|
![](http://www.statmodel.com/discussion/icons/csc/1px.gif) |
CFA with different response options |
![Previous](http://www.statmodel.com/discussion/icons/csc/navgen_arrow_left.gif) ![](http://www.statmodel.com/discussion/icons/csc/1px.gif)
|
Message/Author |
|
Scott Smith posted on Monday, September 16, 2013 - 1:08 pm
|
|
|
Is it methodologically sound to run a CFA to find the fit of two items with likert response options of 1,2,3 and 4 with one item that only has dichotomous response options of 1 and 2? |
|
|
You can combine ordered categorical and binary items. With only three factor indicators, the model is just-identified so you will not be able to assess model fit. |
|
Back to top |
|
|