Question

Explain why one would use non-parametric tests? What are their advantages and disadvantages? At least a 100 word explanation

Answer 1

Non parametric tests refer to tests used in statistics to analyze ordinal or nominal data with small sample sizes. Non parametric models do not require making any assumptions about the distribution of the population, and so are sometimes referred to as a distribution-free methods.

This method can also be used when the data is quantitative but has an unknown distribution, non-normal, or has a sample size so very small that the central limit theorem(conversion of other non-normal distribution to normal distribution) can't be applied.

Non parametric tests have some advantages especially when observations are nominal, ordinal (ranked), subject to outliers or measured imprecisely. In these situations they are difficult to analyze with parametric methods without making major assumptions about their distributions. In parametric tests, we used make some untrue assumptions, when in some cases they are quite irrelevant. So here is a great advantage. Non parametric tests are also relatively simple to conduct. Non parametric test can also be be used to check the randomness. Some commonly used non-parametric tests are Wald-Wolfowitz Run Test, Wilcoxon Signed rank test, Mann Whitney U test, etc.

Disadvantages of Non parametric methods include lack of statistical power as compared with traditional approaches of testing. This is a particular concern if the sample size is small. Non parametric methods are geared toward hypothesis testing rather than estimation. Sometimes it is possible to obtain non parametric estimates and associated confidence intervals, but generally this is not straightforward, it is kind of unfamiliar. Tied values seems to be problematic when these are common, and adjustments to the test statistic may be necessary.