calculation of unstandardized weight

[Max]

I would like to calculate the unstandardized weight (like a regression coefficient) between two latent wariables.

I take it that since PLS operates on the standardiced LV, the path coefficient (weight) should be interpreted as: a change of X standard deviation in LV1 is going to produce a change in LV2s standarddeviation of WEIGHT1-2 * X

If this is the case, how do I calculate a weight, that could be interpreted like a regression coefficient: a change of X in LV1 wil lead to a change in WEIGHT1-2 * X in LV2

[viswadatta]

The path coefficients are the inner weights, which are to be used. The standardised weights are called inner loadings, which mention the degree of correlation between LVs.

[Max]

In other words, you are confirming that the inner weights (path coeficients) are standardized ?

[AAK]

Yes, I also have difficulities to convert the path coeff in PLS into UNstandardized. There are some posts mentioning the converting into UNstandardized LV score, but what if I only need to change the path coeff? Anyway to do that?? Many thanks!~