I want to compare my diversity results between three communities of caridean shrimps (Shannon, Simpson, and beta indices). Hill numbers is an alternative to do so given that they show to you the effective number of species of a community. On the other hand, the t-test of Hutcheson is a method used to analyze statistical differences between Shannon results. So, I decided to perform both of them in order to know the magnitude of the difference ( with the Hill numbers) and if that difference has statistical significance (with the t-test of Hutcheson). So my question is if any of you have done what I've done and what are your recommendations? because I checked this website http://www.loujost.com/Statistics%20and%20Physics/Diversity%20and%20Similarity/How%20to%20compare%20the%20diversities%20of%20two%20communities.htm
and now I have this issue of not being sure of the process I followed because the author says that first you have to convert Shannon to effective number of species and then do the t- test. In other words, using the hill numbers within the t value equation instead of Shannon raw numbers. That seems a little bit unreliable to me because the t-test equation requires Shannon values, which, as you know, comes from an equation that has logarithmic foundations ( H’=–sumatory (Pi ln Pi) and the hill numbers are not logarithmic.
So what do you think? do I use the raw numbers (logarithmic) of Shannon or do I first transform the results to Hill numbers and then do the t-test.
I really really hope I did not confused you with my not-so-good english and my doubts.
Best regards and thank you
Hi, I understood what you did and it seems unnecessary. If you want to compare the diversity of epscies between pairs of communities, I recommend that you use the Hill series. Check the 2007 and 2007 Jost works that provide you with the necessary information. For using the Hutcheson method implicitly implies that the data comply with the premises of the parametric tests. In addition, this last method does not provide the necessary information, since it is more statistical than biological. The comparison using Hill's series tells you how many times a community is more or less diverse.
Article Jost, L. Entropy and diversity. Oikos
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis