Dummy Variable

Frequently Asked Questions about the SmartPLS software. Please refer to this section first, if you have problems with the software.
Post Reply
alnawas
PLS Junior User
Posts: 5
Joined: Thu Mar 01, 2012 4:49 pm
Real name and title:

Dummy Variable

Post by alnawas »

Dear Professors,

It would be a massive help if you could please explain me how to deal with dummy variables using SmartPLS?

More specifically I want to test to the effect of student orientation on student satisfaction and I want to control for the mission of the university over student satisfaction. I have five groups of missions, so should I code them like this:

University Alliance (group1): 1.0.0.0.0
Million+ (group2) : 0.1.0.0.0
Russell (group3 : 0.0.1.0.0
1994 (group4) : 0.0.0.1.0
GuildHE (group5) : 0.0.0.0.1

Does this sound right?

If that is right, then should I put all these five groups into one construct "University Mission" and then draw a path from "University Mission" towards student satisfaction?

I want to see when I control for the mission of the university, does student orientation still affect student satisfaction?

Many thanks in advance
User avatar
Hengkov
PLS Super-Expert
Posts: 1618
Joined: Sun Apr 24, 2011 10:13 am
Real name and title: Hengky Latan
Location: AMQ, Indonesia
Contact:

Post by Hengkov »

Hi,

I think no, you create construct "mission university" with five indicators (mode B) and connect to student satisfaction to control OK.

Best Regards,
Hengky
alnawas
PLS Junior User
Posts: 5
Joined: Thu Mar 01, 2012 4:49 pm
Real name and title:

Post by alnawas »

Hengkov wrote:Hi,

I think no, you create construct "mission university" with five indicators (mode B) and connect to student satisfaction to control OK.

Best Regards,
Hengky
Dear Prof. Hengky,

Many thanks for your kind suggestions.

I'll do exactly as you kindly suggested, but I'm still confused as what value should I give to each indicator?

More specifically, should I treat the five missions as classical control variables (i.e. 0.0.1)?

It would be a massive help if you could please explain me that!
Many thanks
User avatar
Diogenes
PLS Super-Expert
Posts: 905
Joined: Sat Oct 15, 2005 5:13 pm
Real name and title:
Location: São Paulo - BRAZIL
Contact:

Post by Diogenes »

Hi,
Yes, do it in the way that Latan explained (formative), but one of the categories should be used as reference (5 groups --> 4 dummies), for example:

University Alliance (group1): 0.0.0.0
Million+ (group2) : 1.0.0.0
Russell (group3 : 0.1.0.0
1994 (group4) : 0.0.1.0
GuildHE (group5) : 0.0.0.1

See one example in:
Falk, R. F., & Miller, N. B. (1992). A Primer for Soft Modeling. Ohio: The University of Akron Press.

Best regards,

Bido
alnawas
PLS Junior User
Posts: 5
Joined: Thu Mar 01, 2012 4:49 pm
Real name and title:

Post by alnawas »

Dear Prof. Bido,

Many thanks for your clarification and I truly appreciate that.

I did exactly as you said and I added the five categories formatively and one of the categories was 0.0.0.0.0. However, when I wanted to run the model, it didn't run because of that group, but when I changed it to 0.0.0.0.1. and then run the model, it worked.

Any suggestions would be extremely appreciated,
Warmest regards
User avatar
Diogenes
PLS Super-Expert
Posts: 905
Joined: Sat Oct 15, 2005 5:13 pm
Real name and title:
Location: São Paulo - BRAZIL
Contact:

Post by Diogenes »

Hi,

You will have just 4 formative indicators.

See the "Social status" in this example:
http://www.amazon.com/gp/product/images ... 55&s=books

The first indicator was included in the picture, but its weight = 0.

This is a way to include the concept as a control variable, but the weights are not interpretable (as some kind of importance).

Best regards,

Bido
alnawas
PLS Junior User
Posts: 5
Joined: Thu Mar 01, 2012 4:49 pm
Real name and title:

Post by alnawas »

It worked well... Loads of thanks for your extreme kindness,

Warmest regards!
ali.hebaz
PLS Junior User
Posts: 1
Joined: Sun Nov 27, 2022 11:41 am
Real name and title: Ali Hebaz

Re: Dummy Variable

Post by ali.hebaz »

Please can you explain more the apporach, i've did the same but i still have the same problem it cannot calculate
Post Reply