lclogit error - weights must be the same for all observations in a group r(407)

[Rebecca Boehm]

Dear Statalist -

I am using -lclogit- to analyze data from a choice experiment using the following model:

lclogit choice attribute_1 attribute_2 attribute_3 .... attribute_n, id(choice_maker_id) group(choice_experiment_question_id) nclasses(3)

I get the following error when I run this model:

Iteration 0:  log likelihood = -25298.937
weights must be the same for all observations in a group
r(407);

I suspect the error is caused by the way I have numbered the variable "choice experiment question" as 1-8 for all choice makers, since each person had to respond to 8 choice sets. Does anyone know if the choice experiment identifiers must be unique across all choice makers? Instead of 1-8 they must be 1-8 for choice maker 1, 9-17 for choice maker 2, and so forth?

Thanks for your help!

[Victor Chen]

Well, 1-8 for choice maker1, 1-8 for choice maker1 as well. and SO forth.
Based on my experiences.
I am just wondering do you have some groups (choice maker i) who have more than 8 choices options (1-8, 9 or more)? or choice maker choose more than one in their choice?

[Lou Nadeau]

I am wondering if this was ever solved? I am having the same issue. Each time I try to run the lclogit command I get
weights must be the same for all observations in a group Each respondent in my data made 3 choices from a set of 3 options (A, B, and status quo) and represents nine observations in the data. I made sure I had three choice instances from each respondent and that each actually selected an option in each choice question.

Any insight into the source of this error? Seems like it must have to do the data structure since the command itself does not involve specifying a weight. I figured that balancing the data should do it. (There are also no missing values in the data I am using, so no obs are getting dropped for that reason.)

Thanks

[Joel Tona]

I had the same problem. When reading the description of the command,one would expect that the variable that corresponds to id, you should put the identifier of the choice the respondent is making. For example, if each respondent is making 3 choices, every choice consisting of four alternatives, the variable that identifies the choice the respondent is making would be would be 1 1 1 1,2 2 2 2 ,3 3 3 2,1 1 1 1,2 2 2 2,3 3 3 3,1 1 1 1 ,2 2 2 2,3 3 3 3... and so on. What you must do in these case is not discontinuing the identifier of the choice. So, you shold put 1 1 1 1 1,2 2 2 2,3 3 3 3,4 4 4 4,5 5 5 5 and so on. This should resolve the problem