Warning X and Y variable in two level... PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Carina Wolf posted on Thursday, March 26, 2015 - 12:14 pm
I assume that on individual level intergroupcontact opportunity ("opport", single item) predicts intergroup friendship ("friend", three Items, is not modeled as latent factor).
On the context level we have in addition percentage of immigrants ("percent") in district ("distric").
Here I assume that percentage of immigrants predicts intergroupcontact opportunity ("opport"). And percentage of immigrants predicts intergroup friendship - both directly and indirectly via the intergroup contact opportunity in neighbourhood.
If I run the model, the following warning occurs:
*** WARNING in MODEL command
In the MODEL command, the following variable is a y-variable on the BETWEEN
level and an x-variable on the WITHIN level. This variable will be treated
as a y-variable on both levels: "opport"
What should I do now? Thanks a lot in advance!

USEVARIABLES ARE opport percent friend;
MISSING ARE ALL (777);
CLUSTER IS distric;
BETWEEN IS percent;

DEFINE: CENTER percent (GRANDMEAN);
ANALYSIS: TYPE IS TWOLEVEL;
MODEL: %WITHIN%
friend ON opport;
%BETWEEN%
friend ON opport percent;
opport ON percent;
MODEL INDIRECT:
friend IND percent;
 Bengt O. Muthen posted on Thursday, March 26, 2015 - 6:30 pm
Just ignore this; there is no problem.
 James Algina posted on Saturday, October 03, 2015 - 6:05 pm
Hello:

With the code at the end of this message, tech 9 reports the following for every rep

"THE MODEL ESTIMATION TERMINATED NORMALLY

Errors for replication with data file DataX.csv:"

Here X is the rep number.

The only warning produced when analyzing data in a specific rep is

In the MODEL command, the following variable is a y-variable on the BETWEEN
level and an x-variable on the WITHIN level. This variable will be treated
as a y-variable on both levels: COV

I am assuming this warning is the cause of the errors to which tech 9 refers. Is this correct?

Thanks,

Jamie

file is datalist.txt;
type = montecarlo;
Variable:
names = School Treat Cov Dv;
Usevariables = School Treat Cov Dv;
Cluster = School;
Between = Treat;
Define:
Center Cov (Grandmean);
Analysis:
Type = Twolevel;
Model:
%Within%
Dv on Cov*.25;
DV*.85;
Cov*.85;
%Between%
Dv on Cov*.35 Treat*.5;
DV*.15;
Cov on Treat*0;
[Cov*0];
Cov*.15;
 Linda K. Muthen posted on Saturday, October 03, 2015 - 7:51 pm
It sounds like no error is reported in TECH9. The message you show is informational only. It does not stop the analysis.
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: