SmartPLS 2 Data analysis and conclusion

This forum is closed, and read-only.
Locked
azmiabdaziz
PLS Junior User
Posts: 1
Joined: Tue Jun 18, 2019 2:24 am
Real name and title: Mr Azmi Abd Aziz

SmartPLS 2 Data analysis and conclusion

Post by azmiabdaziz »

Dear Sir

I have run my data via SmartPLS 2 and needs some clarification from you.
There are 8 constructs independent (variables) with items and 1 constructs dependent (variables) with items as follows:
Independent
Construct 1 (5 items) AVE 0.8888, R Square = 0
Construct 2 (6 items) AVE 0.4963, R Square = 0
Construct 3 (5 items) AVE 0.7888, R Square = 0
Construct 4 (5 items) AVE 0.7989, R Square = 0
Construct 5 (5 items) AVE 0.7816, R Square = 0
Construct 6 (5 items) AVE 0.7666, R Square = 0
Construct 7 (6 items) AVE 0.6677, R Square = 0
Construct 8 (5 items) AVE 0.8080, R Square = 0

Dependent
Construct 9 (4 items) AVE 0.0000, R Square = 0.7099

1. The program does not provide Goodness of the Fit (GOF)? How to calculate it manually. Please assist me. I will share my data with you later.

2. I have to present my data with consideration (based on SEM ) to add a few more results such as

a. GFI (Good Fit Index) should ve > 0.90 (Note: is this GFI same function like GoF?
b. CFI should be > 0.90
c. RMSEA should [should be 0.05 - 0.08]
d. p [should be 0.01 - 0.05]

Is there any data from SmartPLS that can be used or comparable with or we need to calculate the value for a, b, c and d? Really need to know on this situation.

3. For the Pearson correlation, "r" value can be 0, +ve 1 value or -ve 1 value. From my understanding, the value just shown the direction of data but not the strength of the association. When we got -ve value, what is the best assumption we can made. I got the result as follows:

a. Result under Total effect:
Sample mean (M) = - 0.014750 , still under 0 value and below -ve 1. Can we say that the construct is with good correlation with other variables however the .......I need your help.

b. Other result all shown +ve value under 0 value and below +ve 1. Can we confirm that these variables have good association.

Thank you. Hope to get your response so I can continue my results preparation.
Best regards
Azmi
azmi03aa@gmail.com






Thank you very much.
jmbecker
SmartPLS Developer
Posts: 1282
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: SmartPLS 2 Data analysis and conclusion

Post by jmbecker »

Some of the results that you need are available in SmartPLS 3, which should be the preferred choice anyway.

Some of the results that you request are not available in PLS models. That has nothing to do with the software, but with the underlying algorithm and its conceptual properties. Chi-Square based Goodness of the Fit like CFI or GFI are not meaningful in PLS.
Please also read about the differences between CB-SEM and PLS.
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
Locked