|
|
EFA with ordered categorical variables |
|
Message/Author |
|
EWickwire posted on Saturday, February 18, 2006 - 7:18 pm
|
|
|
Hi, I'm attempting an exploratory factor analysis (n=538) with 50 items. The items are scored on a 5-point scale. The midpoint is "neither" between opposing endpoints ("very good" to "very bad" e.g.). As expected C was the modal score, but variability was acceptable. There are indications of a general factor although switching away from SPSS improved the rotations tremendously. I don't know much about general factors in EFA. I'm using ML (obviously) extraction. It's been suggested to use polychoric correlation matrix in an EFA. Is that the appropriate analysis? Can you show me how to proceed? Many Thanks, Wick |
|
EWickwire posted on Saturday, February 18, 2006 - 7:26 pm
|
|
|
PS: It also may be appropriate to delete items after an initial EFA. (In the initial EFA I end up with a huge factor and a scree plot difficult to interpret with 3-4-5-6-7 factors very close. I run parallel analysis, which indicates retain 5 factors, so that is what I have done). If it's appropriate to delete items, how would you proceed? And would you then rerun the EFA, even continuing to tweak, until "simple structure" is achieved? I very much appreciate your help! |
|
EWickwire posted on Saturday, February 18, 2006 - 8:10 pm
|
|
|
Revised in EFA area. |
|
Back to top |
|
|