Yes you are right, here it's an updated version with the diagram. Let me know if you need other help! Also i would like to know what do you think for my workThanks
Calculation of sample size for t-test
It is considered that, there is a small effective size when Cohen’s d is around 0.2.
So in this case, d is considered equal to 0.2. To calculate the sample sizes for the two groups,
power of analysis, alpha error probability and allocation ratio are needed.
Alpha error probability is 0.05, power of analysis is equal to 1-beta=0.8 and allocation ratio is
1 as the two groups have equal sample sizes.
Using those values for the variables in G*power application, we find that sample size for
each group is 310.
So totally 620 samples are needed. This is a really big number, and in statistical research
sometimes it’s really hard to find so many samples.
t tests - Means: Difference between two independent means (two groups)
Effect size d=0.2
α err prob=0.05
Power (1-β err prob)=0.8
Allocation ratio N2/N1= 1
Output:Noncentrality parameter δ=2.4899799
Sample size group 1=310
Sample size group 2=310
Total sample size=620
As stated before, sometimes it is really difficult to find samples for a statistical analysis. For
such cases, Erdfelder (1984) developed the concept of compromise power analysis.
Using G*Power, and the compromise selection for t-test, alpha and beta probabilities can be
determinded if we take half of the sample size. That means that the sample size for each
group is now 155. We also have to determine the beta/alpha ratio, which we assume to be
1, as in basic research they are considered equally serious.