 Random forests, RFs, are a popular machine learning algorithm used for classification tasks. A case study was conducted to analyze the effects of various input data characteristics on RF classifications. The study found that high-dimensional datasets should be reduced before use in order to reduce the number of correlated variables. Additionally, it was found that the size of the training dataset has a significant effect on the accuracy of the RF model. Furthermore, the study showed that the training data should be randomized or created in a way that reflects the true class proportions in the landscape in order to obtain accurate predictions. Finally, the study demonstrated that spatial autocorrelation can lead to inflated estimates of RF out-of-bag classification accuracy, thus highlighting the importance of reducing this effect. This article was authored by Kareem Millard and Murray Richardson.