Question

The article “Understanding Nonparametric Tests“ explains what a nonparametric test is and provides a real-world example:

A nonparametric test is a hypothesis test that does not require the population's distribution to be characterized by certain parameters. For example, many hypothesis tests rely on the assumption that the population follows a normal distribution with parameters μ and σ. Nonparametric tests do not have this assumption, so they are useful when your data are strongly nonnormal and resistant to transformation. However, nonparametric tests are not completely free of assumptions about your data. For instance, nonparametric tests require the data to be an independent random sample. For example, salary data are heavily skewed to the right, with many people earning modest salaries and fewer people earning larger salaries.

To complete the Discussion activity, please do the following:

Answer each question fully. Use a minimum of 250 words for a complete discussion post. While it is not required in this discussion, feel free to bring in outside resources to support your answers. Outside resources include materials outside of the textbook, such as a website.

• Identify which types of nonparametric test(s) you would apply to the data mentioned in each of the following questions:
  • Is the self-reported height on dating website profiles equal to the median height of the general population?
  • Is the median self-reported salary on paid online dating sites greater than the median self-reported salary on free online dating sites?
  • Is the median age of dating-service users different for personal matchmaking services, paid online dating sites, and free online dating sites?
  • How does education level affect the number of profile likes on paid online dating sites and free online dating sites?
• Describe why you selected the specific nonparametric test for each question. Fully explain your answers.

Answer 1

Solution:

1) One sample sign test is to be used here as only one sample is given here that of the height on dating website which is to be tested against the target median value of height of general population,

2) Moods-median test is to be used here as there are two independent samples here and we are to test the median of both these samples namely salary on paid dating sites and salary on free dating sites.

3) Kruskal-Wallis test is to be used here as there are more than two medians to be compared namely for personal matchmaking,paid online,free online.

4) Spearman rank correlation test is to be used here as we are asked to test for the association between education and dating sites both paid and free.

Please give upvote.

Thank you.