Disjointed Two Stage Approach

This forum is closed, and read-only.
Locked
chenchenchen
PLS Junior User
Posts: 5
Joined: Mon Mar 01, 2021 1:56 am
Real name and title: chentianzhi student

Disjointed Two Stage Approach

Post by chenchenchen »

Dear Smart PLS Community,

I hope you are doing well.
I have a problem with the disjointed two stage approach of my model.
Some paths are not significant in the first stage.Do I need to adjust them? Or do I just need to focus on getting latent variable score in the first stage and focus on path significance in the second stage?

As I really need the program for my thesis, I would be very glad if you could help me.

Thank you in advance.

Best regards
jmbecker
SmartPLS Developer
Posts: 1281
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Disjointed Two Stage Approach

Post by jmbecker »

The first stage is only for obtaining the latent variable scores for the second stage. You generally do not need to worry about significance. However, if a construct has no significant paths to other constructs its estimated latent variable score might be very unreliable. Hence, it is usually good to check if there is at least one significant path per construct in the model.
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
chenchenchen
PLS Junior User
Posts: 5
Joined: Mon Mar 01, 2021 1:56 am
Real name and title: chentianzhi student

Re: Disjointed Two Stage Approach

Post by chenchenchen »

Dear Dr. Jan-Michael Becker,

Thank you very much for your valuable information and support. I really appreciated. But I still have the following question and hope to get your help.

As shown in the picture below, which is the first stage path model, the black lines represent significant paths, while the red lines represent non-significant paths. The variable “GAIN”has only one path point to “SATI”and it is non-significant , but “GAIN”is an important variable in my study and I need its latent variable score, as you said “if a construct has no significant paths to other constructs its estimated latent variable score might be very unreliable”, then what should I do with this situation?

Looking forward to your reply.

Thank you!
Attachments
2.PNG
2.PNG (131.68 KiB) Viewed 18963 times
chenchenchen
PLS Junior User
Posts: 5
Joined: Mon Mar 01, 2021 1:56 am
Real name and title: chentianzhi student

Re: Disjointed Two Stage Approach

Post by chenchenchen »

I would also like to ask whether the calculation of sample size is based on the second stage model in the hierarchical latent variable model,if I am interested in the significance of single path.
jmbecker
SmartPLS Developer
Posts: 1281
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Disjointed Two Stage Approach

Post by jmbecker »

1) Any incoming or outgoing significant path is sufficient. On first glance, it seems that you always have at least one incoming or outgoing path that is significant. Thus, you should be ok.

2) If you want to test significance in the second stage model, the sample size there is important for sufficient power. But usually, you should have the same sample size at both the first and second stage.
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
chenchenchen
PLS Junior User
Posts: 5
Joined: Mon Mar 01, 2021 1:56 am
Real name and title: chentianzhi student

Re: Disjointed Two Stage Approach

Post by chenchenchen »

Thank you very much!
Locked