Collier, Zachary K.Zhang, HaobaiSoyoye, Olushola2022-06-292022-06-292022-06-06Zachary K. Collier, Haobai Zhang & Olushola Soyoye (2022) Alternative methods for interpreting Monte Carlo experiments, Communications in Statistics - Simulation and Computation, DOI: 10.1080/03610918.2022.20824741532-4141https://udspace.udel.edu/handle/19716/31039This is an Accepted Manuscript of an article published by Taylor & Francis in Communications in Statistics - Simulation and Computation on 06/06/2022, available at: http://www.tandfonline.com/10.1080/03610918.2022.2082474. This article will be embargoed until 06/06/2023.Research methodologists typically use descriptive statistics and plots to report the findings of Monte Carlo experiments. But previous literature suggests that Monte Carlo results deserve careful analysis rather than relying on simple descriptive statistics and plots of results, given the complex data conditions in simulation studies. As an alternative, data mining methods can also help readers digest Monte Carlo experiments. Therefore, our paper uses data mining methods to provide two novel contributions. First, we use detailed descriptions and code to illustrate how to use two data mining methods to analyze results from Monte Carlo experiments. Second, we demonstrate how data mining methods can be used in conjunction with interpreting plots, performing analysis of variance tests, and calculating effect sizes. Our study raises the awareness that there are alternative methods to interpretation and serves as a guide to readers for explaining the importance of manipulated conditions in Monte Carlo experiments.en-USdata miningmixture modelMonte Carlo simulationArticle