Have an idea?

Visit Sawtooth Software Feedback to share your ideas on how we can improve our products.

Strategy for Alternative Specific Design

In your help file you advice complete enumeration as primary design strategy for attribute specific designs. I've just checked D-efficiency of complete enumeration and balanced overlap. To my surpprise CE is worse on std err for main effects of attribute specific atributes and overal D-efficiency. How can I explain that?

my design is simple

brand-> promotion (brand1) + special promo (brand7)

asked Nov 24, 2011 by anonymous
retagged Sep 13, 2012 by Walter Williams

1 Answer

+1 vote
How much is the difference in "Strength of Design" (relative D-efficiency) using the Advanced Design Test between Complete Enumeration and Balanced Overlap?  The "Quick Test" that automatically is shown after you generate the design (which estimates standard errors) is not as reliable as the Advanced Test with Simulated Random Respondent Data.
answered Nov 24, 2011 by Bryan Orme Platinum Sawtooth Software, Inc. (198,715 points)