Interpreting intercepts in MLM & FE m... PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Elizabeth Shulman posted on Saturday, June 08, 2013 - 8:23 pm
I am comparing results for two similar models, one run as a MLM, one as a Fixed Effects (FE). Observations are clustered within individuals. The intercepts are really different in the two models even though age is centered at 14 in both. In the FE model, AGE is not strongly related to the DVs in the between level.

Could you please explain the difference in the interpretation of the intercepts? Thanks in advance!

Here are the syntaxes and the estimated intercepts:

!!!!!!!!!! MLM MODEL

WITHIN = Age;
BETWEEN = ;

DEFINE:
DV1 = DV1*100;
DV2 = DV2*100;
Age = Age-14; !intercept at 14

ANALYSIS:
TYPE=TWOLEVEL; COVERAGE=0; ESTIMATOR=MLR;

MODEL:

%WITHIN%
DV1 ON Age ;
DV2 ON Age ;
Age ;


%BETWEEN%

DV1 with DV2 ;

! Means
! DV1 64.019 1.331
! DV2 63.483 1.335

----------------------------------------

!!!!!!!!!! FIXED EFFECTS MODEL

WITHIN = ;
BETWEEN = ;

DEFINE:
DV1 = DV1*100;
DV2 = DV2*100;
Age = Age-14; !intercept at 14

ANALYSIS:
TYPE=TWOLEVEL; COVERAGE=0; ESTIMATOR=MLR;

MODEL:

%WITHIN%
DV1 ON Age ;
DV2 ON Age ;
Age ;


%BETWEEN%
DV1 ON Age ;
DV2 ON Age ;
Age ;

DV1 with DV2 ;

! Intercepts
! DV1 41.839 5.498
! DV2 43.717 6.111
 Linda K. Muthen posted on Saturday, June 08, 2013 - 9:22 pm
Please send the two outputs 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: