1
$\begingroup$

This is probably a very basic question, but I have spent quite some time without finding a starting point for a solution, so I would be grateful for any help.

I want to test the hypothesis that the proportion of a binary variable in a number of samples is equal to a known proportion of that variable.

Specifically, I am mapping weed species in pastures. I want to know whether they occur overproportionally in tall (or in short) parts of a pasture. I am investigating nine pastures. For these, I know the proportion of area covered by tall (or short) vegetation from remote sensing. I also mapped all individuals of the weed species. For each individual, I know the proportion of tall (or short) vegetation within a 1-m buffer around the individual.

As a result, I have three values for each of the nine pastures: The known proportion of tall (short) vegetation, the observed proportion of tall (short) vegetation in the vicinity of weeds, and the number of weed individuals on which this observation is based.

Which test would I use to test whether the vicinity of weeds had a larger or smaller share of tall/short areas than would be expected from the known proportion of tall/short areas? Essentially, my null hypothesis is that observed vs. known proportion follow a 1:1 line, and I want to test for deviations from that.

Any help is greatly appreciated!

$\endgroup$

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.