Level 2 predictors on level 1 outcome PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Steven Howard posted on Thursday, March 02, 2017 - 10:30 pm
I am trying to run multi-level regression analyses, with children nested within preschool rooms. My primary question is the extent to which a level 2 variable (measured at the room level) predicts a subsequent level 1 outcome (measured at the child level). There are also important covariates that I need to control for at both the room level and child level.

I have tried to put together this model – see script below, which I have simplified for the purpose of illustration – but am getting a number of error messages. Are you able to advise what I am doing wrong?

Data:
File = DATASET.csv;
Variable:
Names = Room ! Room is cluster variable
RoomCov1 ! Room-level covariate (categ.)
RoomCov2 ! Room-level covariate
ChldCov1 ! Child-level covariate (categ.)
ChldCov2 ! Child-level covariate
ChildY ! Central child-level outcome
RoomX1 ! Central room-level predictor
RoomX2; ! Central room-level predictor
Usevariables = Room RoomCov1 RoomCov2 ChldCov1 ChldCov2 ChildY RoomX1 RoomX2;
Missing are all (-999);
Cluster = Room;
Analysis:
Type = twolevel;
Estimator = ML;
Model:
%Between%
ChildY ON RoomCov1 RoomCov2 RoomX1 RoomX2;
%Within%
ChildY ON ChldCov1 ChldCov2;
Output:
Tech1 standardized sampstat;
 Bengt O. Muthen posted on Friday, March 03, 2017 - 5:05 pm
You have to put room-level covariates on the Between= list. See UG examples such as 9.1
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: