Kurt posted on Saturday, October 12, 2013 - 9:29 pm
I am trying to run the model as below (from Preacher and colleague's mplus multilevel mediation command) and would like to also take into account that my data is clustered within level 3 clusters. While using Type=twolevel random is there anyway that I can take this account? When I used dummy variables for level 3 cluster, it somehow fails to calculates the standard error showing " WARNING: THE SAMPLE COVARIANCE OF THE INDEPENDENT VARIABLES IS SINGULAR. PROBLEM INVOLVING VARIABLE DUMMY_TE." The macro is about testing mediation among level1 IV, level2 Mediator and level2 DV.
TITLE: 1-2-2 mediation (MSEM) DATA: FILE IS mydata.dat; VARIABLE: NAMES ARE id x m y; USEVARIABLES ARE id x m y; CLUSTER IS id; ! Level-2 grouping identifier BETWEEN ARE m y; ANALYSIS: TYPE IS TWOLEVEL RANDOM; MODEL: %WITHIN% x; %BETWEEN% m y; m ON x(a); ! regress m on x, call the slope "a" y ON m(b); ! regress y on m, call the slope "b" y ON x; ! regress y on x MODEL CONSTRAINT: NEW(indb); ! name the indirect effect indb=a*b; ! compute the Between indirect effect OUTPUT: TECH1 TECH8 CINTERVAL;