What to do when Cronbach's alpha is really low?

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 Ceecy » Tue Jun 08, 2021 9:50 am

I have been trying to test the reliability of my variables.
For one variable, Cronbach's alpha is 0.211, and I did not reverse design the four subquestions' Likert scales for this construct in the survey. So I can not reverse scaled items, it does not help.
I've attached the correlational matrix and the low Cronbach's alpha, can you give me some suggestions to boost my Cronbach's alpha on this matter?

Thank you so much!
iShot2021-06-08 17.41.19.png
iShot2021-06-08 17.41.19.png (97.81 KiB) Viewed 4062 times
iShot2021-06-08 17.41.36.png
iShot2021-06-08 17.41.36.png (100.41 KiB) Viewed 4062 times
Posts: 7
Joined: Sun May 16, 2021 2:23 pm

by Bobafett » Tue Jun 08, 2021 12:02 pm

Sorry - why exactly can't you reverse score the offending item? I'm not clear why you can't take advantage of the inbuilt function of doing this on the fly in jamovi?

Also 4 items is really scraping the barrel when it comes to Cronbach's alpha - usually the recommendation is no less than 10 items as there can be constraints on how high the coefficient can reach. In that case, you might wish to look at other options such as average inter-item correlation, though that too may be limited as you'll only have 6 values to deal with...
Posts: 20
Joined: Thu Jul 18, 2019 11:33 am

Return to Help