How can I calculate the sample size for an animal trial? How many animals per group would be statistically significant for generalization of the results?
Hi, regardless the element of the sample if they are files, records, animals, lab resutls or people, the sample size is calculated similarly.
To do power analysis to estimate your sample size, you have to write your hypothesis, and based on that you decide what statistical test you will use. It should be one of the inferential statistics. so you need to determine the following: alpha {standard to be .05}, power [standard to be .80], effect size {small, moderate, or large, each test has its own value, you can find these values in the net}. Then download free programs to calculate the sample size such as G. power.
How to calculate sample size in animal studies?
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3826013/
That's very helpful. I will be studying histological measures so i guess 3-5 will be enough per group. Thank you for your time.
Yes I think for histological measures 3-5 animals are sufficient but to be sure any way use 5 if you can (at least in experimental groups)
Hi Shahinaz,
as mentioned before there are several factors affecting the number of your sample. FIrst you should decide the sample unit (one animal, one cage etc) and according to the statystical analysis you want to perform, you carry out the power analysis (to get the n°of sample). One really nice website to understand these key-points is http://www.3rs-reduction.co.uk/ from Prof Festing.
Hope it helps, good luck!
Viola
hi, i came across this very helpful website.
http://www.3rs-reduction.co.uk/html/6__power_and_sample_size.html
Hi, regardless the element of the sample if they are files, records, animals, lab resutls or people, the sample size is calculated similarly.
To do power analysis to estimate your sample size, you have to write your hypothesis, and based on that you decide what statistical test you will use. It should be one of the inferential statistics. so you need to determine the following: alpha {standard to be .05}, power [standard to be .80], effect size {small, moderate, or large, each test has its own value, you can find these values in the net}. Then download free programs to calculate the sample size such as G. power.
Hi Elena,
I find the information you provided is very useful? May I know the source of your information? I would like to know more about the justification of sample size.
Usually we have two ways for calculating sample size; one robust power analysis that depends on a host of variables such as type one error (Alpha), effect size, SD, power effect and direction of effect (one-tailed VS two-tailed analysis)- the second method which is more easy but less accurate; the calculation of the E Value; E = Total number of animals − Total number of groups- E between 10 and 20 is acceptable. Despite its simplicity, one could argue that the calculated E value- though being in between 10 and 20, it may suggest low sample size that may lead to missing significant effects even if it exists in population. But to be honest, calculating sample size using the power analysis method needs a professional statistician.
Please guide....
How to calculate animal size and do power calculation for qualitative data (IHC data) ? Please suggest at your earliest. I am not able to justify animal number for my proposal. My whole study is based on only qualitative data (expression of molecular markers on tissue section).
Dear Dr. Elena Redondo Castro and Nikola M Stojanovic,
How did you reached to the conclusion that 3-4 or 5 animals will be suitable and appropriate for histological measures. Is there any calculation for it? I need to submit a grant in which reviewers asked for power calculation for the same.
please suggest...
Yes! 5 animals are sufficient for histology and/or IHC. Among 5 animals depending on the sections you can choose minimum 10 non serial sections from each animal i.e. minimum 50 sections from each group (if the group is consist of 5 animals). To choose non serial sections you will have to consider sections which are minimum 30-50 micrometer apart. If you still feel problem then let me know.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size