|
|
1-1-1 multilevel mod. med.: unusual r... |
|
Message/Author |
|
Sven Lohrey posted on Friday, January 16, 2015 - 9:30 am
|
|
|
Dear Madam or Sir, I tried to run a 1-1-1 multilevel moderated mediation model trying to combine what Preacher, Zyphur, and Zhang (2010) call an "unconflated model" with a moderation of the a path. I tried to test moderation at mean, +1SD, and -1SD levels of the moderator Int1 with the below syntax. The model estimation etc. all terminated normally, however, my problem is that for all three levels of the moderator (mean, +/- 1SD) I received the exact same estimate and SE for the conditional indirect effect and that canīt be accurate so maybe I made a mistake at some point. It would be very nice if you could help me with this issue. Thank you very much! |
|
Sven Lohrey posted on Friday, January 16, 2015 - 9:32 am
|
|
|
The syntax I used is: USEVARIABLES ARE DayAut DayFlow DayExlag DayCP x1mean x2mean x3mean Int1; CLUSTER IS Person; CENTERING IS GROUPMEAN (DayFlow DayExlag DayAut); WITHIN ARE DayAut DayFlow DayExlag Int1; BETWEEN ARE x1mean x2mean x3mean; Missing are all (999); DEFINE: x1mean = Cluster_mean(DayFlow); x2mean = Cluster_mean(DayExlag); x3mean = Cluster_mean(DayAut); Int1 = DayExlag*DayAut; ANALYSIS: TYPE = TWOLEVEL; ESTIMATOR = ML; MODEL: %WITHIN% DayCP ON DayFlow(bw); DayCP ON DayExlag; DayFlow ON DayExlag(aw1); DayFlow ON DayAut(a2); DayFlow ON Int1(a3); [DayFlow@0]; DayExlag WITH DayAut Int1; DayAut WITH Int1; %BETWEEN% x1mean x2mean DayCP; x1mean ON x2mean(ab); DayCP ON x1mean(bb); DayCP ON x2mean; x2mean WITH x3mean; MODEL CONSTRAINT: NEW (indb indw0 indw1 indw2 Autmean Authigh Autlow); Autmean = 3.82; Authigh = 5.07; Autlow = 2.57; indw0 =(aw1+a3*Autmean)*bw; indw1 =(aw1+a3*Authigh)*bw; indw2 =(aw1+a3*Autlow)*bw; indb = ab*bb; |
|
|
Don't see an issue. Please send input, output, data, and license number to Support. |
|
Back to top |
|
|