Page 1 of 1

Constant term

Posted: Tue Jun 03, 2008 9:27 am
by puck16
Dear all,

Is it possible to include a constant term/intercept for each PLS regression equation in the structural model? If so, can anyone tell me how?

Many thanks,
Sandra

Posted: Tue Jun 03, 2008 6:51 pm
by Diogenes
Hi Sandra,
SmartPLS works with standardized variables,
in this way the constants will be zero.

One idea is to copy/paste the unstandardized LV scores to Excel or SPSS and compute the regression there. The standardized results (betas) will be a little different from SmartPLS.

Best regards.
Bido

Posted: Thu Jun 05, 2008 3:55 pm
by puck16
Dear Bido,
Thanks for the useful information.
Sandra

Re: Constant term

Posted: Fri Sep 07, 2018 8:31 pm
by randylaw
Hello,

It is understood that there is no constant in the standardized solution, but I would like to obtain the constant term to use with the raw regression weights that come from the IPMA procedure to create the full regression equation. How would I do that?

Thanks!

Randy

Re: Constant term

Posted: Sat Sep 08, 2018 11:47 am
by jmbecker
You should copy the undstandardized latent variable scores from the IPMA (IPMA > Latent Variables > Unstandardized) into SPSS, Excel or some other software that can run normal regression analysis.
You should get the same results for the path coefficients as in (IPMA > Path Coefficients > Unstandardized), but in addition you will also get the intercept.