I'm new to statistics so am trying to choose the best analysis to find if;
There is a difference between canopy cover (0-100%) at different sites
There is a relationship between canopy cover and nominal environmental variables, for each site.
I can convert canopy cover to binary data without losing information so I think for the second question my best bet is Binary Logistic Regression using the stepwise procedure. Is this correct? Do I need to do any further analysis?
For the first question, I think I need to do a Kruskal-Wallis Test however the number of observations of canopy cover are vastly different between sites. For instance at one site along a 600m transect I have 600 observations of canopy cover and at another site along a 300m transect I have 300 observations of canopy cover. This seems to be an issue because I have to tell the program (XLStat) to ignore missing observations (data). Is this a problem and if so, how do I get around it?
Any help is hugely appreciated