What steps should I take to prepare my DCE data for analysis in STATA

[Suli Stephen]

Greetings, I am a beginner in analyzing DCE data, and I need to know if there is any document that can guide me in preparing my data for analysis. I have read the STATA commands that I can use, on Train book and on STATA manual, I am however stuck since I need to make my data in the way that I can analyze it. In summary my data have 325 observations in which each participant made 9 choices which were randomly drawn from a set of different scenarios, among three alternative each. In either of those alternative two alternative were to take the services and the third one was to remain in status quo. So, I read the clogit command, which was written to be ideal to estimating the coefficients for my attributes to start with and then go further to estimate the utility. Since my data contain nine set of choices for each participant, how do I generate one variable that can represent all of the nine set of choices? Is there a book apart from the Train Book that one can recommend that helps in steps for preparing the DCE data for analysis in STATA?

Thanks and kind regards

Suli

[Bridget Johnston]

Hello,

I haven't come across a guide for preparing the dataset for Stata, but I can make two suggestions.

The first is this guide which has a few useful examples of what your data will need to look like to use in Stata. You'll have to put it into a long format with a line for each choice. So if there are 3 per card, you'll end up with 3 observations for that choice card.

http://www.capacityplus.org/files/re...user-guide.pdf

I've always used support.CEs in R for creating the data set. You can Google the documentation and there is a very useful article in Journal of Statistical Software.

https://www.jstatsoft.org/article/vi...c02/v50c02.pdf

Good luck,
Bridget

[Suli Stephen]

Dear Bridget,

Thank you for the reply. I will have a look at the guide and see if I can get an idea of that I should do with the data.

Regards
Suli