forum.smartpls.com

Hello Everyone,

I'd like to apologise if my question is inappropriate. I am just starting out with the use of smart PLS3 for my statistical analysis and still am figuring out bits and pieces .

I am currently trying to evaluate the influence of 7 Independent variables on one 1 dependent variable. Each of these variables have 3 measurement items and all are measured on a 7 point Likert Scale. I've been reading the J.Hair Book on SEM-PLS. He states that all data needs to be (z) standardised before running the software. his example states that there is an option to select Data Metric ( Mean 0, Var1) which would help convert the data to the required standardised format. I've been trying to use the SMARTPLS3 software, however The option to chose data metric isn't available. My query is as follows

1) Is the interface different in PLS2 as compared to 3 for running the calculations? as the screenshot provided in the book varies from the interface of this software.
2) Does the software automatically convert the 7 point likert scores to standardised values?
3) If it does not automatically standardise the raw data, how can i do it on the PLS3 software?

Would really appreciate any help in this regard and really sorry if my question is repetitive or silly. I am an absolute novice.

Many Thanks

Sussmek

Hi Sussmek,

Yes, the interface is a little bit different from Version 2. I heard that they are actually updating the book to the new version. However, mostly differences are not large.

With regard to your question: In SmartPLS 3 the data is standardized by default (i.e., the mean 0, var1 option in SmartPLS 2). There is no need to do any standardization prior to importing the data to SmartPLS3

Many Thanks Dr Becker ! Your reply is much appreciated.

Dear Dr. Becker,

what If I have already standardized the data, imported them into SmartPLS, and made the whole analysis?

Is it ok? I.e. is it a rule that there is no need for standardization prior to importing the data to SmartPLS3 or is it a choice?

Thank you for the question.

Kind regards,
Antonija

You do not need to standardize the data prior to importing it. SmartPLS will always standardize the data in the algorithm. In fact, it is usually better to import the unstandardized raw data, because standardization might be different if you analyse groups, have missing values, etc. In these cases importing standardized data could lead to wrong results.

Dear Dr. Becker,
thank you very much for the quick answer. I corrected the database and everything is ok.
Kind regards,
Antonija