Correlation analysis is used if you want to know whether there is a correlation between two phenomena. To determine the type of correlation to be used, researchers need to consider whether or not the assumption of normality and data characteristics are met. The purpose of this study was to compare the results of the analysis using several correlation tests with both parametric and non-parametric approaches. The method used is to provide data simulation with three types of data characteristics, namely normal, skewed and data containing outliers. The test used is the correlation in the parametric approach with the Pearson Product Moment Test, while for the non-parametric approach is the Spearman Rank and Kendall Tau Rank tests. Furthermore, a case study is given. The results show that in correlation testing without considering data distribution and data characteristics. Can produce inaccurate conclusions.
Copyrights © 2025