Hi all,
I could really do with some help. A caveat is that I am a compete statistics newbie.
I have a questionnaire with 20 questions. I have 7 variables that these questions are indicators for. The variable with the most indicators has 4 indicators. The rest are 3 or 2. My research (hypthoses) is comparing the effect of one variable on the other 6.
I am struggling to get survey responses so am hoping to hit the minimum number required for analysis. A level of power equal to .80 would suffice.
I have attached some pictures from Smart pls but please let me know what other info you need.
any and all help gratefully accepted.
Thanks,
George.
Need help determining sample size please
-
- PLS Junior User
- Posts: 1
- Joined: Tue Aug 01, 2017 12:51 pm
- Real name and title: George Ardagh
Need help determining sample size please
- Attachments
-
- reliability.PNG (16.9 KiB) Viewed 1999 times
-
- f-square.PNG (20.22 KiB) Viewed 1999 times
-
- boot_path.PNG (53.84 KiB) Viewed 1999 times
-
- SmartPLS Developer
- Posts: 1284
- Joined: Tue Mar 28, 2006 11:09 am
- Real name and title: Dr. Jan-Michael Becker
Re: Need help determining sample size please
First, I would recommend using the G*Power Tool.
Second, I would recommend reading this forum thread: viewtopic.php?f=5&t=4009
Third, you may have a problem in determining your effect size as you have only one predictor for several depenent variables. Thus, using the estimated f² effect sizes from your current model does not help you as they are somewhat misleading (f² is the additional explained variance relative to the unexplained variance of the predictor; if you have only one predictor it is quite a strong effect as you compare against zero explanaition).
I would actually use a weak effect size to determine the minimum sample size (e.g., 0.02 ≤ f² < 0.15) in allmost every study to get a conservative estimate of the required sample size.
Second, I would recommend reading this forum thread: viewtopic.php?f=5&t=4009
Third, you may have a problem in determining your effect size as you have only one predictor for several depenent variables. Thus, using the estimated f² effect sizes from your current model does not help you as they are somewhat misleading (f² is the additional explained variance relative to the unexplained variance of the predictor; if you have only one predictor it is quite a strong effect as you compare against zero explanaition).
I would actually use a weak effect size to determine the minimum sample size (e.g., 0.02 ≤ f² < 0.15) in allmost every study to get a conservative estimate of the required sample size.
Dr. Jan-Michael Becker, BI Norwegian Business School, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de