Should a Poisson regression be carried out with only 3 data points? I'm trying to test the relationship between the number of adults counted and the percentage heather cover over 3 areas. 
The data looks like this:
Area      Cover    Count
Field1     53.7      216
Field2     19.5        2
Field3     39.8     6106

Given that count data is involved I was going to carry out a Poisson regression. 
I know a relationship is unlikely given this data but my question is, is 3 data points enough for a regression analysis? This is important as I also want to test the relationship between adult counts and nest number, adult counts and vegetation height etc. but all of which only have 3 data points as I was only studying 3 areas on a site. 
With this type of data, I don't know if any other test could be used given that it is count data. 
I'm still a beginner when it comes to stats so any help would be much appreciated. 
I'm using R to carry out all my analyses.
 A: With this limited of a data set, you can actually just think about what the data are showing you on a case by case basis. In your case, there is not much to show. In Field 1, you have what seems like a 'large' number of adults (not sure what organism this is - if for example beetles, then perhaps 200 beetles is not a lot of beetles in a field?) and a pretty large amount of heather. Field 2, smaller amount of heather and smaller number of adults. All good. Field 3 breaks the nice little pattern. Without more data, there's no way to know if this is indicative of a 'non-relationship' or just variability. 
If you have an opportunity to gather data to your heart's content, you should probably work on doing a power calculation. I suspect that you have to take data opportunitistically and that it's a lot of work to collect data for a single field.  Something else to keep in mind is that you have percentage as well as count data. Percentage data (capped at something like 100%) sometimes needs to be dealt with as well, either through something like logistic regression, an arc-sine transformation, etc. 
