I'm a beginner when it comes to CFA (in AMOS) and was wondering if anyone could advise me.

I've questionnaire data from two separate samples and I want to confirm the factor structure. I've run an EFA on the first sample (N=395, 36 items, 4 factors) and am trying to confirm that structure in the second sample (N=253). The fit indices are generally poor, despite reasonable factor loadings (>.3, majority >.6): Cmin 1280.25, df=588 CFI =.895 TLI = .888 RMSEA = .055

What are the next steps I should do to investigate the source of the poor fit. I've looked online but can't find a guide that I can follow for my level of knowledge.

| cite | improve this question | | | | |

I would look into differences between the samples. Since you did not tell us anything about what variables you have or who is in the samples, it's hard to be more specific.

But, just hypothetically, suppose your samples are human beings. Do your two samples differ on age, sex, income? Do they differ on other things that are relevant?

If your factor analysis is about, for instance, something psychological, are the samples from populations that have different sorts of psychological issues?

You might then combine the samples and do EFA including the variables you think are accounting for the differences.

| cite | improve this answer | | | | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.