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: 
Check out the JMP® Marketplace featured Capability Explorer add-in
Choose Language Hide Translation Bar
Thestudent
Thestudent
Level I

How to weigh 5 as the best and 0 as the worst. Multiple regression (fit model)

Created: Apr 5, 2022 08:03 AM  |  Last Modified: Jun 10, 2023 4:46 PM (614 views)

Its many years since I used JMP, so sorry for the stupid question. 

I made a survey where the answers is rated from 0 (Not included) to 5 (The best). I need to have this included in my regression model. How do I do that? Its concerning many variables which all explains preferences. Some are very important (rated 5) and some are less important (from 4 to 0) 

I just need to know how I weigh them in my fit model 

 

 

1 ACCEPTED SOLUTION

Accepted Solutions
Mark_Bailey
Staff Mark_Bailey
Staff
Solution

Re: How to weigh 5 as the best and 0 as the worst. Multiple regression (fit model)

Apr 5, 2022 08:23 AM (606 views)  |  Posted in reply to message from Thestudent 04-05-2022

Is not the rating (0-5) the response? If so, you can use Numeric or Character data type, but use the Ordinal modeling type. Enter this column in the Y role after selecting Fit Model from the Analyze menu. This action will automatically change the Personality to Ordinal Logistic regression. You will model the log odds of the cumulative logit of the response.

 

Is that your question?

View solution in original post

0 Kudos
1 REPLY 1
Mark_Bailey
Staff Mark_Bailey
Staff
Solution

Re: How to weigh 5 as the best and 0 as the worst. Multiple regression (fit model)

Apr 5, 2022 08:23 AM (607 views)  |  Posted in reply to message from Thestudent 04-05-2022

Is not the rating (0-5) the response? If so, you can use Numeric or Character data type, but use the Ordinal modeling type. Enter this column in the Y role after selecting Fit Model from the Analyze menu. This action will automatically change the Personality to Ordinal Logistic regression. You will model the log odds of the cumulative logit of the response.

 

Is that your question?

0 Kudos