Date:         Wed, 6 Sep 2006 23:21:27 -0700
Reply-To:     David L Cassell <davidlcassell@MSN.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         David L Cassell <davidlcassell@MSN.COM>
Subject:      Re: Survey data analysis
In-Reply-To:  <200609062029.k86AlCkH031753@malibu.cc.uga.edu>
Content-Type: text/plain; format=flowed

sophiDT@HOTMAIL.COM wrote:
>
>Dear SAS-L experts,
>
>I am trying to calculate the difference of variables of interest between 2
>years of survey data. I have proc surveymeans in SAS and I also have a SAS
>compatible sofeware SUDAAN that specifically for survey analysis (never
>used).  Now my question is how should I lay out the data? stack them with
>indicator of year 1 and 2? or put them side by side and calculate the
>differences first, then use proc surveymeans?  I hate to do the 2nd way if
>I can avoid to do it.  Any advices?
>
>Thanks in advance.
>
>Sophia

There is no point in moving forward this far, until you have checked that
you are doing the right thing.

Are you saying that you have panel data, and individual data points are
tracked
from year 1 to year 2?  Are you saying that the strata and clusters and
weights are unchanging across years?  Unless you have all these
qualifications,
you cannot use your Approach #1.

So I am guessing that you will have to take Approach #2.  If your samples
differ across years, or your design structures vary across years, then you
don't have a meaningful way of 'lining things up' for survey sample
analysis.

HTH,
David
--
David L. Cassell
mathematical statistician
Design Pathways
3115 NW Norwood Pl.
Corvallis OR 97330

_________________________________________________________________
Get real-time traffic reports with Windows Live Local Search
http://local.live.com/default.aspx?v=2&cp=42.336065~-109.392273&style=r&lvl=4&scene=3712634&trfc=1