Chi-Square and Correlation-Topic 4 DQ 2 Response- see upload for question
Chi-Square and Correlation-Topic 4 DQ 2 Response- see upload for question
What are additional downsides to using a non-parametric test?
ORDER A PLAGIARISM-FREE PAPER HERE !!
Solution
What are additional downsides to using a non-parametric test?
Non-parametric tests refer to mathematical methods utilized in the testing hypothesis in statistics, and they are distribution-free; that is, they do not make any assumptions about the frequency distribution of the variables being tested. These types of tests are commonly utilized where the data is skewed and include techniques not dependent on data with a particular distribution. Some of the downsides of a non-parametric test are that; they are less efficient and powerful than parametric tests. Parametric tests rely on an underlying statistical distribution of data, and thus several validity conditions have to be met making the results more reliable; however for the non-parametric tests, they do not rely on any distribution and thus can be applied even when the validity conditions are not met, and thus the results are less reliable. Since non-parametric tests are distribution-free, they may or may not provide accurate results (Asmare & Begashaw, 2018).
Most non-parametric tests utilize data inefficiently/inadequately as they transform observed values into ranks and groups; this can lead to a lack of precision, leading to erroneous acceptance or rejection of the null hypothesis giving misleading information. The tests are also only suitable for smaller sample sizes since the sampling distribution for the non-parametric test is numerous and can be cumbersome. There also exists a threat of violation of the assumptions, which can render the results inaccurate; non-parametric tests are based on two assumptions. Data in each comparison group exhibits similar homogeneity of variance; such assumptions can be easily violated, rendering the results inaccurate. The tests are also only suitable for ratio and interval data (Asmare & Begashaw, 2018).
References
Asmare, E., & Begashaw, A. (2018). Review on parametric and nonparametric methods of efficiency analysis. Biostatistics and Bioinformatics, 2(2), 1-7. https://www.researchgate.net/profile/ErkieAsmare/publication/327670168_Review_on_Parametric_and_Nonparametric_Methods_of_Efficiency_Analysis/links/5b9d445792851ca9ed0d596d/Review-on-Parametric-and-Nonparametric-Methods-of-Efficiency-Analysis.pdf
