Team diversity measure as time invariant predictor of latent growth curve

Question

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).

Answer 1

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.