Re: sample size specification in pwr2::pwr.2way

As for the logic, I dunno, but the source code has

pwr.fA<-function(a=a, b=b, alpha=alpha, size.A=size.A, f.A = NULL, delta.A=NULL, sigma.A=NULL){
N=size.A*a*b
if (is.null(f.A)) {f.A<-sqrt(((1/a)*(delta.A/2)^2*2)/sigma.A^2)
lambda<-N*f.A^2
...
pwr.fB<-function(a=a, b=b, alpha=alpha, size.B=size.B, f.B=NULL, delta.B=NULL, sigma.B=NULL){
N=size.B*a*b
if (is.null(f.B)) {f.B<-sqrt(((1/b)*(delta.B/2)^2*2)/sigma.B^2)
lambda<-N*f.B^2
 

I.e., the N differs between the two power calculations in pwrA and pwrB. You may want to check with the reference they cite...

-pd
   



> On 14 Oct 2025, at 18:39 , Floris van Vugt <[email protected]> wrote:
> 
> Dear R Help list,
> 
> My question is how sample size is specified in 2-way ANOVA power calculation in the pwr2 package.
> 
> In the function pwr2::pwr.2way documentation it gives an example use (source https://rdrr.io/cran/pwr2/man/pwr.2way.html) :
> ```
> ## Example 1
> pwr.2way(a=3, b=3, alpha=0.05, size.A=4, size.B=5, f.A=0.8, f.B=0.4)
> ```
> 
> I have trouble understanding what the total sample size is in this case. The documentation says `a` refers to the number of groups in Factor A, and `size.A` is the sample size per group in Factor A. Similarly for Factor B.
> 
> Based on that, I inferred that Factor A is supposed to have 3 groups with 4 individuals per group, meaning a total of 12 individuals in the experiment as a whole. However, in Factor B there are 3 groups with 5 individuals per group, which would lead to a total of 15 individuals. So my conclusions are contradictory: are there 12 or 15 individuals in this experimental design? Or yet a different number?
> 
> Any help appreciated.
> 
> Best wishes
> Floris
> 
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> [email protected] mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide https://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

-- 
Peter Dalgaard, Professor,
Center for Statistics, Copenhagen Business School
Solbjerg Plads 3, 2000 Frederiksberg, Denmark
Phone: (+45)38153501
Office: A 4.23
Email: [email protected]  Priv: [email protected]