I have data extracted from literature on biomass production and related variables. I would like, if possible to conduct a seried of factor based analyses comparing biomass growth on managed and unmanaged treatments.

However, the data is nonparametric, and the majority of the data is on non-managed plots, (9 managed against 39 umanaged observations).

I'm thinking I would probably want a Mann-Witney test, since the data is unranked, but I'm concerned that the low and uneven sample sizes might preclude that.

In order to be robust about things, I want to make sure there is some valid statistical option, otherwise it may well be that I'll need to use descriptive options instead, but it would be good to be able to say something a bit more definitive.

More Daniel Forster's questions See All
Similar questions and discussions