Phil Rodkin posted on Thursday, May 19, 2011 - 12:50 pm
We have had difficulties in trying to run a CFA with our data using a correlation matrix.
We received the following WARNING:
************** WARNING: THE RESIDUAL COVARIANCE MATRIX (THETA) IS NOT POSITIVE DEFINITE. THIS COULD INDICATE A NEGATIVE VARIANCE/RESIDUAL VARIANCE FOR AN OBSERVED VARIABLE, A CORRELATION GREATER OR EQUAL TO ONE BETWEEN TWO OBSERVED VARIABLES, OR A LINEAR DEPENDENCY AMONG MORE THAN TWO OBSERVED VARIABLES. CHECK THE RESULTS SECTION FOR MORE INFORMATION. PROBLEM INVOLVING VARIABLE ATAGGB. ***************
We thought the error was due to a mistake in our correlation matrix, such as a mistyped number above 1.0 or below -1.0. This was not the case. We do not think there is a problem with the correlation matrix itself. When we run the same model but with a slightly different correlation matrix (i.e, with 3 sig. figures instead of 6), the model runs just fine. However, the latter correlation matrix (i.e., with 3 sig. fig.) fails to run when we introduce paths models.
We would like to send you our correlation matrix and MPLUS input file so that you might advise us. To whom should we send these files?