What are the cutoff values for sample and residual correlations and correlations nt indicators and components?

[Miguel Mayol]

First thanks for all

Second sorry for my lack of knowledge, I started with SEM for my thesis just half a year ago, and I am learning by doing (writing searching and reading) and used trial PLS-SEM plus some favors using the university's pro one until I found this free software

I was not able to find the cutoff or recommended values for the values indicated in these tables.

A) Sample correlations (lower diagonal) & Residual correlations (upper diagonal)

B) Correlations between indicators and components

I supposed it is <1, but I am not sure, and that is why I ask.

Thanks in advance for answering this question, and if it is not a lot to ask with some references.   

[heungsun.hwang]

As you know, all correlations are between 1 and -1. Typically r = .1, .3, and .5 indicate small, medium, and large linear associations, respectively. In PCA or factor analysis, a standardized loading (correlation) for each indicator smaller than |.3|  may be paid less attention for interpretation, although this can be situation-dependent. The correlation residuals are between -2 and 2 because they are the differences of two correlation matrices (sample and model-implied). A correlation residual close to these limits can be indicative of a seriously bad fit (Bollen, 1989, p. 258), whereas a correlation residual close to zero may be indicative of an acceptable fit. But, I am not sure whether there are really cutoff values for the correlation ressiduals. 

Best,

Heungsun