Message/Author |
|
|
Hello, I need to make a moderation analysis in Mplus within the actor partner independence model but have no clue how to include a moderator variable (humor frequency) in the regression of SAM (stress level) ON Pos_m/Pos_w (Positive Humor). Find the current syntax below. Can anybody help? DEFINE: posm = CLUSTER_MEAN(Pos_m); posw = CLUSTER_MEAN(Pos_w); CENTER day Pos_m Pos_w (GROUPMEAN); CENTER posm posw (GRANDMEAN); VARIABLE: NAMES ARE cluster day Pos_m Pos_w SAM_f_m SAM_f_w SAM_a_m SAM_a_w; MISSING = ALL(-99); USEVARIABLES = cluster day Pos_m Pos_w SAM_a_m SAM_a_w posm posw; BETWEEN = posm posw; WITHIN = day Pos_m Pos_w; CLUSTER = cluster; ANALYSIS: TYPE = twolevel random; MODEL: %WITHIN% ind_m | SAM_a_m ON Pos_m; ind_w | SAM_a_w ON Pos_w; cross_mw | SAM_a_w ON Pos_m; cross_wm | SAM_a_m ON POs_w; SAM_a_m SAM_a_w ON day; SAM_a_m WITH SAM_a_w; %BETWEEN% SAM_a_m ON posm posw; SAM_a_w ON posm posw; SAM_a_m WITH SAM_a_w ind_m ind_w cross_mw cross_wm; SAM_a_w WITH ind_m ind_w cross_mw cross_wm; ind_m WITH ind_w cross_mw cross_wm; ind_w WITH cross_mw cross_wm; cross_mw WITH cross_wm; OUTPUT: sampstat CINTERVAL; |
|
|
For moderation - in the form of a cross-level interaction - you would have to regress your random slopes on level-2 covariates, not just the random intercepts as you do now. |
|
|
Thank you for your reply!! Could you show me this within the Syntax? Or is there any example given in form of a syntax like I have found for the basic APIM? |
|
|
See the section on Cross-level Interaction on our Mediation page: http://www.statmodel.com/Mediation.shtml |
|
Back to top |