Date:         Tue, 11 Jul 2000 12:04:23 -0400
Reply-To:     Paige Miller <paige.miller@KODAK.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Paige Miller <paige.miller@KODAK.COM>
Organization: Eastman Kodak Company
Subject:      Re: Pairwise Conjoint Analysis
Content-Type: text/plain; charset=us-ascii

Toni Price wrote:
>
> Does anyone know where I can find information on how to implement pairwise
> conjoint analysis with SAS? I have found a number of examples for
> choice-based and other forms of conjoint analysis, (eg. in tech. report
> R-109) but couldn't see anything specifically on pairwise.
>
> I need to find out how to generate efficient designs, how to compute the
> number of runs needed for a given accuracy of the analysis, and exactly how
> to apply the analysis - if there are any technical papers explaining this
> (not necessarily in the SAS context) that would really help too.

Not specifically related to conjoint analysis, an article by Kempthorne
may be of help. It describes how to create incomplete cyclic paired
designs, do ANOVA on the results, and also provides estimates of the
design's efficiency.

Kempthorne, O. (1953) A class of experimental designs using blocks of
two plots, Ann. Math. Stat., 24, 76-84.


--
Paige Miller
Eastman Kodak Company
paige.miller@kodak.com

"It's nothing until I call it!" -- Bill Klem, NL Umpire
"Those black-eyed peas tasted all right to me" -- Dixie Chicks