In the research variable scale I want to use, one factor is measuring only two items. Is it okay to use a scale that only measures 2 items on one factor, since this will lower the reliability? Or can I remove the factor that only measures 2 items and use it?
You have to check first the factor loading. If it's high you don't have to remove the factor, but you have to do so, if it's not high
That depends on what you want to do. If you want to do a CFA, it's possible to have a factor with only two items, if you have other factors with enough items to have more than zero degrees of freedom. However, it is important to remember that this will make that factor less reliable and you should think if those two items are really important for the theory.
- Badges
- Science topic
- Similar topics
- Statistics
- Survey