Re: Cohen data suspect at best

["Rad health" <healthrad@HOTMAIL.COM>]

Then Ruth how do you explain their findings?  This just points out the 

difference between grouped data and individual level data by state.  It is 

far from linear.



Here is just one other paper to support my contention concerning the 

uncertainty with SES economic aggregate data.  There are many.



Am J Public Health  2001 Apr;91(4):632-6



Using aggregate geographic data to proxy individual socioeconomic status: 

does size matter?



Soobader M, LeClere FB, Hadden W, Maury B.



Department of Pediatrics, University of Rochester, Rochester, NY, USA.



OBJECTIVES: This study assessed whether aggregate-level measures of 

socioeconomic status (SES) are less biased as proxies for individual-level 

measures if the unit of geographic aggregation is small in size and 

population. METHODS: National Health Interview Survey and census data were 

used to replicate analyses that identified the degree to which aggregate 

proxies of individual SES bias interpretations of the effects of SES on 

health. RESULTS: Ordinary least squares regressions on self-perceived health 

showed that the coefficients for income and education measured at the tract 

and block group levels were larger than those at the individual level but 

smaller than those estimated by Geronimus et al.



CONCLUSIONS: Researchers should be cautious about

use of proxy measurement of individual SES even if proxies are calculated 

from small geographic units.









>From: RuthWeiner@aol.com

>To: healthrad@hotmail.com, blc+@PITT.EDU

>CC: radsafe@list.vanderbilt.edu

>Subject: Re: Cohen data suspect at best

>Date: Mon, 28 Jan 2002 21:59:11 EST

>

>I do not have the wherewithal to validate census figures independently, so 

>I

>checked on a couple of minority populations that I happen to know about 

>from

>local (state) figures:

>

>New Mexico lists 42.5% Hispanics of all ages, which is the same figure 

>quoted

>by state agencies, the Hispanic Chamber, everybody.

>

>Washington DC lists 60% African American, and Maryland lists 27.9%.  Both 

>are

>consistent with numbers I have heard quoted.  I have heard that DC is 70%

>African American, but that would be 86% accurate if the 70% figure is

>accurate.

>

>Washington State lists 5.5% Asian, which again is consistent with other

>estimates I have read and heard. (we are talking the whole state, not just

>Seattle).

>

>Again, I didn't say there wasn't undercounting -- I just said it wasn't by

>80%.  Incidentally, I got the figures from www.census.gov.

>

>Ruth Weiner, Ph. D.

>ruthweiner@aol.com





_________________________________________________________________

Send and receive Hotmail on your mobile device: http://mobile.msn.com



************************************************************************

You are currently subscribed to the Radsafe mailing list. To unsubscribe,

send an e-mail to Majordomo@list.vanderbilt.edu  Put the text "unsubscribe

radsafe" (no quote marks) in the body of the e-mail, with no subject line. You can view the Radsafe archives at http://www.vanderbilt.edu/radsafe/