statistical power analysis

estherxing · Post by **estherxing** » Fri Apr 13, 2007 9:33 am

Dear All,

I am now calculating the necessary sample size to get a good power value (i.e. above 0.8). When I refer to Cohen (1988), I realize there are power analysis for different statistical test methods, like 't-test', 'chi-square test', 'variance and covariance', 'multiple regression and correlation analysis' as well as 'set correlation and multivariate methods'. The power value/sample size calculation is different for each case.

May I know which one shall I follow in PLS case? Is it 'multiple regression and correltion analysis'?

Thank yoU!

Best,
Esther

dhenderson17 · Post by **dhenderson17** » Fri Apr 27, 2007 7:04 pm

I had the same problem and used the tables for multiple regression from Cohen (1988). I clarified this with a professor in the department who is familiar with PLS as well....

estherxing · Post by **estherxing** » Tue May 01, 2007 7:16 am

Thank you!

I have followed the table for 'multiple regression' as well :)

dhenderson17 · Post by **dhenderson17** » Tue May 01, 2007 12:58 pm

No problem...this also passed my committee!