How EXACTLY calculate Q2

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
Post Reply
leonieraible@hotmail.com
PLS Junior User
Posts: 2
Joined: Sun Sep 12, 2010 10:34 pm
Real name and title:

How EXACTLY calculate Q2

Post by leonieraible@hotmail.com »

Hi everyone, I have a question about how to calculate Q2. I think I know how to do it but I am not sure.

I calculated Q2 of the latent variable "A" this way:

- Blindfolding
- Select Omission Distance 7
- Select Construct "A"
- Construct Crossvalidated Redundancy (Total)
- 1-SSE/SSO of construct "A" is the Q2-value of construct "A"
- I ignored the other Q2-values of the other constructs, to calculate then every value by its own procedure. Therefore the other values are not relevant to me, right?

To calculate Q2 of construct "B", I would do the same procedure, blindfolding, selecting "B",.....looking only at the Q2-value of "B"

Is this right?

Thanks for answering!
christian.nitzl
PLS Expert User
Posts: 248
Joined: Sat Jul 25, 2009 1:34 pm
Real name and title:

Post by christian.nitzl »

Hey Leonie,

As far as I can see your procedure to calculate Q^2 is right for SmartPLS 2.0!

The only thing you have to pay attention is that the number of your cases should not be a multiple of your omission distance or in other words if you divide the number of your cases by your omission distance the result should not be a integer number. In general the omission distance could be a number between 5 to 10.

Best regards,

Christian
leonieraible@hotmail.com
PLS Junior User
Posts: 2
Joined: Sun Sep 12, 2010 10:34 pm
Real name and title:

Post by leonieraible@hotmail.com »

Thank you very much for your answer!!
safearch
PLS Junior User
Posts: 2
Joined: Tue Jun 28, 2011 8:13 am
Real name and title:

Q Square for second order Endogenous

Post by safearch »

Hi All,
I have a model with second order endogenous variable which is reflected by three other first order latent variables, the question is that I want to report Q square at the endogenous variable, so should I report the Q square at the second order endogenous variable or the other three reflective first order variables?

I hope that I am clear as I am still very new to PLS.
Your kind help is very much appreciated. thanking you in advance.
Post Reply