Robust standard errors in PROCESS module
I am wondering what type of robust standard errors are used in the PROCESS module.
In the Hayes PROCESS modules for R/SPSS etc., one can choose between four different heteroscedasticity-consistent (HC) standard errors. I've tried to find out which one is used in JASP, but I couldn't find more detail than that it's using a sandwich-type covariance estimator. I am not sure how that maps onto the four HC types.
Therefore, I'd like to ask for more details on the type of estimator used. Thanks!
[JASP version 0.98.1, Process module version 0.95.4-release.803]
Comments
I've analysed the same test data set using lavaan and I can reproduce the same robust standard errors as in JASP using the following specifications for standard error estimation:
se = "robust.huber.white", information_meat_hc = "HC0"
Hence, it looks like robust standard errors in the process module are of type HC0.
[sorry for replying to my own post, just thought that this way someone else who is googling this might find the answer]
that's very useful @ulrike_senftleben thanks a lot for doing this, and props to you for being resourceful and finding a way to figure out the answer
Would be good to see this added to the help files though...