Very Small Sample size
Very Small Sample size
I have small sample size (18 cases) with seven constructs. I ran the model by using PLS and I got reasonable results. however, i have not run the model totally, I ran each two constructs separately. Please, is there any problem with what I did (ran the model with each two constructs separately)?
- Diogenes
- PLS Super-Expert
- Posts: 899
- Joined: Sat Oct 15, 2005 5:13 pm
- Real name and title:
- Location: São Paulo - BRAZIL
- Contact:
Hi,
If 18 is not the population…
This sample makes sense?
With only two cases we could have R2 = 100%, certainly this not make sense, but with 18, we could have the same problem.
See:
MARCOULIDES, G. A.; SAUNDERS, C. Editor’s comments: PLS: a silver bullet? MIS Quarterly, v. 30, n. 2, p. iii-ix, 2006.
Best regards,
Bido
If 18 is not the population…
This sample makes sense?
With only two cases we could have R2 = 100%, certainly this not make sense, but with 18, we could have the same problem.
See:
MARCOULIDES, G. A.; SAUNDERS, C. Editor’s comments: PLS: a silver bullet? MIS Quarterly, v. 30, n. 2, p. iii-ix, 2006.
Best regards,
Bido