Can I use PLS SEM to validate my instrument which i developed from grounded theory

Frequently asked questions about PLS path modeling.
Post Reply
micpatlem
PLS Junior User
Posts: 3
Joined: Sun Jan 26, 2020 2:38 am
Real name and title: drg. Michael Andreas Leman, MMedEd

Can I use PLS SEM to validate my instrument which i developed from grounded theory

Post by micpatlem » Sun Jan 26, 2020 3:19 am

My name is Michael, I am studying about the characteristics of healthy role model in medical institutions. From Grounded theory, I developed new instrument that I used for survey in the next phase of my study. I have some questions according to my study, especially when I got data from survey that I intended to use for validating my instrument:

- Could I use PLS SEM to validate my new developed instrument?" The sample size is 79 sample which I consider meet the minimum sample size to use PLS SEM and my research aims was exploratory.
- Is my path model right? It has 7 construct (PH, SH, MH, SpH, WTP, LLL, IHB) which I categorized as 7 exogenous variables and the endogenous variable is the characteristics of healthy role model (HRM). For better understanding to my model, please look the path model I attached here.
- I couldn't continue to report R2, f2 since the indicator of 7 exogenous variable (PH, SH, MH, SpH, WTP, LLL, HB) is equivalent with endogenous variable (HRM)

Please help me to answer this questions.

Sincerely,
Michael
Attachments
path model paling awal.png
path model paling awal.png (106.58 KiB) Viewed 2572 times

jmbecker
SmartPLS Developer
Posts: 1079
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Can I use PLS SEM to validate my instrument which i developed from grounded theory

Post by jmbecker » Sun Jan 26, 2020 9:22 pm

You describe a reflective-formative higher-order construct. And it seems you model it by using the repeated indicator approach. Such a model is generally valid, but comes with some problems.

1) As you describe it does not allow you to assess the R² etc. because the endogenous construct in your model is composed of its lower-order constructs. They are meant to describe it entirely.

2) Your model as such is not identified without a link of the higher-order construct (HRM) to another construct which is not one of its own lower-order components.

3) You might want to estimate the higher-order component (HRM) where you use the repeated indicators using Mode B for the indicators.

Please read:
Sarstedt, M., Hair Jr, J. F., Cheah, J. H., Becker, J. M., & Ringle, C. M. (2019). How to specify, estimate, and validate higher-order constructs in PLS-SEM. Australasian Marketing Journal (AMJ).
Becker, J. M., Klein, K., & Wetzels, M. (2012). Hierarchical latent variable models in PLS-SEM: guidelines for using reflective-formative type models. Long Range Planning, 45(5-6), 359-394.
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de

micpatlem
PLS Junior User
Posts: 3
Joined: Sun Jan 26, 2020 2:38 am
Real name and title: drg. Michael Andreas Leman, MMedEd

Re: Can I use PLS SEM to validate my instrument which i developed from grounded theory

Post by micpatlem » Mon Jan 27, 2020 2:33 pm

Dear Dr. Jan-Michael Becker

Thanks for your helping information. I have read the references you mentioned but still have some questions.
I read your publication " Hierarchical latent variable models in PLS-SEM..." (Becker et al., 2012) and try to answer the guidelines you wrote in reporting guidelines p.377. First, my model is reflective-formative type. Second, I chosed repeated indicator with mode A measurement (because my first order constructe are reflective), and use path for the PLS-SEM algorithm. But as the literature said, the R2 is 1.0 because the same indicator on dependent and independent variable. How could I solve this problem?

I read the references and the answer that we must proposed is an extended repeated indicator approach or two stage approach. How could I perform this approach since I dont have any antecendent (endogenous) or exogenous construct which has a number of indicators? Here I attached the two stages model that I could not be continue to calculate. I wish I have some of your suggestions in my problem, thank you.


Sincerely yours
Michael
Attachments
REPEATED INDICATOR MODE A.png
REPEATED INDICATOR MODE A.png (84.24 KiB) Viewed 2563 times
2 STAGE.png
2 STAGE.png (9.25 KiB) Viewed 2563 times

micpatlem
PLS Junior User
Posts: 3
Joined: Sun Jan 26, 2020 2:38 am
Real name and title: drg. Michael Andreas Leman, MMedEd

Re: Can I use PLS SEM to validate my instrument which i developed from grounded theory

Post by micpatlem » Tue Jan 28, 2020 2:01 am

Dear. Dr. Dr. Jan-Michael Becker

I tried to estimate my hierarchical model by using two-stages approach. Is it a right? thanks

Sincerely,
Michael
Attachments
Two stage-1st stage-reflective-formative type ANALISIS.png
Two stage-1st stage-reflective-formative type ANALISIS.png (66.8 KiB) Viewed 2561 times
Two stage- 2nd stage reflective-formative type-ANALISIS.png
Two stage- 2nd stage reflective-formative type-ANALISIS.png (23.24 KiB) Viewed 2561 times

Post Reply