Have an idea?

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

Testing attribute significance in Logit

I have run a logit analysis on my data and now i would like to know where i need to look or what I should calculate to check for attribute significance.
I hear somebody saying that with latent class you look at the t-ratio does this also hold for logit?

Additional on this question: i have some respondents who have filled in their conjoint questions but havent filled in the demographics. Can i still just use them in the analysis of latent class and then only use the demographic data that is available to describe the classes?

Many thanks in advance!!
asked Jul 24, 2017 by roelev (120 points)

1 Answer

0 votes

Yes, that also holds for logit - the output will give you what you need for those t-tests of attributes' levels' significance.

The Latent Class software doesn't require any demographics at all (this is a latent class version of multinomial logit, not a traditional cluster analysis-like latent class model) so the fact that you have incomplete demographics won't hurt you at all.
answered Jul 24, 2017 by Keith Chrzan Platinum Sawtooth Software, Inc. (95,675 points)