User Community
cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Show  only  | Search instead for 
Did you mean: 
  • New to JMP? Join us Sept. 23-24 for the Early User Edition of Discovery Summit, tailor-made for new users. Register now for free!
  • Use World Cup data to build models, explore spatial relationships, and create informative visualizations in JMP. Register. July 17, 2 pm US Eastern Time.
  • Your voice matters! Tell us how you prefer to receive JMP updates, so we can tailor our communication to your needs. Take short survey.

JMP Wish List

We want to hear your ideas for improving JMP. Share them here.
Choose Language Hide Translation Bar
0 Kudos

Allow for disallowed combinations in Choice Designs

Status: New Submitted by 8 hours ago
Comment

What inspired this wish list request? 

I am currently using a JMP Student License (version 19) to to design a Discrete Choice Experiment using the Choice Design platform of DOE. I am using both the prior means and prior variances capabilities allowed for by the choice design platform. 
 
I would like to prevent certain combinations of attribute levels from appearing together order to remove illogical combinations for the situation of my experiment and reduce respondent fatigue. As far as I am aware, there is no ability to do this in the choice design platform.
 
I am aware that the custom design platform offers the ability to disallow certain combinations, but you cannot use prior means/distributions as you would be able to in the choice design platform.

What is the improvement you would like to see? 

I would like to see the ability to disallow combinations integrated into the Choice Design platform as it is in the custom experiment platform, so that users are able to create designs that simultaneously make use of Bayesian priors, partial profiles, and the ability to disallow combinations. 

Why is this idea important? 

I was driven to use JMP to develop my choice design due to how it handles partial profiles and Bayesian priors, which is more built-in than other DCE softwares, but was surprised that it didn't have the ability to disallow combinations. Other DCE design softwares do have the ability to disallow combinations, but don't work with partial profiles in the way that JMP does, so as of now it feels that there is no "perfect" option to move forward with. Given than disallowing combinations can be very important in designing a DCE, adding this ability to JMP would be very beneficial in terms of allowing users to create more complex designs.

As of now, I have the option to manually remove disallowed combinations from choice sets generated by JMP, but I have concerns that doing this manually would then effect the statistical properties associated with the Bayesian priors. Although of course disallowing combinations generally has an effect on the efficiency of a design, integrating it into the platform would lessen this worry about doing it manually.