sen-Yuan Wu
> New Jersey Department of Labor
> Division of Labor Market & Demographic Research
> Tel. 609-292-0077, Fax 609-984-6833
> <http://www.state.nj.us/labor/lra>
John -
Thank you for the work on this. I am going to forward to the FSCPE listserve as well. It is interesting in that even a smaller community down the road from which is even smaller looks numerically better. If anyone wants to see the article that came out on Searchlight and a lesson on dealing with the press check http://www.lvrj.com/lvrj_home/2002/Jul-29-Mon-2002/news/19282077.html
"Blodgett, John G." wrote:
> There appear to be some problems with this. We ran a test with our DP datasets, comparing the 100% and sample counts for all places. The summary report can be viewed at http://mcdc2.missouri.edu/pub/data/sf3prof/check_totpops.pdf . The biggest problem, in terms of pct difference in the counts, is definitely in the very small places. There are 593 places in the country where the difference was 25% or more and 566 of these were for places with 500 people or less.
> The report also includes a listing of these 593 places, sorted by state and descending Pct Difference. The winner of the worst sample estimate award is Blacksville CDP, Ga. They had a 100% count of 4 people, but the sample estimate was 52.
>
> John Blodgett
> OSEDA - Office of Social & Economic Data Analysis
> U. of Missouri Outreach and Extension
> 626 Clark Hall - UMC
> Columbia, MO 65211
> (573) 884-2727
> blodgettj(a)umsystem.edu
> URL: http://oseda.missouri.edu/jgb/
>
> -----Original Message-----
> From: jeff hardcastle [mailto:jhardcas(a)UNR.EDU]
> Sent: Friday, July 26, 2002 11:30 AM
> To: FSCPE(a)LISTSERV.LOUISVILLE.EDU
> Subject: Demographic Profile Errors
>
> I am not sure if this has happened to states for the Demographic
> Profiles that include SF1 and SF3 data but in Nevada's case there are
> serious problems that suggest that the whole set of profiles needs to be
> reviewed for errors. These errors appear to be more than standard
> sampling and response errors. During a quick review counties look
> better than places however it appears that there may be geocoding
> errores in the sample data. There also appears to be differences in
> what the sample data is weighted against.
>
> The place that brought this to my attention was Searchlight NV where tge
> DP-1 pop is 576 and the housing unit count is 444. On DP-2 through 4
> the pop is 768 and the unit count is 595. A 33% and a 34% difference
> respectively. The way I tumbled to this was that a reporter had seen
> that Searchlight had no native Nevadans living there. He went to
> Searchlight and interviewed people and found that most of them if not
> all were natives. (Searchlight is an old mining down south of Las Vegas
> and in the middle of very open country.)
>
> Is this kind of error being found elsewhere?
>
> --
> peace,
>
> Jeff Hardcastle
> (775) 784-6353 Phone
> (775) 784-4337 Fax
> jhardcas(a)unr.edu e-mail
> "shifts happen" http://publicconversations.org/
Hi all,
Our MPO is currently checking the newly-released 2000 CTPP data. We have
found some discrepancies between CTPP and other sources. According to
Census, "CTPP Part 1 data will not agree with SF 1 or Redistricting (Full count
data). The CTPP data are based on the census long form, and are subject to
weighting, just like other sample surveys. The SF1 and Redistricting data
are full counts, and are more accurate. The best source to do spot-checks
on the CTPP data is SF3".
Even keeping this in mind, we still found the comparision result alarming.
For the population data, we aggregated census block population from the
redistricting file into each TAZ, hoping the aggregation result is not
significantly different from the CTPP data. We have 602 TAZs in our area,
of which 511 have somebody living in them according to the redistricting
file. However, according to CTPP, only 474 TAZ have population. The
population in the 37 zones that CTPP ommited varies from 1 to 40, which we
were told are not significant.
We do have some zones that have significant differences in population (at
least we consider "significant"). We have a zone that has a population of
1277 in the redistricting file. The CTPP says that this zone has a
population of 455. The difference between these two data sources are 822.
We are not sure how to deal with this problem. Do any of the other MPOs
have the same kind of problems? We know we need to use the CTPP
information with care, but we assume that no matter what sampling method
and what weighting factors the Census is using, a difference of almost
200% is still too much.
We would like to pose this question to all of the CTPP users. If any of
you can give us a good explanation or how your agency is addressing the
data issue, please drop me a note so that we all learn from it. Thanks.
Sincerely
Zhongze (Wykoda) Wang
Associate Transportation Planner
Chatham County - Savannah Metropolitan Planning Commission
P. O. Box 8246, 110 East State Street
Savannah, GA 31412-8246
Phone: (912) - 651-1452 Fax: (912) - 651- 1480
Email: wangz(a)thempc.org
A new report, "Journey to Work Trends in the United States and its Major Metropolitan Areas, 1960-2000," is now available (FHWA report # FHWA-EP-03-058) . It was prepared for FHWA, authored by Nancy McGuckin and Nanda Srinivasan. If you regularly receive a mailed copy of the CTPP Status Report, you should receive a printed copy of the report in the mail very shortly. It is currently available in electronic form at www.trbcensus.com , with thanks to Chuck Purvis at the Metropolitan Transportation Commission in Oakland, CA. (It will be posted to the FHWA website, after all the spreadsheets are converted into a FHWA "508 compliant" format.)
If you would like a single copy of the printed report, please email your request to ctpp(a)fhwa.dot.gov, and be sure to include all the appropriate name and address information.
If you would like MULTIPLE copies of the printed report, please FAX your request to the DOT warehouse at 301-386-5394. Please be sure to include the FHWA report number (FHWA-EP-03-058) in your request!
This report is based primarily on Census Bureau Census 2000 Summary File 3 data, and includes data for the U.S. total, and for metropolitan areas (1999 definition) with population of 1 million or more.
Key findings:
-- Nearly 60 percent (57.4 percent) of the U.S. population live in one of the large metro areas (1 million pop or over)
-- 3 metro areas added more than 1 million population between 1990 and 2000: Dallas-Ft Worth; Atlanta and Phoenix.
-- Commute times in the large metro areas averaged about 28 minutes, with over 40 percent of workers spending 30 minute or longer on their (one-way) commute to work. For areas outside of the large metro areas, the average commute time is 22 minutes.
-- In the New York/New Jersey metro area about 25 percent of commuters use transit for work. The national average is about 5 percent of commuters using transit, and for all large metro areas, 7.4 percent using transit.
Errata-- OOPS! I already found the first mistake! On page 4-3, it should read that nearly 40 percent (2.3 million of 6.1 million) of the country's transit commuters live in the New York/Northern New Jersey metropolitan area.
Note: Each FHWA Division and Metropolitan office will receive a copy (or 2, I can't remember what I asked for!) in the mail. Each FHWA Resource Center office will receive a minimum of 10 copies.
Elaine Murakami
206-220-4460
Attention: State DOT and MPO staff evaluating CTPP 2000, Part 1:
I received a number of e-mails/phone calls since my last e-mail asking for staff to evaluate different levels of geography.
Some of the common issues are:
1. CTPP Part 1 data will not agree with SF 1 or Redistricting (Full count data). The CTPP data are based on the census long form, and are subject to weighting, just like other sample surveys. The SF1 and Redistricting data are full counts, and are more accurate. The best source to do spot-checks on the CTPP data is SF3.
2. The count of households WILL NOT EQUAL Occupied Housing Unit counts (in general).
This is because the all the household data in the census are derived by tabulating the "person weight" of the householder. Weights for occupied housing units are assigned based on housing unit counts.
3. Please be aware that all CTPP tables (except the mean, median, and standard deviation tables) are subject to the following rounding rules:
0 is kept as 0.
1-7 are rounded to 4
Anything above 7 is rounded to the nearest multiple of 5 (eg: 9 is rounded to 10, 17 is rounded to 15 etc)
Columns for Totals in each table may not match the sum of the categories because Totals are rounded independently of the cells. For example:
0 vehicle households = 6 is rounded to 4
1 vehicle households =14 is rounded to 15
2 vehicle households=8 is rounded to 10
3 vehicle households=8 is rounded to 10
4 vehicle households=3 is rounded to 4.
Total households=39 is rounded to 40 (and NOT 4+15+10+10+4 = 43 rounded to 45).
The difference between the totals and the sum of the parts is substantial for tables with more than 30 cells.
4. Missing geography: Sometimes, a few TAZs or BGs may not show up in your geography/map viewer or tables. This is because geography are shown only if at least ONE table in CTPP has some values. This means that those TAZs or BGs that don't show up have NO population or households in them (based on long form data).
5. Wrong Column headers: We noticed that the CTPP Access Tool wrongly labels column headers for one dimensional tables - this error will be fixed for the final version.
6. MSA/CMSA names are wrong: We think that the names for MSA/CMSA or PMSAs are incorrect, and should be fixed in the final version.
If you have issues DIFFERENT from the above 6, please call/e-mail me.
Thank you!
Nanda Srinivasan
202-366-5021
The Census Bureau just released its latest American Community
Survey-Alert (an ACS newsletter) Number 17. In this Alert there are
articles on
* ACS 2002 Data Are Now Available
* Redesigned ACS Web Site Launched
* ACS Content Fact Sheets
The Alerts and "lots of" information on the ACS can be found on the ASC
website at http://www.census.gov/acs/www/
--
Ed Christopher
Planning Specialist
Resource Center
Federal Highway Administration
19900 Governors Drive
Olympia Fields, Illinois 60461
708-283-3534 (V)
708-574-8131 (cell)
Hello from Utah.
You may be interested in a recent report:
"Commuting Patterns in Utah: County Trends for 1980, 1990, and 2000"
It is based on the CTPP 2000 county level data as well as the
county-to-county journey-to-work file.
It is available here:
http://www.business.utah.edu/bebr/uebr/uebrMayJun2003.pdf
Do you know of similar studies focusing on other states?
Best regards,
Pam
Pam Perlich
Senior Research Economist
Bureau of Economic and Business Research
University of Utah
1645 E. Campus Drive, Room 401
Salt Lake City, Utah 84112-9302
Voice: (801) 581-3358
Fax: (801) 581-3354
Email: Pam.Perlich(a)business.utah.edu
Greetings folks out there in CTPP-land -- I was wondering whether anyone
had yet tabulated or otherwise produced tables 2-003, 2-004, and 2-005 for
New York State's counties. (That's sex by occupation, industry, and class
of worker by place of work.) I don't have access to the advance release of
the CTPP myself but am trying to compile a report on the Upstate NY economy
based on place of work. If it's not kosher to release these tables to me,
don't -- but if you can I would appreciate it. Thanks. Please post directly
to me, not back to the list.
Rolf Pendall
-------------------------------------------------
Rolf Pendall, AICP, Ph.D.
Associate Professor, City & Regional Planning
Cornell University
212 W. Sibley Hall, Ithaca, NY 14853
607-255-5561 / 530-678-8103 (fax) / rjp17(a)cornell.eduhttp://www.people.cornell.edu/pages/rjp17/
