Date: Fri, 20 Dec 2002 16:53:53 -0800
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: "David L. Cassell" <Cassell.David@EPAMAIL.EPA.GOV>
Subject: Re: help on power analysis
Content-type: text/plain; charset=us-ascii
Steve Albert <salbert@AOL.COM> replied [in part]:
> sampling theory here. You need to understand the sampling design, and
> issues involved. (You also haven't described how your health care
> are to be recruited, which may introduce some biases of its own, or
> they're told to select the n individuals from their communities, which
> introduce other biases, affect the correlation between observations in
> cluster, etc.)
Exactly. The OP is going to have a host of potential problems
with response errors and non-response errors. So a large part
of the process of designing the study is likely to be the design
of the *logistics* involved, so the fieldwork produces the design
that the statisticians planned.
> 185. A more mathematical treatment is in William Cochran's "Sampling
> Techniques", 3rd edition (a classic on sampling),
Cochran's book also has a list of the 11 steps involved in the
process of sampling. I have just written about this in an invited
paper I'll be doing at SUGI with AnnMaria de Mars (Rousey). I have
found more than one study where a good statistical design was ruined by
poor logistical planning and implementation. So the OP really needs to
talk with people who have experience *implementing* sampling designs.
The American Statistical Association has a short series of pamphlets
whch might be helpful for the beginner. They are written for the
general public, and are produced by:
Section on Survey Research Methods
American Statistical Association
1429 Duke St.
Alexandria, VA, USA
The titles I know of are:
"What is a survey?"
"How to plan a survey"
"How to Collect survey data"
"What are the main sources of survey error?"
"What about surveys in the media?"
BTW, some of the authors on these are Westat folks.
David Cassell, CSC
Senior computing specialist