Can i modify the results of my moderating effect

Frequently asked questions about PLS path modeling.
Post Reply
amira-mehdi
PLS Junior User
Posts: 1
Joined: Mon Jan 22, 2018 10:53 am
Real name and title: amira

Can i modify the results of my moderating effect

Post by amira-mehdi » Mon Jan 22, 2018 11:14 am

m.PNG
m.PNG (20.24 KiB) Viewed 644 times
moderating.PNG
moderating.PNG (30.46 KiB) Viewed 644 times
Hello, in my model i have the social media use as a moderator variable that amplifies the positive impact of social CRM on firm performance. My proble is that moderating effect when testing it with the smartpls3 was not significant (p-value= 0.549) + the direct relation between Social CRM and firm performance was also not significant (p-value= 0.664)which is total illegical !
My question is how can i do to make my results better and to prove that the moderator+ the direct impact both have a significant impact ??

User avatar
cringle
SmartPLS Developer
Posts: 796
Joined: Tue Sep 20, 2005 9:13 am
Real name and title: Prof. Dr. Christian M. Ringle
Location: Hamburg (Germany)
Contact:

Re: Can i modify the results of my moderating effect

Post by cringle » Tue Jan 23, 2018 11:30 am

We strongly suggest not to "torture" methods until it confesses what you expect.

In case of your situation, the coefficients are so small that no variations would lead to a different solution. The problem may be with your data. Please check them again (or collect more data). But if the data is good, take the results as they are: There may be no relevant moderating effect!

Sorry, there is (as always) not a straightforward solution when estimations do not meet expectation (what is the usual case in statistical analyses).

Best regards
Christian Ringle

Post Reply