2
$\begingroup$

I am doing a confirmatory factor analysis (CFA), and used the reliability() in R statement to get the average variance extracted (AVE). I also calculated the Mean Inter-Item-Correlation (MIC) manually from the item-correlations based on CFA output. The results are the same.

Are these the same indicators of reliability? I thought they are different?

$\endgroup$
1
$\begingroup$

To calculate AVE: sum up each squared factor loading, divide it by the number of indicators

The average inter-item correlation is simply the average or mean of all your items correlations

So they are different things.

I believe AVE is an indicator of construct validity as how much variance is explained by a given latent variable. if your latent variable is not explaining an accepted amount of variance in its indicators, then it is a poorly defined latent variable but MIC is to test the internal consistency reliability of your items and concerns the reliability of your latent variable

$\endgroup$
1
  • $\begingroup$ Thank you for your response. That's exactly what I thought. However, when I use the reliability() statement in R to get the AVE, the numbers are exactly the same as when manually calculating the MIC. This caught my attention and made me wonder whether the AVE in the reliability statement is not the AVE as you describe it. Particularly since it is a statement requesting reliability, and as you say AVE is a measure of validity. $\endgroup$ – Linda P Sep 4 '16 at 20:23

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.