Hi Cliff,
A few of us (Ed Christopher, Brian McKenzie and I) had a brief discussion on your question offline, and here are our thoughts:
The short answer is that there is no single source of guidance on this. There are cases where people use the resident population because it very often corresponds with and serves as a proxy for the working population,
like the CTPP transportation profiles. When we did our profiles at state and sub-state geographies CTTP tables we used the resident populations as well as the state-level design favors. Here is the link to the various profiles we have done.
http://ctpp.transportation.org/Pages/profiles.aspx
We have also seen guidance that recommends using the state-level or even national-level design factor using the 2000 Census documentation, because workers can potentially come from anywhere within the state or
cross state boundaries.
To help you with the design factors they are imbedded in the worksheet tool developed by CUTR which can make your work a lot easier:
http://www.nctr.usf.edu/spreadsheet/77802.xls. They have a report that can be found here along with more documentation.
http://www.nctr.usf.edu/abstracts/abs77802.htm.
Thanks,
JJ
Jingjing (JJ) Zang
Cambridge Systematics, Inc.
t +1 301 347 9130 (ext. 53130)
From: ctpp-news [mailto:ctpp-news-bounces@chrispy.net]
On Behalf Of Cook, Cliff
Sent: Tuesday, January 09, 2018 12:01 PM
To: ctpp-news@chrispy.net
Subject: [CTPP] Computing Standard Error for 2000 CTPP Part 2
I am trying to compare the size of the workforce in my city, as reported by the 2012-16 ACS, to the size reported by the 2000 CTPP Part 2 Table. While I can do a straightforward comparison of figures, I’d like to evaluate the difference
taking into account the 90% margin of error.
I can use 2000 Census documentation to determine how to compute MOEs for figures reported for my community’s own population, households and housing stock. However, I am not sure how to approach this problem for the workforce, since the
workforce is based on a sample from a larger population, perhaps county, state or national. Any suggestions are appreciated.
Cliff Cook
|
||||||||||||
|
||||||||||||
