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

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