Solution:
The sample size allows researchers to know how many individuals are necessary to study, in order to estimate a certain parameter with the desired degree of confidence, or the number necessary to be able to detect a certain difference between the study groups, assuming that it really exists.
According to the protocol followed in the study which would be the basis for further studies by the FDA, the number of individuals chosen (large population of genetically similar rats) was necessary to study and estimate the determined parameter with an optimal degree of confidence.
Therefore, if the population is small, the degree of confidence in the results will not be the desired one since the study sample would not be representative of the population of interest and could increase the various types of error in the given scientific study. So, in this case, according to the protocol followed by the researchers, the larger the sample, the better the results, therefore the study with 100 rats is more valid than with 20 rats.