finnigan posted on Wednesday, February 14, 2007 - 9:01 am
I'm using a levels of analysis framework and would like to know how I might model change in a collective construct such as shared mental models over time. I understand that growth modeling is used to model individual change,but not sure if there is a variant of growth models that address changes in collective constructs over time?
I'm afraid I don't understand what you mean by change in collective constructs.
finnigan posted on Thursday, February 15, 2007 - 1:27 am
I am working off a theory that an individual construct may operate at a group level. For example at an individual level a person's ability to manage their time may be influenced by a similar construct which operates at the group level. Typically in the organizational research, researchers use intra class correlations to decide on whether to aggregate the individual measuress on the construct at the individual level into a team or collective construct which is in this case a shared time mamangement practice within the team. I expect to use the shared or collectively held construct as a predictor variable of individual level constructs in a growth model,but I'd like to know if there is an approach in MPLUS to model a change in a collective constructs mean/variance over four measurement occasions, something similar to what happens in a growth model except that the construct of interest is a group and not an individual one. Thanks
We refer to this as multilevel modeling. A growth model can be specified on the between level for cluster (collective) units and the growth factors can be used as predictors of the between-part of individual-level variables. See Chapter 9 of the Mplus User's Guide which is available on the website.