New to MLM - Basic help request PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 James Hamilton posted on Wednesday, May 07, 2014 - 2:41 pm
I am analyzing text from 175 blog posts obtained 16 different medical patients; 8 with one disorder and 8 with another (case code). Some of the blogs for each case are written by the patient and some about the patients by friend or relative (self code) I want to test the main effect of caseness on text parameters (e.g. fm_negemo), as well as the interaction of caseness and self-other on text parameters. Can someone tell me if I have set this up right in the code below.

Thanks.



USEVARIABLES = Subject casecode selfcode fm_negemo casexself;

CLUSTER = subject;
BETWEEN = casecode;
WITHIN = selfcode casexself;

Define: casexself = casecode * selfcode;

ANALYSIS: TYPE = TWOLEVEL;

Model: %WITHIN%
fm_negemo ON selfcode casexself;

%BETWEEN%
fm_negemo ON casecode ;
 Linda K. Muthen posted on Thursday, May 08, 2014 - 2:13 pm
You have too few cluster units to do multilevel modeling. Some say you need a minimum of 20. Others say a minimum of 30-50.

You might want to ask on MultilevelNet how you might handle your data.
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: