Non-parametric testing is done when which of these are applicable? (Note: There are 3 correct answers).

Question

Options

• AWhen the traditional t tests don't produce the results we need
• BA Hypothesis Test for the Median of the population is in question
• CIt does not require data to come from Normally Distributed populations
• DThey look at the Median rather than the Mean of populations
• EWhen there are no parameters to measure in the process

• A
  6% (2)
• B
  79% (26)
• E
  15% (5)

Explanation

Non-parametric tests are used when assumptions of normality cannot be met (C), and they are specifically designed to test hypotheses about the median rather than the mean (B, D) - making all three of these statements accurate descriptions of when and why non-parametric methods apply. B and D are essentially two sides of the same coin: if your hypothesis concerns a population median, non-parametric tests (like the Mann-Whitney U or Wilcoxon signed-rank) are the appropriate tool.

Why A is wrong: Choosing a test because it gives you a preferred result is p-hacking/result manipulation - non-parametric tests are selected based on the nature of the data, not desired outcomes.

Why E is wrong: "Non-parametric" does not mean there are no parameters to measure. It means the test makes no assumptions about the distributional parameters (like μ or σ) of the population - the data still has measurable characteristics.

Memory tip: Think of "non-parametric" as "no normal distribution required." The key triggers are: skewed data, ordinal data, small samples, or median-focused hypotheses - any of these signals that non-parametric is the right route. If your question mentions "median," non-parametric is almost certainly involved.

No community discussion yet for this question.