I have a nutrition experiment in which I will use laboratory rats. The number of rats is large, 80 rat. I want to estimate insulin at the beginning and end of the experiment, but the cost is high. Is it possible to use serum samples collected from the rats in one group and compare the groups to each other after that?
Hi! No, I wouldn’t recommend doing that. You’d lose biological variability, and if one of the animals has an abnormally low or high insulin level, you’d lose the entire data point and wouldn’t even know it — leading to a false positive or false negative result.If I understood correctly, you're not comparing an experimental group to a control group, but rather "before" and "after" measurements in the same animal — in other words, you have a dependent (paired) sample. In this case, statistical analysis should be based on comparing parameters within the same animal, not between different ones.So your dataset should include the same animals before and after treatment. Most likely, you’ll need to reduce the sample size or select the most representative animals based on other tests you've conducted, if applicable.
Thank you very much for your reply and for trying to help me. I am comparing the groups to each other and comparing the insulin levels of each group before and after. What are your suggestions for reducing the cost?
Hello Noura M. Abd Al-Maqsoud
As suggested by Alina Chaplygina , follow the dependent (paired) sample, if comparing before and after.
Regarding the suggestion for selecting the most representative animals based on other tests you've conducted, do so only if applicable and with solid data and supporting hypotheses.