Latent Variable Value

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
Post Reply
ayunurma
PLS Junior User
Posts: 1
Joined: Tue Jul 18, 2017 2:34 pm
Real name and title: Ayu Nur Maulidian

Latent Variable Value

Post by ayunurma »

I am a new PLS user, and I still learn about PLS too.

Here, my team has learnt from many source, then we conclude that:
1. indicator is observed variable, so the presence is to measure the latent variable
2. Latent variable is unobserved variable, so it needs the help of indicator which formed it (in this case we use formative indicator) to measure them.

is that statement correct?

and we also have some question:
1. if the LV has measured by the indicators, so where is the value? does the measurement from the indicators give us a value of LV as the result?
2. if there is no values, does the PLS only estimates the path coefficient?
3. can we interpret the path coefficient just like the regression (even for reflective and formative)? so if the path coef value is 0.46, does it means that when the X1 change 1 point, then the Y1 will change 0.46 point?

hope you can understand what I mean exactly
we would like to say thank you for all of your responses

Regards,
Ayu
User avatar
cringle
SmartPLS Developer
Posts: 818
Joined: Tue Sep 20, 2005 9:13 am
Real name and title: Prof. Dr. Christian M. Ringle
Location: Hamburg (Germany)
Contact:

Re: Latent Variable Value

Post by cringle »

Yes that's correct.

1. if the LV has measured by the indicators, so where is the value? does the measurement from the indicators give us a value of LV as the result?
--> Run the PLS algorithm and take a look at the results report which includes the latent variable scores.

2. if there is no values, does the PLS only estimates the path coefficient?
--> see the previous response.

3. can we interpret the path coefficient just like the regression (even for reflective and formative)? so if the path coef value is 0.46, does it means that when the X1 change 1 point, then the Y1 will change 0.46 point?
--> yes but as standardized regression coefficients.

Best regards
Christian Ringle
Post Reply