Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : Parameter: Journal of Statistics

ROBUST PERMUTATION TEST FOR SPEARMAN CORRELATION AND ITS APPLICATION TO TESTING THE RELATIONSHIP BETWEEN OPEN UNEMPLOYMENT RATE AND NUMBER OF CRIMES Indriyani, Indriyani; Suliadi, Suliadi
Parameter: Journal of Statistics Vol. 5 No. 1 (2025)
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam Universitas Tadulako

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22487/27765660.2025.v5.i1.17444

Abstract

Pearson correlation coefficient is often unreliable for data that is not bivariate normally distributed or outliers are present, which affects the accuracy of measuring the strength of the linear relationship. Alternatively, Spearman correlation coefficient can be used to measure monotonic relationships without the assumption of normal distribution and can overcome the presence of outliers. Although the t-test approach to Spearman correlation is commonly used in theory, it is not always appropriate and can result in a Type I error when data are not normal or the sample size is small. To overcome these limitations, Yu & Hutson in 2022 proposed a robust permutation test method on the Spearman correlation coefficient using studentized statistics designed to overcome deviations from bivariate normality and small sample sizes, providing better control of Type I error. This research discusses the application of that method to analyze data of the open unemployment rate and the number of reported crimes in Indonesia. Using the permutation test, it was obtained the value of =0.2437 with a p-value <0.05, indicating a significant correlation between the two variables. The findings of this study are expected to provide a basis for effective policy recommendations in reducing crime rates by considering unemployment factors.