How to model variables if they have a bidirectional relationship

[Chudamani]

In my study, I am trying to perform multi-level (three levels) longitudinal mixed-modeling on panel data with 6-time points. There is a bidirectional relationship between my predictor (education level) and the continuous outcome variable (cognitive score). I am assessing if education is associated with cognitive trajectories over time. How should I model my variables?  

[sbxkoenk · Accepted solution]

Hello,

I have moved your question to the "statistical procedures" board.

When I started doing multi-level (2+ levels) longitudinal mixed-modeling, I started here :

Using SAS PROC MIXED to Fit Multilevel Models, Hierarchical Models, and Individual Growth Models
Judith D. Singer
Journal of Educational and Behavioral Statistics
Vol. 23, No. 4 (Winter, 1998), pp. 323-355 (33 pages)

Good luck,

Koen

[Chudamani]

Thank you for these resources.