Message/Author |
|
|
Hi, I have repeated measures data (patients’ reports of symptoms at each therapy session), and I’m trying to model the within-person effect of the IV (amount of a specific therapeutic technique during each session) on next-session symptom change. I’m new to Mplus, so I’m not sure how to account for the lagged auto-regressive effect of the DV (symptoms at one session affecting symptoms at the next session). I only have data for the IV for 3 sessions. I’ve been looking at Example 9.2 (two-level regression analysis) of the user guide as a template, but I’m not sure if I am adapting it correctly. I would like to run a main effect model, as well as moderation with a between-person variable. Here is the code I've tried (_p indicates prior session): USEVARIABLES ARE Patno Session S_bet_p S_win_p BDI_p BDI; missing = .; WITHIN = S_win_p; BETWEEN = S_bet_p; CLUSTER = patno; DEFINE: CENTER S_win_p (GROUPMEAN); ANALYSIS: TYPE = TWOLEVEL RANDOM; MODEL: %WITHIN% s | BDI ON S_win_p; %BETWEEN% BDI s ON S_bet_p; BDI WITH s; Any help would be greatly appreciated. Thanks! |
|
|
What's your number of time points for the symptom outcome? |
|
|
Thank you for your quick response! I have 5 time points for the outcome (sessions 1-5),and IV data for sessions 2-4. |
|
|
Also, this is the code I have tried with a moderator (TRS): Within = S_win_p ; Between = TRS ; Cluster = patno ; DEFINE: Center x (groupmean); ANALYSIS: TYPE = TWOLEVEL RANDOM ; MODEL: %WITHIN% s | BDI ON S_win_p; BDI ON BDI_p ; %BETWEEN% BDI ON TRS; s ON TRS; BDI WITH s; Thanks again! |
|
|
Looks fine but X is not in the model. |
|
Back to top |