📈 Section 4 · Statistics
🟣 Checkpoint Quiz
MODULE 23 · QUIZ
Section Quiz — Statistics & Probability
Your progress in Section 4100%
🎯 What this checkpoint covers: This is a review lesson, not new teaching. It's a 12-question assessment pulling from everything in Section 4 — Lessons 19 through 22: descriptive statistics, probability distributions, hypothesis testing and p-values, and correlation and regression. Skim the recap below, then take the quiz.
Recap
Section 4 in a Nutshell
Before the quiz, here's a compressed recap of the four lessons you've completed. If any of these points feel unfamiliar, it's worth a quick re-read of that lesson before you continue.
19
Lesson 19 — Descriptive Statistics Deep Dive
Mean, median, mode describe central tendency — the median is robust to skew and outliers, the mean isn't. Variance (s² = Σ(xᵢ − x̄)²/(n−1)) and standard deviation measure spread; NumPy's
ddof=0 default differs from pandas' ddof=1 default. The IQR outlier rule flags anything below Q1 − 1.5×IQR or above Q3 + 1.5×IQR. df.describe() computes all of this in one line.20
Lesson 20 — Probability Distributions
The normal distribution follows the 68-95-99.7 empirical rule for 1/2/3 standard deviations from the mean. The binomial distribution models counting successes across n independent yes/no trials. The uniform distribution makes every value in a range equally likely.
numpy.random draws samples; scipy.stats (e.g. norm.pdf, binom.pmf) computes exact theoretical probabilities.21
Lesson 21 — Hypothesis Testing & p-values
H₀ assumes no real effect; H₁ is the effect you suspect. A p-value is the probability of data this extreme IF H₀ were true — never the probability that H₀ itself is true. At α = 0.05, p ≤ α means reject H₀.
scipy.stats.ttest_ind() runs a two-sample t-test. A Type I error is a false positive (rejecting a true H₀); a Type II error is a false negative.22
Lesson 22 — Correlation & Regression Analysis
The Pearson correlation coefficient r (−1 to 1) measures linear relationship strength — but correlation is not causation; watch for confounding variables.
df.corr() plus sns.heatmap() visualizes correlations across many columns at once. Linear regression fits ŷ = mx + b via scipy.stats.linregress() or np.polyfit(); r² measures how much variance the line explains.🧩 Statistics & Probability Checkpoint — 12 Questions
Answer all 12 questions to test your mastery of Section 4. Instant feedback on every answer.
1. A dataset of hourly wages has a mean of $45 and a median of $22. What does this most likely indicate?
2. Why does sample variance in pandas (.var()) default to dividing by (n − 1) rather than n?
3. Using the standard IQR method, which value counts as an outlier?
4. Which pandas/NumPy call gives you count, mean, std, min, 25/50/75%, and max for every numeric column in one step?
5. According to the empirical rule, roughly what percentage of a normal distribution falls within 1 standard deviation of the mean?
6. Counting the number of successful free throws out of 15 independent attempts, each with the same success probability, is modeled by which distribution?
7. What does a p-value of 0.04 actually mean?
8. At the standard alpha = 0.05, a t-test returns p = 0.20. What's the correct conclusion?
9. Rejecting a null hypothesis that was actually true is called:
10. Two variables have a Pearson correlation of r = 0.95. What is the safest correct conclusion?
11. In the line of best fit ŷ = mx + b fit with scipy.stats.linregress(), what does r-squared (r²) represent?
12. Ice cream sales and drowning incidents rise together every summer, with a strong positive correlation. The best explanation is:
Finished the checkpoint?
Mark it complete to track your progress.
Module 23 of 23
Section 4 — Statistics & Probability for Data Science