Hi everyone
Do anyone know how to measure control variables using SmartPLS?
Appreciated
Measure Control Variables
-
- PLS Junior User
- Posts: 3
- Joined: Tue Sep 16, 2014 7:32 pm
- Real name and title: Hebat-Allah
- Hengkov
- PLS Super-Expert
- Posts: 1599
- Joined: Sun Apr 24, 2011 10:13 am
- Real name and title: Hengky Latan
- Location: AMQ, Indonesia
- Contact:
Re: Measure Control Variables
Hi,
First, you need a good reason why the use of control variables, and why it is important. Many researchers do not have a strong justification when using the control variables. Generally in the form of demographic control variables such as age, class, gender, and so on. So, you do not need to measure it, just put it only into the model, and see the effect.
Greetings,
First, you need a good reason why the use of control variables, and why it is important. Many researchers do not have a strong justification when using the control variables. Generally in the form of demographic control variables such as age, class, gender, and so on. So, you do not need to measure it, just put it only into the model, and see the effect.
Greetings,
-
- PLS Junior User
- Posts: 3
- Joined: Tue Sep 16, 2014 7:32 pm
- Real name and title: Hebat-Allah
Re: Measure Control Variables
Hi Hengkov,
Thank you very much for your reply. There is a good reason in the literature for including the control variables (CV) in my model. Based on the literature, those CVs have effect on my DVs, so, I have to include them in the model to reduce the error term.
I want to know that using SmartPLS, do I need to use CV as IVs, but represent them by one indicator (which calculated previously) or there are any other methods of measuring them?
Then, how to comment about their effect, is it the value of R2, or the path coefficient, or both?
Best
Thank you very much for your reply. There is a good reason in the literature for including the control variables (CV) in my model. Based on the literature, those CVs have effect on my DVs, so, I have to include them in the model to reduce the error term.
I want to know that using SmartPLS, do I need to use CV as IVs, but represent them by one indicator (which calculated previously) or there are any other methods of measuring them?
Then, how to comment about their effect, is it the value of R2, or the path coefficient, or both?
Best
- Hengkov
- PLS Super-Expert
- Posts: 1599
- Joined: Sun Apr 24, 2011 10:13 am
- Real name and title: Hengky Latan
- Location: AMQ, Indonesia
- Contact:
Re: Measure Control Variables
Hi,
Yes, use CV as IVs and connect to the DVS who want to be controlled. To determine whether the IVs have any effect or not, see its significance. Surely only the R ^ 2 value will always be greater if no CV, but the number of samples required will also be greater.
Greetings,
Yes, use CV as IVs and connect to the DVS who want to be controlled. To determine whether the IVs have any effect or not, see its significance. Surely only the R ^ 2 value will always be greater if no CV, but the number of samples required will also be greater.
Greetings,
-
- PLS Junior User
- Posts: 3
- Joined: Tue Sep 16, 2014 7:32 pm
- Real name and title: Hebat-Allah
Re: Measure Control Variables
Hi,
Thank you very much.
Appreciated
Best,
Thank you very much.
Appreciated
Best,
-
- PLS Junior User
- Posts: 6
- Joined: Thu Oct 30, 2014 7:04 pm
- Real name and title: Marian Dragt MBA - PhD Student
Re: Measure Control Variables
How can I check whether the change in R^2 is significant? I included a control variable, which has a significant effect, but I don't know how to check whether the increase of R^2 s significant or not.
- Hengkov
- PLS Super-Expert
- Posts: 1599
- Joined: Sun Apr 24, 2011 10:13 am
- Real name and title: Hengky Latan
- Location: AMQ, Indonesia
- Contact:
Re: Measure Control Variables
Hi,
It's easy, you simply test the model with and without the control variables, and see the changes in the value of R^2.
Greetings,
It's easy, you simply test the model with and without the control variables, and see the changes in the value of R^2.
Greetings,
-
- PLS Junior User
- Posts: 6
- Joined: Thu Oct 30, 2014 7:04 pm
- Real name and title: Marian Dragt MBA - PhD Student
Re: Measure Control Variables
Thanks Hengkov, that's wat I did :). But how can I check whether that difference is significant?
- Hengkov
- PLS Super-Expert
- Posts: 1599
- Joined: Sun Apr 24, 2011 10:13 am
- Real name and title: Hengky Latan
- Location: AMQ, Indonesia
- Contact:
Re: Measure Control Variables
Hi,
Usually if the control variables have a significant effect, it will automatically provide a significant effect also on the R^2. Significant differences between the R^2 before and after the control variables included clearly be seen.
Greetings,
Usually if the control variables have a significant effect, it will automatically provide a significant effect also on the R^2. Significant differences between the R^2 before and after the control variables included clearly be seen.
Greetings,