I have adopted and validated a 54 item School connectedness scale for my thesis. After the cultural adaptation 2 items removed. After EFA 11 items dropped and following CFA 9 items dropped and only 31 items were retained. Want to know whether this 31 item scale is valid the proceed futher for my thesis?
Dear Indika Nupehewa
Yes you can move forward with 31 items for further analysis.
But try to use modification indices while doing CFA in AMOS, if not used.
Use the bellow attached link for your reference.
https://www.youtube.com/watch?v=JkZGWUUjdLg
I hope it helps! :)
Thank you Bhumika Bunkar for replying to my question. I have already used modification indices. Finally, I've got model fit by getting IFI Delta 2 0.901. Hope this model will suite to my country Sri Lanka. Thanks again.
Hi all,
the most prevalent question would be how many and which latent variables these 31 items shall measure and what the phrase "validated measure" really means. Most scales were developed using PCA and not factor modeling and even if, the majority failed a strict test.
For each latent variable, 2-3 pure and clean items should suffice as--when you have a set of k items that all measure an attribute--every set of lower than k will also measure the attribute. Yes, you reliability will be lower but validity comes first and for that a strict test is essential.
Best,
Holger
- Badges
- Science topic
- Similar topics
- Philosophy
- Moral Philosophy
- Free Will