Module 01 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. A statistical project needs an analysis-plan folder because…
- A. journals require it
- B. choices made after seeing the data are unfalsifiable
- C. it keeps the folder structure symmetric
- D. Python cannot import modules otherwise
2. A notebook that runs but fails on a fresh kernel usually indicates…
- A. a package version problem
- B. hidden state from a cell that no longer exists
- C. a corrupted file
- D. a random seed issue
3. np.random.default_rng(seed) is preferable to np.random.seed(seed) because…
- A. it is faster
- B. its state travels with the object rather than being process-wide
- C. it produces better random numbers
- D. it works with pandas
4. scipy and statsmodels differ mainly in that…
- A. scipy is faster
- B. statsmodels gives a model you can add covariates to
- C. scipy handles categorical data
- D. statsmodels has no p-values
5. A results table should contain…
- A. only the significant comparisons
- B. every planned comparison
- C. whichever metrics moved
- D. the primary outcome alone
---
Answer Key and Explanations
Check these only after attempting every question.
1. B — choices made after seeing the data are unfalsifiable
There is always a version of the test choice that gives the answer you hoped for. Dating the plan is what rules that out.
2. B — hidden state from a cell that no longer exists
Cells can run in any order, and a variable defined in a deleted cell is still in memory.
3. B — its state travels with the object rather than being process-wide
Legacy seeding sets global state that any library call can consume, shifting every result after it.
4. B — statsmodels gives a model you can add covariates to
For a two-group comparison they give identical numbers. Only the statsmodels version accepts + covariate as the next term.
5. B — every planned comparison
A reader who cannot see how many tests were run cannot judge any of them.
Practical Check
Apply this module to your own work: complete the module activity for *The Python Statistical Workflow*, 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.