I am turning my work from medical statistics into biology statistics (fishery). I am wondering if there is a big differences in methodology or focus between these two fields? In medical field, we do a lot clinical trials or patient cohort studies, and we try to find associations by applying different types of significance tests and regressions. In the case of biology, there is very limited control of the sampling method. I have a feeling that the statistician's main focus would be to find the best way to sampling the data. Since I am really fresh in this field, can someone give me some ideas about the main focus, common methodologies in biology statistics? Is it really different from medical statistics? Do you have any nice books to recommend? Thanks
A couple things come to mind. As a disclaimer, my exposure to the statistics of Biology come from undergrad, and the occasional times my research overlaps with Ecologists.
- First, "biological statistics" is a staggeringly broad definition. What might be routine for the microbiology people may be impossible for the population ecologists, for example.
- As mentioned by Michael Chernick, there may be some sampling methods that fisheries research relies on that don't generally show up in clinical or population-based health research. Capture-Recapture comes to mind as one. The sheer difficulty in observing animals, who by their nature tend to be poor research subjects, and sample size issues. To take a somewhat extreme example, if one is doing Black Rhino research, your maximum N will be 4,240. But even for more abundant species, sampling large numbers can be extremely difficult.
- Multi-species analysis is generally of greater interest to Biologists than medical researchers. While one could make an argument that some infectious disease research is multi-species analysis, attempting to estimate species diversity, predator-prey relationships, ecological networks etc. are fairly field specific.
- While spatial statistics are becoming a thing in population health research, they're already a major aspect of biological research.
I suspect one of the hardest transitions will just be figuring out what counts as "standard" in the field in terms of analysis.