Hi listers,
I am trying to analyse data from a discrete choice experiment (DCE). Each
patient is given 10 scenarios, each with two treatment choices A and B
(coded 0 and 1). Each choice has different combinations of the following
attributes of the treatment: length frequency duration benefit. Each
attribute has 3 levels, for instance length is either 6 weeks, 10 weeks and
16 weeks. Data looks like this:
id scenario choice_made A_length A_ freq A_dur A_ben B_length B_freq...
1 1 0 1 1 1 1 2 3
1 2 1 2 1 3 2 3 3
1 3 1 3 2 2 1 1 1
...
2 1 0 1 1 1 1 2 3
2 1 0 2 1 3 2 3 3
...
Data also has a list of socioeconomic and demographic variables. My
preferred model is a random effects model. I am not sure whats the best way
to proceed. Is the following approach correct?
xtprobit choice_made A_length A_freq A_duration A_benefit B_length B_freq
B_duration B_benefit age sex education occupation, i(id)
I am new to DCE design, so any help would be greatly appreciated.
Many thanks,
Shehzad
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
