Is there a standard formula for calculating sample size for animal (rat) studies based on the number of variables to be used?
Ethically we should always chose the least number of animals (rats) for experimental research. But why do most papers show 5,6 or 7 in comparing groups?
Do we consider animals as a population or a sample?
You can conduct what is known as a power analysis using the publicly available G*power program.
Hello Stephanie,
I agree with David L Morgan , that doing an a priori power analysis is the preferred tactic in planning the sample size for a study. The key element is deciding how large or small an effect (often expressed as the standardized mean difference between treatment conditions) is worth detecting, should it exist. In animal studies (such as, does food as a reward reinforce bar-pressing behavior), the ES is often very large, which allows for very modest sample sizes to be used. However, for other kinds of comparisons, maybe not; that's why identifying the target ES is critical.
Your second question, why do so many use 5, 6, or 7 as the sample size, probably hinges on (a) the fact that published studies these folks refer to did likewise and made it into print, therefore they reason that "it must be ok"; and/or (b) large ESs were observed.
Third question. Formally, your study animals are a sample. Lab rats as furnished by commercial vendors are usually inbred for multiple generations so as to be as functionally interchangeable as possible. That's to help with generalizing your results beyond your sample (to a population of similarly interbred rats).
Good luck with your work!
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size