You are not logged in. You can browse but not post. Login or Register by clicking 'Login or Register' at the top-right of this page. For more information on Statalist, see the FAQ.
@Jared. Thank you. My worry will be that among the responses, there are some cells that has very few counts. I am thinking this will affect my result. Can I use multinominal regression.
You can look to combine such choices into an "other" category. There are also tests for combining dependent categories in multinomial logit, so these may suggest combining the seldom chosen options with those more frequently chosen. If your scale is ordinal, then the combination should be with adjacent categories. See Richard Williams's notes on tests for combining dependent categories: https://www3.nd.edu/~rwilliam/stats3/Mlogit2.pdf
Just to add a bit to what Andrew said: if you have a very skewed distribution where some response categories are sparse, using multinomial logit regression will not improve the result. I think it will make things worse.
Be aware that it can be very hard to answer a question without sample data. You can use the dataex command for this. Type help dataex at the command line.
When presenting code or results, please use the code delimiters format them. Use the # button on the formatting toolbar, between the " (double quote) and <> buttons.
Comment