|
|
Assessing convergence of measures ove... |
|
Message/Author |
|
|
I am trying to assess the convergence between 2 measures of fidelity in a treatment study (clinician self-rated vs. expert). In other words, does how clinicians rate themselves match how experts rate clinicians' fidelity each each of 12 treatment sessions. I am trying to discern the best approach for this. Would it be better to calculate the growth curve of, say, expert rating, and enter self rating as a time varying predictor? If so, would the correct syntax be the same as what i would use to enter a time varying covariate (e.g., example 6.10) Or would it be better to conduct a parallel process model (ex. 6.13) to examine the association between self and expert at each assessment point. Part of my question is that I'm confused by the idea of regressing slope of process 1 on the intercept of process 2, and not sure if this answers the question of whether scores "converge" over time. thank you for any advice! |
|
|
This is a good question for SEMNET. |
|
Back to top |
|
|