Hi all
I am in problem with my sample size (26 cause difficulty to get response) , with a lot variables. can i perform a PLS and establish my structural model.
addition PCA does not adequate in case of smaller sample.
what can i do.thanks a lot for all
Small sample
Small sample
PhD candidate managment
>Hengkov
thanks Hengkov
i know about this constaint but my model dont still in this condition
- i studies relation: motivation, antecedant, aptitude , skill, Human capital and their effect to firm success (in 4variable all dicotomic (y/n)).
my simple size (25) and have minimum 8 variable for each latent construct.
- second not all my variable is likert , do i change its on likert skale to perform in PLS???
if have idea plz show me
regards
i know about this constaint but my model dont still in this condition
- i studies relation: motivation, antecedant, aptitude , skill, Human capital and their effect to firm success (in 4variable all dicotomic (y/n)).
my simple size (25) and have minimum 8 variable for each latent construct.
- second not all my variable is likert , do i change its on likert skale to perform in PLS???
if have idea plz show me
regards
PhD candidate managment
Re: >Hengkov
Hi Ramzidz,
Greetings,
Dennis
Based on your description (6 variables) you'll need >60 respondents. The alternative (a reduction of the complexity of your model to 3 variables) is also difficult as a result of your sample size. You'll really need to increase your sample.ramzidz69 wrote:i studies relation: motivation, antecedant, aptitude , skill, Human capital and their effect to firm success (in 4variable all dicotomic (y/n)). my simple size (25) and have minimum 8 variable for each latent construct.
SmartPLS will accept dichotomous variables.ramzidz69 wrote:second not all my variable is likert , do i change its on likert skale to perform in PLS???
Greetings,
Dennis
Dificulty in sample size
To M. Dennis von Bergh
special thanks M. Dennis von Bergh , i know this , but here in tunisia is so difficult to get respondants who accept to fill an survey as our convinance.
i thisk i shood reduce my latents variable to 3 as you see.
is it oblige to use the same scale for each latent construct or we can have different scale for items (VM), like likert scale (1 to 5) and dichotomus (0,1) in the same LV. or we oblige to use the same and unique scale for each LV.
best regard
special thanks M. Dennis von Bergh , i know this , but here in tunisia is so difficult to get respondants who accept to fill an survey as our convinance.
i thisk i shood reduce my latents variable to 3 as you see.
is it oblige to use the same scale for each latent construct or we can have different scale for items (VM), like likert scale (1 to 5) and dichotomus (0,1) in the same LV. or we oblige to use the same and unique scale for each LV.
best regard
PhD candidate managment
- Diogenes
- PLS Super-Expert
- Posts: 899
- Joined: Sat Oct 15, 2005 5:13 pm
- Real name and title:
- Location: São Paulo - BRAZIL
- Contact:
Hi,
With so small sample size, just huge effects will be detected as significant.
See:
COHEN, J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. New York: Psychology Press, 1988.
MURPHY, K. R.; MYORS, B.; WOLACH, A. Statistical power analysis: a simple and general model for traditional and modern hypothesis tests. 3rd ed. New York: Routledge – Taylor & Francis Group, 2009.
ZILIAK, S. T.; MCCLOSKEY, D. N. The Cult of Statistical Significance: how the standard error costs us jobs, justice, and lives. Ann Arbor, MI: The University of Michigan Press, 2008.
Best regards,
Bido
With so small sample size, just huge effects will be detected as significant.
See:
COHEN, J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. New York: Psychology Press, 1988.
MURPHY, K. R.; MYORS, B.; WOLACH, A. Statistical power analysis: a simple and general model for traditional and modern hypothesis tests. 3rd ed. New York: Routledge – Taylor & Francis Group, 2009.
ZILIAK, S. T.; MCCLOSKEY, D. N. The Cult of Statistical Significance: how the standard error costs us jobs, justice, and lives. Ann Arbor, MI: The University of Michigan Press, 2008.
Best regards,
Bido