Urgent help with SEM PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 olena mazurenko posted on Monday, June 11, 2012 - 3:42 pm
Dear Mplus users,
I urgently need your assistance.
I have data with complex sampling design. I'm trying to estimate the effect of two mediating variables (f4 and f5) on relationship between 3 IV (f1, f2, f3)and 2 DV (f6 and f7).
So far, my model looks like this:
TITLE: SEM
DATA: File is mplus2.dat ;
VARIABLE: NAMES ARE inc area spec mcare hmo unem pov pop crs con rbemlptnew rbphnptnew rbemldrnew rbphndrnew itcommnew itclinnew ithospnew emrusenew tmemlpt_new tmphnpt_new tmemldr_new tmphndr_new nothosp_new notoutp_new notintrp_new notrefs_new strata nsf weight;
CATEGORICAL ARE rbemlptnew rbphnptnew rbemldrnew rbphndrnew itcommnew itclinnew ithospnew emrusenew tmemlpt_new tmphnpt_new tmemldr_new tmphndr_new nothosp_new notoutp_new notintrp_new notrefs_new area crs
STRATIFICATION = strata ;
CLUSTER = nsf ;
WEIGHT = weight ;
ANALYSIS : TYPE = COMPLEX ;
MODEL: f1 by inc area spec mcare;
f2 by hmo unem pov pop ;
f3 by crs con;
f4 by rbemlptnew rbphnptnew rbemldrnew rbphndrnew;
f5 by itcommnew itclinnew ithospnew emrusenew ;
f6 by tmemlpt_new tmphnpt_new tmemldr_new tmphndr_new ;
f7 by nothosp_new notoutp_new notintrp_new notrefs_new ;
MODEL INDIRECT :
f6 IND f4 f5 f3 f2 f1 ;
f7 IND f4 f3 f2 f1;
f6 WITH f7 ;
f1 WITH f2:
f2 WITH f3 ;
f1 WITH f3 ;
Can someone tell me whether I'm on the right track!!!
 Linda K. Muthen posted on Monday, June 11, 2012 - 5:04 pm
This will not work. You do not have any ON statements in the MODEL command related to the indirect effects you specified in MODEL INDIRECT. Also, WITH statements are not allowed in MODEL INDIRECT. Please see MODEL INDIRECT in the user's guide to see the relationship between the MODEL command and the MODEL INDIRECT command.
 olena mazurenko posted on Monday, June 11, 2012 - 7:18 pm
Dear Linda,
Thank you for prompt response!!!!!
Should I write:
MODEL: f1 by inc area spec mcare;
f2 by hmo unem pov pop ;
f3 by crs con;
f4 by rbemlptnew rbphnptnew rbemldrnew rbphndrnew;
f5 by itcommnew itclinnew ithospnew emrusenew ;
f6 by tmemlpt_new tmphnpt_new tmemldr_new tmphndr_new ;
f7 by nothosp_new notoutp_new notintrp_new notrefs_new ;
f6 ON f5 f4 ; (regress DV1 on Mv1 MV2)
f7 On f5 f4 ; (regress DV2 on MV1 MV2)
f5 ON f1 f2 f3; (regress MV1 on IV1 IV2 IV3)
f4 On f1 f2 f3; (regress MV2 on IV1 IV2 IV3)
f6 WITH f7 ;
f1 WITH f2:
f2 WITH f3 ;
f1 WITH f3 ;
MODEL INDIRECT :
f6 IND f4 f5 f3 f2 f1 ;
f7 IND f4 f3 f2 f1;
 Linda K. Muthen posted on Monday, June 11, 2012 - 7:57 pm
I can't see any obvious problems. Give it a try. If you have problems, send the output and your license number to support@statmodel.com.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: