path coefficients

This forum is closed, and read-only.
Locked
wupanying
PLS Junior User
Posts: 2
Joined: Fri May 03, 2013 11:36 am
Real name and title:

path coefficients

Post by wupanying »

Dear all:
I would be very grateful if you tell me the difference between the path coefficients of smartpls and the standardized regression coefficients of spss. Are they the same?
User avatar
Diogenes
PLS Super-Expert
Posts: 899
Joined: Sat Oct 15, 2005 5:13 pm
Real name and title:
Location: São Paulo - BRAZIL
Contact:

Post by Diogenes »

Hi,
yes, they are the same.
If you want to test:
- run the one model in SmartPLS, for example: (x1, x2, x3 --> y)
- report / PLS / calculation results / Latent variable scores
- copy these scores to a spreadsheet / import them to SPSS
- Run the regression analysis (structural model).
- compare the beta values with path coefficients.

Best regards,
Bido
wupanying
PLS Junior User
Posts: 2
Joined: Fri May 03, 2013 11:36 am
Real name and title:

Post by wupanying »

Diogenes wrote:Hi,
yes, they are the same.
If you want to test:
- run the one model in SmartPLS, for example: (x1, x2, x3 --> y)
- report / PLS / calculation results / Latent variable scores
- copy these scores to a spreadsheet / import them to SPSS
- Run the regression analysis (structural model).
- compare the beta values with path coefficients.

Best regards,
Bido
Dear professor,
Why are they not equal when I test? Following is my test steps, there are five measured variables for each latent variable in my model. I take the mean of the five measured variables. Then I run the linear regression of the means and get the standardized coefficients. The standardized coefficients are not equal with the path coefficients. Is there any wrong with my steps?
User avatar
Diogenes
PLS Super-Expert
Posts: 899
Joined: Sat Oct 15, 2005 5:13 pm
Real name and title:
Location: São Paulo - BRAZIL
Contact:

Post by Diogenes »

Hi,

The LV scores from SmartPLS are not simple mean of the indicators, they are weighted by the outer weights).

Suggestion to compare regression in a simple way:
- use just one indicator by LV
- run the model in SmartPLS, for example: (x1, x2, x3 --> y)
- Run regression analysis (x1, x2, x3 --> y) in SPSS.
- compare the beta values with path coefficients.

Best regards,
Bido
bradley7680
PLS Expert User
Posts: 51
Joined: Tue Nov 04, 2014 5:27 am
Real name and title: Dr. Bradley Wilson. Senior Lecturer in Advertising. School of Media and Communication. RMIT University.

Re: path coefficients

Post by bradley7680 »

using latent variable scores/factor scores is common in research and especially in factor analysis and regression.

c.f., My PLS Handbook chapter which outlines the process of doing this.

Good luck
Brad
Bradley Wilson. Ph.D. Senior Lecturer in Advertising. School of Media and Communication. RMIT University. Melbourne, Australia.
SEE: http://www.researchgate.net/profile/Bradley_Wilson2 http://scholar.google.com.au/citations? ... AAAJ&hl=en
Locked