The error says your variance and covariance values are probably wrong somewhere or your model maybe is incorrect. Check out these things and rerun. Usually this will fix the problem Best wishes, D.Booth.

Mugaahed Abdu Kaid Saleh The determinant of a matrix indicates the generalized variance of a set of variables in the matrix. It is used in finding the inverse of a matrix, which is then used to compute eigenvalues. In order to to do this, the determinant must be positive (greater than 0). If the determinant is 0 or negative, there are no eigenvalues and no solution is possible. *This is what R means by the matrix is "not positive definite".

Checking the determinant of matrix (to confirm that it's positive) can be done simply with the det() function. Let's say my variance-covariance matrix is called "mycov" in R.

det(mycov) will compute the determinant. If the determinant is 0 or less, it's possible the data were entered incorrectly somewhere (check for negative values) or the matrix is not a square (i.e., equal number of rows and columns). A square matrix is required to compute the inverse of the matrix.

Mugaahed, just another thought: In order to trace the problem I would suggest that you analyze small models first, i.e., the measurement model of communication, than the measurement model of networking, and so on. Hopefully, you will find the source of your problem then.

In lavaan, by default, the factor loading of the first indicator of a latent variable is fixed to 1 in order to set the scale of the latent variable. Sometimes, the first indicator only has a small factor loading, so that the latent variance of the factor may be non-significant. In this case it would be better to chose a different indicator as the scaling variable.

HTH, Karin