In addition, I want to calculate a misclassification table on aggregate level. What is the right procedure for that?

I'm using Lighthouse Software, conducted a CBC.

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

In addition, I want to calculate a misclassification table on aggregate level. What is the right procedure for that?

I'm using Lighthouse Software, conducted a CBC.

0 votes

Eline,

For aggregate logit the effect sizes are the odds ratios. You can get these by simply exponentiating the part worth utilities. You can do this with the exp() function in Excel, for example.

Our software does not print out a classification table, so you would have to do that yourself. You know the actual choices so you would simulate the choices for each respondent in each choice set and then cross the actual by the predicteds. Our software is primarily used for choice experiments where a classification table usually does not make sense: the identity of alternatives A, B and C are not constant but change from one choice set to the next.

For aggregate logit the effect sizes are the odds ratios. You can get these by simply exponentiating the part worth utilities. You can do this with the exp() function in Excel, for example.

Our software does not print out a classification table, so you would have to do that yourself. You know the actual choices so you would simulate the choices for each respondent in each choice set and then cross the actual by the predicteds. Our software is primarily used for choice experiments where a classification table usually does not make sense: the identity of alternatives A, B and C are not constant but change from one choice set to the next.

...