The researchers studied the presence of a harmful bacteria in wild deer poop from different locations. They used a special test to see if the proportions of positive samples were the same at all locations. With only a few positive samples, they found that a specific test called Fisher's exact test gave more reliable results than the usual chi-square test. The results could change a lot with just a small change in the number of positive samples at each location. Also, the exact test took longer to compute in a software program.