There is an issue using a within level variable as a grouping variable because the groups are not independent. I think a better approach for dyad data is a multivariate approach where you use a wide format for the data. This is more flexible than the long approach. See the following paper which is available on the website:
James L. Peugh, David DiLillo & Jillian Panuzio (2013) Analyzing Mixed-Dyadic Data Using Structural Equation Models Structural Equation Modeling. Pages: 314-337 DOI: 10.1080/10705511.2013.769395
E.I. Montoya posted on Thursday, September 19, 2013 - 5:41 pm
Actually in this case the grouping variable is a between-level variable. Both members of the dyad are the same gender.