Abstract

As many real datasets do not satisfy the normality assumption, the need for simulation studies related to nonnormality has arisen. To conduct such simulation studies, it is important to generate appropriate non-normal data. Although many non-normal data generation methods have been developed, there is a lack of studies that discuss the entire process of non-normal data generation. Therefore, this study offers guidelines that researchers can consider for conducting simulation studies by outlining the process of non-normal data generation with specified steps. Specifically, the non-normal data generation process will be discussed based on the power method, which is the most widely used technique for generating non-normal data, and the program and functions that researchers can utilize will also be examined. Furthermore, by providing a real example of non-normal data generation based on a two-factor CFA model along with the program code, we expect that this study will enhance understanding of the data generation process while providing tools that researchers can use.