Hello,
I'm trying to perform an post-hoc power analysis for my thesis reserach. I downloaded G*power, choose the multiple regression analysis. But I'm not sure which number should be used as the number of indicators.
I have a model with one dependent LV and several independent LVs. And these independent LVs are in turn again predicted by some other LVs.
let's say my model looks like this:
A->U->X
B->Y->X
What should I use as the number of indicators? 2 or 4? Thank you very much for your help!
Power Analysis
How to compute statistical power of complete PLS model?
How to compute statistical power of complete PLS model?
Plz clarify me.
Plz clarify me.
Computing statistical power
How you calculated statistical power of whole PLS model?
If you use G*Power, what we should select?
Test family: t test?
Statistical test: Linear multiple regression: Fixed model, single regression coefficient?
Type of power analysis: Post hoc: Compute achieved power - given alpha, sample size, and effect size?
Input Parameter
Tail(s): One
Effect size f^2: ? (what we should give in this input)
Alpha err prob: ? (what we should give in this input)
Total sample size: (this is fine)
Number of predictor: Largest number of IV impacting a DV.
What should be the above parameter? Where we can get it from our SmartPLS result?
Plz clarify me, I am struggling since long.
Thank you.
If you use G*Power, what we should select?
Test family: t test?
Statistical test: Linear multiple regression: Fixed model, single regression coefficient?
Type of power analysis: Post hoc: Compute achieved power - given alpha, sample size, and effect size?
Input Parameter
Tail(s): One
Effect size f^2: ? (what we should give in this input)
Alpha err prob: ? (what we should give in this input)
Total sample size: (this is fine)
Number of predictor: Largest number of IV impacting a DV.
What should be the above parameter? Where we can get it from our SmartPLS result?
Plz clarify me, I am struggling since long.
Thank you.
this is better
http://www.smartpls.org/forum/viewtopic ... r+analysis
http://www.smartpls.org/forum/viewtopic ... r+analysis