Module 03 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. np.std(x) and pd.Series(x).std() differ because…
- A. numpy is less accurate
- B. their default ddof differs: 0 against 1
- C. pandas skips missing values
- D. numpy uses the median
2. On skewed money or duration data the more robust pair is…
- A. mean and standard deviation
- B. median and IQR
- C. mode and range
- D. mean and IQR
3. df.groupby('depot') on a frame where some rows have a missing depot…
- A. raises an error
- B. silently drops those rows
- C. puts them in their own group
- D. fills them with the mode
4. Two 95% confidence intervals that overlap…
- A. prove there is no difference
- B. are compatible with a clearly significant difference
- C. mean the test is invalid
- D. cannot both be correct
5. An honest statistical sentence contains…
- A. the word significant
- B. the estimate, the interval, the p-value and n
- C. the p-value alone
- D. the estimate alone
---
Answer Key and Explanations
Check these only after attempting every question.
1. B — their default ddof differs: 0 against 1
On 61 delivery times that gave 26.74 against 26.97, and the gap grows as n shrinks.
2. B — median and IQR
One value in 61 nearly halved the standard deviation while the median moved by 0.35.
3. B — silently drops those rows
82 rows in, 81 counted. dropna=False keeps them visible.
4. B — are compatible with a clearly significant difference
In the lesson the intervals overlapped while the difference interval was [0.87, 9.13] with p = 0.0186.
5. B — the estimate, the interval, the p-value and n
'Significant' replaces four pieces of information with none of them.
Practical Check
Apply this module to your own work: complete the module activity for *Descriptive Statistics and Statistical Visualisation*, 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.