Module 06 Knowledge Check
5 questions. Pass mark 4 out of 5. Answer every question before checking the answer key below, then retry after reading the feedback.
1. Before-and-after measurements on the same people should be analysed…
- A. as two independent groups
- B. as paired differences
- C. with a chi-square test
- D. with ANOVA
2. Welch's t-test is R's default because it…
- A. is faster
- B. does not assume equal variances
- C. requires smaller samples
- D. gives smaller p-values
3. A significant chi-square test tells you…
- A. which cells drive the association
- B. that an association exists, and nothing about its direction
- C. the effect size
- D. that one variable causes the other
4. After a significant ANOVA, pairwise comparisons should be…
- A. run as ordinary t-tests
- B. adjusted, for example with Tukey's HSD
- C. skipped entirely
- D. run only on the largest groups
5. A non-significant result from 10 per group…
- A. shows there is no effect
- B. is close to uninformative, since a real 5-point effect would be missed 90% of the time
- C. proves the effect is small
- D. should be reported as a trend
---
Answer Key and Explanations
Check these only after attempting every question.
1. B — as paired differences
In the lesson the unpaired test gave p = 0.66 and the paired test p = 0.008 on identical numbers.
2. B — does not assume equal variances
With unequal spreads and unequal group sizes it reported 14.6 degrees of freedom where the equal-variance version claimed 73.
3. B — that an association exists, and nothing about its direction
The standardised residuals give the direction: premium at +3.04 on churn, basic at -2.72.
4. B — adjusted, for example with Tukey's HSD
Unadjusted pairwise p-values reintroduce the multiplicity the F-test was designed to control.
5. B — is close to uninformative, since a real 5-point effect would be missed 90% of the time
Power at n = 10 was 10%. The study was never capable of finding the effect.
Practical Check
Apply this module to your own work: complete the module activity for *Comparing Groups and Categorical Data*, then write one sentence naming what your result shows and one naming what it does not.
Strong Answer Pattern
A strong answer names the task, the evidence used, the check performed, and the remaining limitation. It avoids "proved", "guaranteed", or "always" unless the evidence genuinely supports it.