1
$\begingroup$

I am specifying a latent growth curve model in MPLUS with a time-invariant predictor of the growh curve.

The time-invariant predictor is a team diversity measure (standard deviation) which is based on a 7 item survey scale. Each team has 5 team members.

My question is whether the correct approach would be to calculate the standard deviation on the basis of each individual's average score on the 7 items and include this in the model as an observed variable, or calculate the standard deviation per item across the 5 members, and include the SD per item as indicators of a latent variable (assuming that scale reliabilities and Rwg/ICC - within group agreement and intraclass correlation - are all satisfactory).

$\endgroup$
1
$\begingroup$

If the 7-item scale measures some meaningful construct, then I would recommend the first approach. If each item itself is informative or relevant to your research question, then I would go with the latter.

$\endgroup$

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.