Dear forum,
I hope you're well. I've been using the -conjoint- command/package. This enables the analysis of conjoint experiments (a popular research design in the social sciences) in Stata and details of the command/package are here: https://www.stata.com/meeting/uk21/s...UK21_Frith.pdf
However, I've been having some problems. In my experiment, I have nine variables with three levels each. Everything works when I run models with any three variables, but when I run models with more than three variables I get the 'no observations' red error line. However, the included variables do have observations.
When I remove the levels from the variables (i.e. just test for the influence of the variables themselves, rather than their different levels), I can run models with eight variables. However, when I add a ninth variable, I get the 'not possible with test' red error line.
Would anyone know what the problem(s) might be? Any help would be greatly appreciated. Best and thanks,
Tom
I hope you're well. I've been using the -conjoint- command/package. This enables the analysis of conjoint experiments (a popular research design in the social sciences) in Stata and details of the command/package are here: https://www.stata.com/meeting/uk21/s...UK21_Frith.pdf
However, I've been having some problems. In my experiment, I have nine variables with three levels each. Everything works when I run models with any three variables, but when I run models with more than three variables I get the 'no observations' red error line. However, the included variables do have observations.
When I remove the levels from the variables (i.e. just test for the influence of the variables themselves, rather than their different levels), I can run models with eight variables. However, when I add a ninth variable, I get the 'not possible with test' red error line.
Would anyone know what the problem(s) might be? Any help would be greatly appreciated. Best and thanks,
Tom
Comment