The size of the sample used to train species distribution models affects how well they predict where species are found. This study looked at how the size of the sample used to test these models also impacts their accuracy. By simulating different scenarios, researchers found that as the sample size increases, the accuracy of the predictions also improves. Both the AUC and Se* statistics used to measure accuracy were equally precise and unbiased. The more data available, the better the models performed. It was found that a sample size of at least 20-30 is needed for accurate predictions, especially for rare or poorly known species.