Yes, I noticed something similar: 417 respondents who stated that they
voted for Barack Obama also indicated that they would never vote for him. I
contacted the ANES about this back in 2009, and a representative indicated
that the labels for Q15a were incorrectly reverse coded; Barack Obama was
supposed to be coded lower than John McCain on all items, but this did not
happen in the CSES section. The representative indicated that an erratum
would be posted, which is here:
http://www.electionstudies.org/studypages/2008prepost/2008prepost_errata20090324.htm.
Looks like your student found a similar error.
L J Zigerell
http://ljzigerell.wordpress.com/
**
On Fri, Feb 17, 2012 at 3:16 PM, Paul Gronke <[log in to unmask]> wrote:
> Hey all! A student came to me today who trusted his instincts and not the
> mathemagic of a survey, and I think he has discovered a miscoding in the
> 2008 NES.
>
> He is using the SDA Archive version of the dataset at Berkeley. He was
> examining vote choice against "V085188b", an 11 point "Dislike / Like"
> measure for Obama.
>
> The results struck him as bizarre; the more a respondent said the liked
> Obama, the LESS likely they were to vote for Obama.
>
> The variable correlates at a -.50 with V083037a, which is the "Feeling
> Thermometer: Democratic Presidential Candidate."
>
> I can only assume that V085188b is coded in reverse somewhere, either by
> NES or when placed into SDA. Besides giving the student an A for effort,
> I'm wondering if anyone else has seen this?
>
> Thanks!
>
> ---
> Paul Gronke Ph: 503-517-7393
> Fax: 734-661-0801
>
> Professor, Reed College
> Director, Early Voting Information Center 3203 SE Woodstock Blvd.
> Portland OR 97202
>
> **********************************************************
> Political Methodology E-Mail List
> Editors: Diana O'Brien <[log in to unmask]>
> Jon C. Rogowski <[log in to unmask]>
> **********************************************************
> Send messages to [log in to unmask]
> To join the list, cancel your subscription, or modify
> your subscription settings visit:
>
> http://polmeth.wustl.edu/polmeth.php
>
> **********************************************************
>
**********************************************************
Political Methodology E-Mail List
Editors: Diana O'Brien <[log in to unmask]>
Jon C. Rogowski <[log in to unmask]>
**********************************************************
Send messages to [log in to unmask]
To join the list, cancel your subscription, or modify
your subscription settings visit:
http://polmeth.wustl.edu/polmeth.php
**********************************************************
|