I have “which test to use” stats question if anyone could please assist.  

dependent variables= violet rosettes/m2 (density) and % cover of 
vegetation classes (both transformed)

I sampled 60 plots in 2008 and then subjected them to one of three 
treatments (treatment=independent variable) (there is potential for two 
additional independent variables, plot # and plot type, though there 
shouldn't be a significant influence)

I resampled density and % cover classes in 2009.  

I want 2008 plots paired to 2009 plots (i.e. plot 1 2008 to plot 1 2009, 
etc) and I want to determine any effect of the treatments  
My data is not normal, even when transformed

Researching online has led me to think I should use a repeated measures 
ANOVA…yes, no?

The other option I was thinking about was to create a third dependent 
variable which would be the difference in density (and % cover) from 2008 
to 2009 and then analyzing this difference against treatment effect (the 
data for the difference as a dependent variable is still not normal)

I will also be resampling the 60 plots in 2010 so analysis may need to be 
modified and will be repeated after that data is collected.  

Please let me know if you have any suggestions.  

Thanks alot for your time!

Reply via email to