QM210 / QM310
P-Value Poker
Given a test statistic and context, estimate the p-value range and decide whether the result is statistically significant. Sharpen your hypothesis testing intuition.
How It Works
- You will face 10 hypothesis test scenarios with real business context
- Each round shows the test type (Z or T), test statistic, degrees of freedom, and tail direction
- Estimate which p-value band the result falls into (4 options)
- Then decide: is the result significant at the α = 0.05 level?
- Both correct: 10 pts. One correct: 5 pts. Neither: 0 pts.
- After answering, see the exact p-value and shaded rejection region
- Grade scale: A (90+), B (80+), C (70+), D (60+), F (below 60)
Challenge Complete!
You scored out of 100
Round-by-Round Summary
| # | Test | Exact p | Your Band | Your Sig. | Pts |
|---|
Common P-Value Misinterpretations
- A p-value is NOT the probability that the null hypothesis is true. It is the probability of observing data at least as extreme as what was collected, assuming H0 is true.
- A small p-value does not mean the effect is large or practically important. Statistical significance is not the same as practical significance.
- p = 0.05 is an arbitrary threshold, not a magic boundary. A result with p = 0.049 is not fundamentally different from p = 0.051.
- Failing to reject H0 (large p-value) does not prove H0 is true. It only means we lack sufficient evidence to reject it.
- P-values depend on sample size. With very large samples, trivially small effects can produce very small p-values.