Transformed (logarithmized variables)

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
Post Reply
Jerry007
PLS Junior User
Posts: 2
Joined: Sat Aug 05, 2017 1:31 pm
Real name and title: Jeremias Meyer

Transformed (logarithmized variables)

Post by Jerry007 »

Hi,

my skewness at some variables (items) were pretty bad ( above 1 or -1) so I logarithmized the specific ones.
Now I´m modelling and have 4 factors having around 4-6 items each.
Some of the item were logarithmized, some were not.

My first question: could that be a problem using logarithmized and NOT logarithmized items at the same time?

My second question: Do I have do anything else? Cause the Min and Max values of a construct with logarithmized and not logarithmized values differ obviously now...

Kind regards & Thanks

Jerry
jmbecker
SmartPLS Developer
Posts: 1284
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Transformed (logarithmized variables)

Post by jmbecker »

I would not mix log and non-log variables in one construct. It will be hard (impossible) to interpret.
Either log all or none of the indicators for one constructs.

Different constructs can of course have log and non-log indicators, but not mix them in the same construct.
Dr. Jan-Michael Becker, BI Norwegian Business School, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
Jerry007
PLS Junior User
Posts: 2
Joined: Sat Aug 05, 2017 1:31 pm
Real name and title: Jeremias Meyer

Re: Transformed (logarithmized variables)

Post by Jerry007 »

Thanks!
Post Reply