I developed a brief questionnaire with 20 items and 10 components, each component has 2 items. factor analysis confirmed the structure of the scale, and the total Cronbach alpha was 0.88. can I report the Cronbach of components that have 2 items?
I see the articles of BFI-10 (the brief version of the big five inventory) that they report Cronbach alpha for components. BFI-10 has 10 items and 5 components.
But some professors believe that for two items, the researcher shouldn't report Cronbach and the researcher should report a correlation between two items.
Do high-level journals accept the reporting the Cronbach's alpha of components with two items?
ideally any factor should have more than 3 items (Andy Field, 2013; Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. 2010).
However, we can calculate Cronbach Alpha with even 2 items. No of item is not a criterion for publication in a good journal.
Hope it works!
I agree with your supervisor, simple correlation will be better indicator.
I think 10 item brief personality scale was not developed for academic purposes. There are many double barrel items in that scale and it is impossible to get high Cronbach's Alpha coefficient values. If i were you i would not use it for academic purposes. If i were the author of that scale. I would suggest people not to use it for covariance based analysis. It just makes no sense, sorry.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Applied Statistics
- Survey Design
- Questionnaire