Call me dense, but I don’t see how this problem is any different than working with moving averages.  I mean, I do understand that this data is not an average but a period estimate.  Still, the issues created by dropping the first year of the period and adding the last year as the data series moves forward in time seems the same to me. 

 

What am I not understanding properly?

 

Frank

 

Frank Lenk

Director of Research Services

Mid-America Regional Council

600 Broadway, Suite 200

Kansas City, MO 64105

www.marc.org

816.474.4240

flenk@marc.org

816.701.8237

 

From: ctpp-news-bounces@chrispy.net [mailto:ctpp-news-bounces@chrispy.net] On Behalf Of Patty Becker
Sent: Tuesday, February 01, 2011 3:17 PM
To: ctpp-news@chrispy.net
Subject: Re: [CTPP] 2000 and 2006-08 work trip comparison question.

 

If you see the point about the problem of comparing overlapping 3 year ACS estimates, it gets even worse with consecutive 5 year numbers, in which 4 of the 5 years are the same data.

Personally, I think there's a strong argument for releasing 5 year data only twice a decade, by definition non-overlapping. We may get to that point in a couple of years.

Patty Becker



At 03:53 PM 2/1/2011, you wrote:

Frank-I totally agree and I would like to see the statisticians get involved and think about this. People are going to compare overlapping periods and we have to get smart about how to deal with this.

As for why some say we should not compare overlapping years the story goes like this.  If you have 2006 to 2008 data and 2007 to 2009 data you would have two overlapping years.  According to the logic you would in fact only be comparing the first year 2006 to the last year 2009 as the middle two years would cancel themselves out.  While this sounds logical on the surface there has to be a way to deal with overlapping 5-year periods or it doesn't make sense to have an ACS.

I am not sure if anyone would out right admit it but is anyone designing performance measures or other regional metrics calling for tracking annual multi-year period estimates at tracts or places?

Frank Lenk wrote:

I would also be interested in a more detailed explanation of why we should not compare overlapping period estimates.  From a practical standpoint, people are going to do it anyway, especially the press.  What do we need to know - other than the inherent issues of what is a statistically significant vs. insignificant change and the fact that ACS was designed to examine characteristics of the population rather than its level - to be able to correct naïve interpretations of easily calculated tract-level year-to-year differences in the estimates of things like poverty rate, unemployment rate, educational attainment rates, etc.? What is a correct interpretation of these differences?
Frank
Frank Lenk
Director of Research Services
Mid-America Regional Council
600 Broadway, Suite 200
Kansas City, MO 64105
www.marc.org < http://www.marc.org>
816.474.4240
flenk@marc.org< mailto:flenk@marc.org>
816.701.8237
From: ctpp-news-bounces@chrispy.net [ mailto:ctpp-news-bounces@chrispy.net] On Behalf Of Kendra Watkins
Sent: Monday, January 31, 2011 10:05 AM
To: ctpp-news@chrispy.net
Subject: RE: [CTPP] 2000 and 2006-08 work trip comparison question.
This is interesting. I recall hearing this guidance in the past. However on the Census website on the ACS page it specifically states; "Generally, you can compare American Community Survey (ACS) 3-year and 5-year estimates with Census 2000 data. There are differences in the universe, question wording, residence rules, reference periods, and the way in which the data are tabulated which can impact comparability."
The qualifier in the second sentence addresses the conflict (different methods, time periods etc) but I can't find anywhere on the ACS pages where the Census recommends we not compare ACS to decennial Census. And when I search by subject the Journey to Work topic states that I can 'Compare' it to the 2000 Census.
It does specifically state not to compare overlapping years on multiyear estimates.

Kendra Watkins
Senior Data Analyst
Mid-Region Council of Governments
809 Copper Ave. NW
Albuquerque, NM 87102
Phone: (505)724-3601
Fax: (505)247-1753
Email: kwatkins@mrcog-nm.gov< mailto:tgaudette@mrcog-nm.gov>


This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to which they are addressed. If you have received this email in error please notify the system manager. This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail.

From: ctpp-news-bounces@chrispy.net [ mailto:ctpp-news-bounces@chrispy.net] On Behalf Of Weinberger, Penelope
Sent: Monday, January 31, 2011 8:41 AM
To: ctpp-news@chrispy.net
Subject: RE: [CTPP] 2000 and 2006-08 work trip comparison question.
The CB does not recommend comparing an ACS based data set to a CB Long Form based data set.  The 2000 data represent a point in time estimate, the ACS data represent a period estimate.  Furthermore, the Census Bureau recommends not comparing period estimates with overlapping years.
Penelope Weinberger
CTPP Program Manager
AASHTO
202-624-3556
http://ctpp.transportation.org/Pages/default.aspx
It's just as bad to not make a plan as to blindly follow the one you already have.
From: ctpp-news-bounces@chrispy.net [ mailto:ctpp-news-bounces@chrispy.net] On Behalf Of Seidensticker, Dan
Sent: Friday, January 28, 2011 4:18 PM
To: (ctpp-news@chrispy.net)
Subject: [CTPP] 2000 and 2006-08 work trip comparison question.
We downloaded the2006-2008 ACS county-to-county worker flow for Dane County, Wisconsin from http://ctpp.transportation.org/Pages/3yrdas.aspx.
The question we now have...can that data be compared to the county-to-county 2000 CTPP work trips  to determine any statistically significant increase/decrease?  If so, how would one calculate the margin of error?
Dan Seidensticker
GIS Specialist
Madison Area Transportation Planning Board:
A Metropolitan Planning Organization (MPO)
City of Madison Planning Unit
121 S. Pinckney Street, Suite 400
Madison, WI 53703
Voice: 608-266-9119
Fax: 608-261-9967
Email: dseidensticker@cityofmadison.com< mailto:dseidensticker@cityofmadison.com>
www.MadisonAreaMPO.org < http://www.madisonareampo.org/>

------------------------------------------------------------------------
_______________________________________________
ctpp-news mailing list
ctpp-news@chrispy.net
http://www.chrispy.net/mailman/listinfo/ctpp-news


--
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@chrispy.net
http://www.chrispy.net/mailman/listinfo/ctpp-news

-------------------------------------------------------------------------------------------------
Patricia C. (Patty) Becker         248/354-6520
APB Associates/SEMCC       FAX 248/354-6645
28300 Franklin Road                   Home 248/355-2428
Southfield, MI  48034                     pbecker@umich.edu