Sample Size Planning

library(pwr)

## Warning: Paket 'pwr' wurde unter R Version 4.2.3 erstellt

# based on the difference observed in the previous study
pwr.t.test(d=(-0.2-(-0.55))/1.5,
           power=0.8,
           sig.level=0.05,
           type="two.sample",
           alternative="greater")

## 
##      Two-sample t test power calculation 
## 
##               n = 227.7938
##               d = 0.2333333
##       sig.level = 0.05
##           power = 0.8
##     alternative = greater
## 
## NOTE: n is number in *each* group

At least 228 subjects per condition. However, note that an additional criterion was reaching a specific estimation precision (defined as a maximum width of the 95 CIs around the group means.). Another was that equally many subjects were supposed to be in each condition:

1. no 95%CI of the means should be wider than \(0.5\) points on the rating scale and

2. all conditions should have the same number of subjects. Both additional criteria were reached after \(n~=240\) subjects, which terminated the data collection.

These criteria were fulfilled with \(n~=240\) subjects in each condition, which terminated data collection.