Background: Before commencing treatment 180 participants completed a baseline 17-item Hamilton depression rating scale (HDRS), which is a likert scale. Because of the treatment, half developed depression and the other half did not. So i have two groups; depressed and non-depressed.
Analysis: I want to know whether the factor structure of baseline HDRS differs between the two groups. Thus, I performed multiple group confirmatory factor analysis (CFA).
I used a theoretically validated and relevant HDRS factor model as the basis of the MGCFA to compare goodness of fit measures and determine measurement invariances between the groups. However, my initial CFA indicated a poor model fit for both my groups. I am mindful that my sample sizes may be too small for this.
Questions
- Is multigroup CFA appropriate for comparing the measurement model of my two groups?
- What should I do given the poor model fit of the CFA in both groups?
- Is a total sample size of 180 too small for this multiple group CFA?
Best Answer
In general, multiple group CFA is a good tool for comparing the equivalence of a measurement model across two groups. However, you first need to show that the measurement model makes sense in at least one group.
First, analyse the entire sample
There are many different ways of tackling factor exploration. Here are some thoughts
Multiple group CFA
So, if, and only if, you can get a good model at the overall level would I proceed to multilple group CFA.