st: a stat question

["Data Analytics Corp." <dataanalytics@earthlink.net>]

Hi,

I have a client who has been conducting a survey for several years, each 
year being a "wave".  This year, the whole survey was changed: most of 
the questionnaire, sample design and size, locations (prior waves had 
only four cities, the new one has 8 cities).  I was asked about doing 
some significance testing on a few variables.  I was very uneasy about 
this because everything about the surveys is different.  I feel that any 
comparisons would be shaky because so much has changed.  Stat theory 
doesn't seem to say anything about comparing across different waves such 
as this.  Any thoughts on doing significance tests on waves of surveys 
where everything about the surveys have changed from one wave to the 
next?  Any literature I can be pointed to?  I've run into this before 
with other clients, so it's a common occurrence.

Thanks,

Walt

--
________________________

Walter R. Paczkowski, Ph.D.
Data Analytics Corp.
44 Hamilton Lane
Plainsboro, NJ 08536
________________________
(V) 609-936-8999
(F) 609-936-3733
dataanalytics@earthlink.net
www.dataanalyticscorp.com

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/