Describe a situation where you would prefer using a non-parametric test instead of a parametric test?

During my final year, I conducted a statistics project analyzing survey results from students about their study habits. The data collected was ordinal, with responses ranging from 'very poor' to 'excellent.' Since the responses did not meet the normality assumption required for parametric tests, I chose to use the Kruskal-Wallis test to compare the different groups. This allowed me to effectively analyze the differences in study habits among various majors without compromising the integrity of the results. The findings helped highlight that certain majors had significantly different study habits, which was valuable information for our department.

While volunteering for a local charity, I helped analyze the effectiveness of different fundraising events. The data collected was skewed, with a few events raising significantly more than others. Instead of using a parametric test, I opted for the Mann-Whitney U test to compare two fundraising events. This decision allowed us to understand which event was more successful without being misled by the outliers. The insights gained helped the charity plan future events more effectively, leading to increased participation and funding.

In my first job as a junior analyst, I was tasked with analyzing customer satisfaction survey results. The data included responses on a scale from 1 to 5, which is ordinal. Knowing that parametric tests wouldn't be appropriate due to the nature of the data, I used the Wilcoxon signed-rank test to compare customer satisfaction before and after a service change. This approach not only yielded reliable results but also provided actionable insights for improving customer service strategies, demonstrating the practical importance of selecting the right statistical method.