forum.smartpls.com

Hi
I need to do power analysis using the technique mentioned in this paper. For that, I need error terms. I have reflective and formative variables in my research model. How can I find error terms for these variables(formative)?

Aguirre-Urreta, M., & Rönkkö, M. (2015). Sample Size Determination and Statistical Power Analysis in PLS Using R: An Annotated Tutorial. Communications of the Association for Information Systems, 36, 3. doi:10.17705/1CAIS.03603

Thank you

For the approach you need the error variance. For reflective measures these can easily be calculated as 1-loading^2. For example, a loading of 0.7 has 1-0.7^2=0.51 error variance and a loading of 0.8 has 1-0.8^2=0.36 error variance (as also shown in their example).

Formative measures are assumed to be error free in PLS (i.e., composite indicators). Thus, the error variance of the construct should be zero.

Thank you Dr. Becker. This is very helpful.