Dear Polmeth members,
I have a question about drawing a sample and random assignment in a survey experiment. Any responses and comments will be greatly appreciated.

I am designing a survey experiment to test the effect of partisan cues on a policy issue. I will be drawing a nationally representative sample(s). Many studies I looked at draw a single sample and then randomly assign respondents to control and experimental groups. Is this common practice? Can I start with two samples? In other words, can I draw two separate samples in the beginning and choose one to be the control and the other to be the experimental group? which method would be more appropriate?  The total sample size will be kept at 1000 in both cases (in case this information has any relevance).

thank you very much for your comments

Sabri

 
-- 
Sabri Ciftci
Assistant Professor
Department of Political Science
Kansas State University
226 Waters Hall
Manhattan, KS 66506
p: +1 785 532 3514
f: +1 785 532 2339
@: [log in to unmask]
   [log in to unmask]

**********************************************************
             Political Methodology E-Mail List
   Editors: Xun Pang        <[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

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