Have an idea?

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

base-case-scenario for sensitivity analysis

Dear community,

i am assessing a cbc study of 330 participants, having  8 attributes with 3-5 levels, 10 random & 3 fixed holdout tasks with 4 concepts + none option.

After chosing my model I want to conduct a sensitivity analysis in my master thesis.  As  number of products for the base-case-scenario were not specified I am wondering how many products I should include.

Should I  use 4 products (equals the number of level for attribute  "brand"and the amount of concepts the participant had to decide from)?

Thank you in advance,

asked May 28, 2018 by botmar (395 points)

1 Answer

+1 vote
Keeping the number of brands to equal the number of products in the base case seems like a reasonable choice.   Also, using a number of alternatives that's the same as the number in your designed choice sets seems like another easily defensible  choice.
answered May 29, 2018 by Keith Chrzan Platinum Sawtooth Software, Inc. (111,275 points)