2 Parallel mediation anlysis problem
Message/Author
 Lee Yong Seok posted on Thursday, March 10, 2016 - 2:16 am
Hi, I'd like to get estimates in 2 parallel mediation analysis. Especially, estimation via 2-mediator separately and difference between these two mediation effects. What a question i really want to ask is that any other problems exist in my programming code in Mplus 7.
The command is presented below. the code ranging from TITLE to ANALYSIS were well specified as had been confirmed by CFA model.

MODEL:
X by Ps_5 Ps_9_r Ps_11 Ps_13
Ps_18 Ps_21_r Ps_25 Ps_30_r
Ps_31 Ps_33 Ps_35 Ps_37_r
Ps_39 Ps_41 Ps_44_r;
M1 by IE_1 IE_2 IE_3 IE_4
IE_5 IE_6;
M2 by EE_7 EE_8 EE_9 EE_10
EE_11 EE_12 EE_13 EE_14
EE_15 EE_16;
Y by r_1-r_10;
[Y] (b0);
Y ON M1 (b1);
Y ON M2 (b2);
Y ON X (cdash);
M1 ON X (a1);
M2 ON X (a2);

MODEL CONSTRAINT:
NEW(a1b1 a2b2 TOTALIND TOTAL DIND);
a1b1 = a1*b1;
a2b2 = a2*b2;
TOTALIND = a1*b1 + a2*b2;
TOTAL = a1*b1 + a2*b2 + cdash;
DIND=a1*b1-a2*b2;

OUTPUT: STAND CINT(bcbootstrap);

The result from this code was embarrassing me because unreasonable value has been reported. low and upper 5% estimates in bootstrapping CI have all the same value and the two-tailed p-value of all coefficents has 0.9 ~ 1.0,for instance. any problems in my command?
 Linda K. Muthen posted on Thursday, March 10, 2016 - 7:17 am
Post:
This is a private posting area. Only registered users and moderators may post messages here.