Hello dedicated Stata users,
I am a doctoral student who is preparing for a dissertation. I have found out that I need to employ conjoint experiments for my dissertation, and I am trying to learn how to create efficient experimental designs using the "dcreate" command in Stata.
For my experiments, I have 8 variables (i.e., x1, x2, x3, x4, x5, x6, x7, x8) which have 2, 3, 2, 2, 2, 4, 3, and 2 levels, respectively. Therefore, there are 1,152 total bundles (i.e., full factorial). I would like to display 3 alternatives per question, so I typed the below code in Stata.
dcreate i.x1 i.x2 i.x3 i.x4 i.x5 i.x6, nalt(3) nset(???) bmat(b)
My question is: How do I know the value of "nset"?
I understand that "nset" indicates the number of choice sets in the design. In that case, should I type "384" (i.e., 1,152 / 3)?
The reason that I am asking this question is that, in the 1st example of the Stata's help file for the 'dcreate' command, it uses 6 variables that have 4, 4, 2, 2, 2, and 2 levels respectively. In that case, the full factorial is 256. It uses 2 alternatives per question, but then it uses "nset(16)" instead of "nset(128)".
To be more clear about my question, how is the value for "nset" is determined? 1) Should it be always 'the number of full factorial / the number of alternatives per question'; or 2) can I choose its value? If I can choose the value of "nset," how do I know what values should be used?
Thank you very much for your attention to my inquiry. I am having a very hard time figuring out how to construct conjoint experiments, and it would be deeply appreciated if you can please help me understand the aforementioned issue.
I am a doctoral student who is preparing for a dissertation. I have found out that I need to employ conjoint experiments for my dissertation, and I am trying to learn how to create efficient experimental designs using the "dcreate" command in Stata.
For my experiments, I have 8 variables (i.e., x1, x2, x3, x4, x5, x6, x7, x8) which have 2, 3, 2, 2, 2, 4, 3, and 2 levels, respectively. Therefore, there are 1,152 total bundles (i.e., full factorial). I would like to display 3 alternatives per question, so I typed the below code in Stata.
dcreate i.x1 i.x2 i.x3 i.x4 i.x5 i.x6, nalt(3) nset(???) bmat(b)
My question is: How do I know the value of "nset"?
I understand that "nset" indicates the number of choice sets in the design. In that case, should I type "384" (i.e., 1,152 / 3)?
The reason that I am asking this question is that, in the 1st example of the Stata's help file for the 'dcreate' command, it uses 6 variables that have 4, 4, 2, 2, 2, and 2 levels respectively. In that case, the full factorial is 256. It uses 2 alternatives per question, but then it uses "nset(16)" instead of "nset(128)".
To be more clear about my question, how is the value for "nset" is determined? 1) Should it be always 'the number of full factorial / the number of alternatives per question'; or 2) can I choose its value? If I can choose the value of "nset," how do I know what values should be used?
Thank you very much for your attention to my inquiry. I am having a very hard time figuring out how to construct conjoint experiments, and it would be deeply appreciated if you can please help me understand the aforementioned issue.
