[RESOLVED] How to use oridinal variables for analysis...?

General help and assistance with jamovi. Bug reports can be made at our issues page: https://github.com/jamovi/jamovi/issues . (If you're unsure feel free to discuss it here)

by abhtony » Fri Aug 24, 2018 6:34 pm

Hi,

So if I have a column called "Opinion: Do you like this", and the data is populated as either "Agree", "Neutral", or "Disagree", how do I use this for analysis? The program seems to only accept ordinal variables that isn't text. (But I guess I want Jamovi to understand that Agree could be "3", and Disagree "1".)

I don't want to go through my whole data set and manually convert "Agree"s to 1s. Is there a smart way to do this?

Much thanks for the help!
abhtony
 
Posts: 2
Joined: Fri Aug 24, 2018 6:28 pm

by jonathon » Mon Aug 27, 2018 1:09 am

hi,

you can set variables which are text to be ordinal, and have been able to for a long time.

are you using a recent version?

kind regards
User avatar
jonathon
 
Posts: 869
Joined: Fri Jan 27, 2017 10:04 am

by abhtony » Sun Sep 16, 2018 9:02 pm

Hi,

I'm using the latest version on Windows... and yeah so I can set the texts to be ordinal and so on, but that disables the function to conduct inferential analysis for some reason.
abhtony
 
Posts: 2
Joined: Fri Aug 24, 2018 6:28 pm

by jonathon » Sun Sep 16, 2018 10:35 pm

hi,

the way you add labels to numeric values is through the variable editor (select a column and press F3, or select 'Setup' from the data tab). if you select each of the levels there, you can type in the labels "Agree", "Disagree", etc.

if you simply change the values in the spreadsheet to text values, then you're losing that numeric information, and the variable won't be usable in analyses that work on numeric values.

feel free to attach a data set if you stil can't get it to work.

jonathon
User avatar
jonathon
 
Posts: 869
Joined: Fri Jan 27, 2017 10:04 am


Return to Help

cron