My main background is biology and I am going to ask a probably naive statistic question here. To make sure it get viewed by statistics experts here so I post my question here. If you find the question should be somewhere else, please move it for me.
My problem is:
There are to forests that I want to assess their plant diversity, say forest F1 and F2. I take 3 areas for sampling/counting. In each area I count number of species and number of individual for each species. If we name species as S1, S2, S3, etc, the tables are like as follow. Because I can't attach a file so please copy it, save as .csv file and import by Excel to see the data.
1. What mathematic techniques and how do I use to measure the consistency of data in each forest? The aim here is to find out how TRULY the data, with three replicates (area1, 2 and 3), represents distribution of plant species in each forest. I thought about using one-way ANOVA but later found that it is not meaningful to do that, because number of individuals for one species can not be used to compensate for count of individual of other species. Two-way ANOVA will work then? I am not sure.
2. Similarly how do I compare forest F1 to forest F2 in order to find out if distribution of plant species in two forest is the same or different?