Speaking of lumpy flows: my first survey in NY (a 1% hh face to face survey) generated a commuter coming to Manhattan every day down the Hudson in a canoe! So every morning from my window at tri-state I watched for the flotilla of 100 canoes that the survey expected - it never showed up. From: ctpp-news-bounces(a)chrispy.net [mailto:ctpp-news-bounces(a)chrispy.net] On Behalf Of Bruce Spear Sent: Wednesday, June 06, 2012 9:28 PM To: ctpp-news(a)chrispy.net Subject: Re: [CTPP] ACS/CTPP vs. LEHD (Spear, Bruce) Hi Nick, In addition to what Todd mentioned, we found a few other factors that are likely to create differences in the ACS vs. LED totals. 1. The ACS is based on a relatively small sample (~ 2.5% per year) of all households nationwide, but this sample is expanded based on county population to produce the totals you see. The problem is that low frequency events such as work trips between distance zones are unlikely to be captured in a small sample. When the sample is expanded, the low frequency trips are still zero, while trips between zones that were sampled must be given a higher weight to match the county population totals. As a result, the ACS flows look a lot lumpier than the LED flows, which do capture the low frequency home and work locations. 2. The ACS only asks about a worker's primary job, while the LED captures all workers and employers for covered trips. Therefore the LED picks up flows of workers with secondary jobs. 3. Many of the very long distance commutes picked up in the LED are probably people who are on temporary assignment, living away from their reported place of residence (as far as their federal tax return is concerned), or weekend commuters who live in an apartment during the week and travel home on weekends. There are far too many of these folks reported in the LED and even in the 2000 CTPP for it to simply be data error. I suggest you read the final report we did for NCHRP 8-36, 98, title Improving Employment Data for Transportation Planning. The report presents the findings from a study we did that examines the differences between CTPP and LED data, and contains some interesting and useful tables that compare differences at the state, county, and even Census Tract levels of geography. You can download a copy of the report from the TRB website at the following URL: <http://onlinepubs.trb.org/onlinepubs/nchrp/docs/NCHRP08-36(98)_FR.pdf> http://onlinepubs.trb.org/onlinepubs/nchrp/docs/NCHRP08-36(98)_FR.pdf Bruce D. Spear, PhD Cambridge Systematics Inc. 100 Cambridgepark Drive, Suite 400 Cambridge, MA 02140 tel 617-234-0417 fax 617-354-1542 bspear(a)camsys.com www.camsys.com From: ctpp-news-request(a)chrispy.net To: ctpp-news(a)ryoko.chrispy.net Date: 06/06/2012 05:31 PM Subject: ctpp-news Digest, Vol 100, Issue 4 Sent by: ctpp-news-bounces(a)chrispy.net _____ Send ctpp-news mailing list submissions to ctpp-news(a)ryoko.chrispy.net To subscribe or unsubscribe via the World Wide Web, visit <http://ryoko.chrispy.net/mailman/listinfo/ctpp-news> http://ryoko.chrispy.net/mailman/listinfo/ctpp-news or, via email, send a message with subject or body 'help' to ctpp-news-request(a)ryoko.chrispy.net You can reach the person managing the list at ctpp-news-owner(a)ryoko.chrispy.net When replying, please edit your Subject line so it is more specific than "Re: Contents of ctpp-news digest..." Today's Topics: 1. Re: ACS/CTPP vs LEHD (Graham, Todd) 2. Re: ACS/CTPP vs LEHD (Kim, Kyeongsu (DOT)) ----- Message from "Graham, Todd" <todd.graham(a)metc.state.mn.us> on Wed, 6 Jun 2012 16:11:25 -0500 ----- To: "'ctpp-news(a)chrispy.net'" <ctpp-news(a)chrispy.net>, "'lindeman(a)nashvillempo.org'" <lindeman(a)nashvillempo.org> Subject: Re: [CTPP] ACS/CTPP vs LEHD Reading the documentation of the two data resources: always a good place to start. At regional or state level of summary, totally different data frames is your main issue: a. Census ACS counts workers through a household survey, relying on individually-provided info on work. Many kinds of work are counted, including self-employment and occasional casual work. b. Meanwhile, the LED uses the data frame of BLS QCEW, counting wage & salary jobs that are "covered" by the Unemployment Insurance system. In Minnesota, over 20% of BEA-counted jobs (a broader employment concept) are outside of this frame, are *not* covered, and thus not counted in LED. At local level, there are still other issues: a. How well can Census ACS represent job locations? b. And does the LED program move its workplace locations around on the map, in order to obscure exact location? Hope that helps. Good luck. Todd Graham | Principal Forecaster Metropolitan Council | 390 North Robert Street | Saint Paul, MN 55101 tel: 1+651-602-1322 | fax: 1+651-602-1674 | e: <mailto:todd.graham(a)metc.state.mn.us> todd.graham(a)metc.state.mn.us in: <http://www.linkedin.com/in/toddgraham> www.linkedin.com/in/toddgraham Visit <http://www.metrocouncil.org/data> www.metrocouncil.org/data for the latest in regional information. From: ctpp-news-bounces(a)chrispy.net [ <mailto:ctpp-news-bounces(a)chrispy.net> mailto:ctpp-news-bounces(a)chrispy.net] On Behalf Of Lindeman, Nicholas (MPO) Sent: Wednesday, June 06, 2012 3:22 PM To: 'ctpp-news(a)ryoko.chrispy.net' Subject: [CTPP] ACS/CTPP vs LEHD I've sent the following query to both the ACS and CTPP but have not (yet) received a response. I was hoping someone on the mailing list might be able to provide assistance. I've been working with Longitudinal Employer-Household Dynamics (LEHD)in On the Map. Specifically worker inflow/outflow analysis at the county level. I've discovered that there is a significant difference between the LEHD data and ACS Place of Work data from the 2008-10 3-Year ACS product. I've found that for the Middle Tennessee (Nashville) region the % of workers residing outside their home county is nearly 17 percentage points higher in the LEHD data than in the ACS data. Please see the attached table. I've been tasked with determining the reason(s) for this difference. Any assistance you can provide would be greatly appreciated Thank you, Nick Lindeman Nicholas Lindeman Economic & Systems Data Analyst Nashville Area MPO 800 2nd Ave. S. P.O. Box 196300 Nashville, TN 37219-6300 (615) 862-7198 <mailto:lindeman(a)nashvillempo.org> lindeman(a)nashvillempo.org <http://www.nashvillempo.org/> http://www.nashvillempo.org _____ This email is intended to be read only by the intended recipient. This email may be legally privileged or protected from disclosure by law. If you are not the intended recipient, any dissemination of this email or any attachments is strictly prohibited, and you should refrain from reading this email or examining any attachments. If you received this email in error, please notify the sender immediately and delete this email and any attachments. ----- Message from "Kim, Kyeongsu (DOT)" <kkim(a)dot.state.ny.us> on Wed, 6 Jun 2012 17:00:13 -0400 ----- To: "ctpp-news(a)chrispy.net" <ctpp-news(a)chrispy.net> Subject: Re: [CTPP] ACS/CTPP vs LEHD Hi Nick, The difference comes from different data sample. As you know, ACS estimates are based on the survey samples. The upcoming 5 year CTPP will be based on same samples. On the other hand, the LEHD incorporates diverse datasets. There exists possibility of having significant difference between ACS and LEHD. I believe, however, Census Bureau tries to minimize the gap. FYI, please find the link that NYCDOT under Nathan Erlbaum (who just left us in April) examined. <https://www.dot.ny.gov/divisions/policy-and-strategy/darb/dai-unit/ttss/ctt p_acs> https://www.dot.ny.gov/divisions/policy-and-strategy/darb/dai-unit/ttss/cttp _acs Hope this help, Kyeongsu Kim Census and Travel Survey NY Metropolitan Transportation Council 199 Water Street 22nd fl New York, NY 10038 kkim(a)dot.state.ny.us 1-212-383-7233 From: ctpp-news-bounces(a)chrispy.net [ <mailto:ctpp-news-bounces(a)chrispy.net> mailto:ctpp-news-bounces(a)chrispy.net] On Behalf Of Lindeman, Nicholas (MPO) Sent: Wednesday, June 06, 2012 4:22 PM To: 'ctpp-news(a)ryoko.chrispy.net' Subject: [CTPP] ACS/CTPP vs LEHD I've sent the following query to both the ACS and CTPP but have not (yet) received a response. I was hoping someone on the mailing list might be able to provide assistance. I've been working with Longitudinal Employer-Household Dynamics (LEHD)in On the Map. Specifically worker inflow/outflow analysis at the county level. I've discovered that there is a significant difference between the LEHD data and ACS Place of Work data from the 2008-10 3-Year ACS product. I've found that for the Middle Tennessee (Nashville) region the % of workers residing outside their home county is nearly 17 percentage points higher in the LEHD data than in the ACS data. Please see the attached table. I've been tasked with determining the reason(s) for this difference. Any assistance you can provide would be greatly appreciated Thank you, Nick Lindeman Nicholas Lindeman Economic & Systems Data Analyst Nashville Area MPO 800 2nd Ave. S. P.O. Box 196300 Nashville, TN 37219-6300 (615) 862-7198 <mailto:lindeman(a)nashvillempo.org> lindeman(a)nashvillempo.org <http://www.nashvillempo.org/> http://www.nashvillempo.org _______________________________________________ ctpp-news mailing list ctpp-news(a)ryoko.chrispy.net <http://ryoko.chrispy.net/mailman/listinfo/ctpp-news> http://ryoko.chrispy.net/mailman/listinfo/ctpp-news