RE: [CTPP] With ACS make sure to Read the Small Print

Nate et al: they have told us in the past that the number of question and the probing that goes on is way more than the ACS/decennial could absorb. I did a paper on this as I recall last time -- maybe in CIAIII. My main thought is at least they owe us a side by side discussion of the numbers from the three sources and a discussion of the likely differences within groups. My note to others on this is just below: This (the difference in results) suggests that the disparity is because it was weak last time. It further indicates that the growth levels would occur in those areas of marginal workers and therefore perturb the modal shares trend between years, etc. There are other oddities that this would explain -- or confound with the actual trend. It would be very helpful if they provided some tabular reference points -- such as mode usage by income class or occupation and auto ownership by race, or income, etc. AEP Alan E. Pisarski 6501 Waterway Drive Falls Church Va. 22044 703 941-4257 alanpisarski(a)alanpisarski.com -----Original Message----- From: ctpp-news-bounces(a)chrispy.net [mailto:ctpp-news-bounces@chrispy.net] On Behalf Of Nathan Erlbaum Sent: Thursday, October 01, 2009 8:24 AM To: ctpp-news(a)chrispy.net Subject: Re: [CTPP] With ACS make sure to Read the Small Print a logical person would then ask why aren't the questions that BLS or CPS uses part of the acs? would this increase the level of confidence and reliability of these data? Earlier this week I was part of an email chain where it was discovered that with the new 2008 ACS data which was just release there was a national change in workers from 2007 of about 4.8 million for a growth of about 3.4 percent. Of course this seemed odd given the year that we have been through. I further learned that BLS was showing a loss of workers during the same time which prompted me to pass the issue over to Melissa Chiu our CTPP coordinator at the Census Bureau. Here is what Melissa had to say. Hi Ed, I spoke with the subject matter branch chief and it turns out that the Bureau recommends using caution when making labor force comparisons between data from 2008 or later and data from prior years (make sure to look at the footnote at the bottom of the labor force tables). This is because there was a question change starting in 2008 that results in more people responding that they worked last week. The question in use before 2008 missed many marginal workers. Essentially the data are better now and this also puts the ACS data more in line with the Current Population Survey (CPS)-based BLS data. In addition, as far as labor force estimates go, the BLS data are considered the official estimates, and many more questions are asked on the CPS regarding employment than on the ACS. So there may still be differences between CPS and ACS due to this and other issues. There is more information on this question change in the research note: http://www.census.gov/hhes/www/laborfor/researchnote092209.html The note references the following 2006 content test report that resulted in the question change: http://www.census.gov/acs/www/AdvMeth/content_test/P6a_Employment_Status.pdf Finally, for other information on labor force data, please start here: http://www.census.gov/hhes/www/laborfor/laborforce.html I hope this helps. If you have further questions, please let me know and I will refer you to the appropriate person. Thanks, Melissa -- Ed Christopher 708-283-3534 (V) 708-574-8131 (cell) FHWA RC-TST-PLN 4749 Lincoln Mall Drive, Suite 600 Matteson, IL 60443 _______________________________________________ ctpp-news mailing list ctpp-news(a)chrispy.net http://www.chrispy.net/mailman/listinfo/ctpp-news _______________________________________________ ctpp-news mailing list ctpp-news(a)chrispy.net http://www.chrispy.net/mailman/listinfo/ctpp-news