Distribution and coverage of areas in relation to second set of areas. As expected?

Discussion created by dj77 on Feb 20, 2012
HI everyone,

I'm analyzing a set of parks in relation to a set of important species areas developed from distribution models and a ranking algorithm. Because of how the ranking algorithm works, important areas tend to be clustered.  The parks cover approximately 15% of my total study area.  A simple analysis finds that approximately 15% of my important species areas are found within parks.  What I would like to know is, can we say that this is what we would expect (or better or worse)  given the coverage of parks in the study area? 

I suppose that to test this we could assume that the parks are distributed randomly throughout the study area (even though we know they are really not).  We could then shift the parks a thousand times within the study area to get an expected coverage of important areas.  We could then see if the value we are getting is within a standard deviation of what would be expected.

Does this method make sense?
If so what tools are available to do this kind of analysis? 
Do I have to hold the shape/area of the parks constant?
Better methods and leads for this sort of analysis?

Thanks in advance,