Marketing Research, Conjoint, Discrete Choice, Experimental Design
Reports TS-643 TS-630A TS-622 TS-621 TS-273 can be obtained from:
http://www.sas.com/service/techsup/tnote/tnote_stat.html
These reports go into great detail on finding experimental designs for
conjoint and choice models. Your goal should not be to create an
orthogonal design; it should be to create a good efficient design that
may or may not be orthogonal.
In article <[EMAIL PROTECTED]>, pokh <[EMAIL PROTECTED]> writes:
|> I need to conduct conjoint analysis.
|> I have four factors each with three levels of attributes.
|> Total number of cards is 81. It's too much for a person to range.
|>
|> Thus I generated an orthogonal array using SPSS. It has 27 cards. It's
|> much
|> less than 81 ;-). But why does it have duplicated cards?
|> If I delete duplicates of the cards, than this plans is not orthogonal.
|>
|> What should I do?
|>
|> Thanks in advance,
|> Irina.
|>
--
Warren F. Kuhfeld SAS Institute Inc. (919) 677-8000 x17922
Multivariate Models R&D Cary, NC 27513-2414 (919) 677-4444 (Fax)
R5116 SAS Campus Dr. www.sas.com The power to know
Opinions expressed here are mine and not necessarily those of SAS.
Marketing Research, Conjoint, Discrete Choice, Experimental Design
Reports TS-643 TS-630A TS-622 TS-621 TS-273 can be obtained from:
http://www.sas.com/service/techsup/tnote/tnote_stat.html
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
