Thread: which statistical test to use? please help!

Hello!

I really need some help on which kind of statistical test i am supposed to use for my study.
I am trying to predict species richness/abundance of beetles based on some tree's characteristics such as sun exposure, shape of tree hole, diameter and height, canopy cover etc.
Some of the predictor variables (tree's characteristics) are continuous (e.g. height, diameter) and some are categorical (e.g. shape of tree hole). The dependent variable (species richness/abundance) are both categorical (at least so i assume: species richness is an index, and abundance is a quantitative variable, but it's discrete because it only includes natural numbers).
So, if my assumptions are right, what kind of statistical test should i use? I think I can't use linear regression. How about AIC?
Considering that I have never done a statistical analysis I am looking for the easiest test to perfom.

I hope someone can solve my doubts!
Thank you in advance!

Try switching your categorical dependent variable with a continuous variable. Then you can examine the statistical significance of species richness on this continuous variable (tree height for example) using ANOVA. If determining the statistical significance of one factor at a time is sufficient for your purposes that is. This might be do-able with a general linear model also, but i don't even want to think about that right now.

Hi,
I believe there are many useful sites which can help you with your problem. I have recently come across a site Online Math Tutor which has proved to be a very useful site. It provides online help in topics like algebra, trigonometry, statistics. You might get to learn and decide about the statistical test once you get in touch with this site.