I've conducted a first order EFA and after that I conducted a second order EFA. In second order EFA one of the factor loadins gave a negative value. In first order EFA I know that this situation is because the item means the oposity and I have do reverse the coding if I want for exemple calculate the internal consistency in SPSS. But in second ordem EFA, what I have to do for calculating internal consistency or orther analysis?
Factor loading should be negative. You need to do some reverse coding for the particular items
If you expect all your second-order factor to be positively correlated, then they should indeed all be coded so that they worded in the same direction. In addition, if a subset of the items is negatively correlated with the rest, this could also have affected your first-order EFA, by grouping those items to together in distinction from the positively worded items.
I recommend making sure that your items are all in the same direction, from the beginning, and then re-running the initial EFA.
I believe what Prof Zainudin Awang means was factor loading should not be in a negative value.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS