Module 07 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 reporting a correlation you should always…
- A. compute r-squared
- B. plot the scatter
- C. test for significance
- D. remove outliers
2. A fitted trend line should be drawn…
- A. alone, for clarity
- B. with the points, and with r-squared reported
- C. only for strong correlations
- D. with a confidence band only
3. A strong overall correlation that vanishes within each group means…
- A. the groups are too small
- B. a third variable drives both
- C. the correlation was miscalculated
- D. the groups should be merged
4. Which claim does observational data support?
- A. Ad spend drives sales
- B. Ad spend explains 40% of the variation in sales
- C. Raising ad spend by 10 will raise sales by 7
- D. Sales depend on ad spend
5. Sampling points to fix overplotting…
- A. has no cost
- B. may drop the rare points, and should be stated
- C. is always wrong
- D. preserves density exactly
---
Answer Key and Explanations
Check these only after attempting every question.
1. B - plot the scatter
Curves and clusters both produce respectable coefficients and neither is a linear relationship.
2. B - with the points, and with r-squared reported
A clean line over a hidden cloud misrepresents its own evidence.
3. B - a third variable drives both
Colour the points by any grouping variable you have - it is a cheap check.
4. B - Ad spend explains 40% of the variation in sales
The others assert causation or the result of an intervention that was never run.
5. B - may drop the rare points, and should be stated
If outliers are the finding, sampling removes it.
Practical Check
Apply this module to your own work: complete the module activity for *Relationship Charts: Scatter, Regression and Grouped Patterns*, 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.
