Intercept-only model, bivariate corre... PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Felice K. posted on Friday, June 30, 2017 - 1:52 am
Dear Dr.es Muthen,

I want to run a twolevel analysis without predictors on level 1 or 2, just considering the nested structure of the data. The objective is to get the correlation coefficients of 2 dependent variables (CFT-SUM and MC_SUM), the intraclass correlation and the estimated sample statistics for "within" and "between". This is my syntax:

data: file = cftmc.dat;

variable: NAMES = CFT_SUM klasse MC_Sum;
MISSING = all(-99);


USEVARIABLES = MC_Sum CFT_SUM;


CLUSTER = klasse;

Analysis: TYPE = TWOLEVEL;

Model: %WITHIN%

%BETWEEN%

output: sampstat;

My fit indices are bad (RMSEA .404, CFI .004, TFL .004), is this due to a syntax error? I get the warnings: "variable is uncorrelated with all other variables CFT-SUM" and "variable is uncorrelated with all other variables MC-SUM". As I have no predictors an level 1 and 2, this is what I intended, isn't it?

Thank you for your help!
 Bengt O. Muthen posted on Friday, June 30, 2017 - 5:06 pm
The variables are probably correlated in the data so use WITH on both levels.
 Felice K. posted on Saturday, July 01, 2017 - 4:02 am
Thanks a lot for pointing that out!
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: