Complex Concepts, Clarified

As we move to Chapters 4 and beyond, statistics deals with continuous chaos, complex conditions, and uncertainty. Here are the tools to tame them.

---

1. The Central Limit Theorem (Galton Board)

How does the bell curve arise from nature? The Galton Board (or Bean Machine) shows this physically. Balls drop through a grid of pegs. At each step, they randomly bounce left or right (Bernoulli trials).

By the time they hit the bottom, the piles form a Binomial Distribution. As row count increases, this becomes the Normal Distribution.

Watch the balls accumulate. Order emerges from random bounces.

---

2. Bayes’ Theorem (Tree Diagram)

Conditional probability is notorious for tripping up our intuition. “If a test is 99% accurate, and you test positive, do you have the disease?”

A Tree Diagram splits the world into paths, making the logic visible.

Scenario:

• Disease Rate: 1% ($P(D) = 0.01$)
• Test Accuracy: 90% ($P(+|D) = 0.9$)
• False Positive Rate: 10% ($P(+|D^c) = 0.1$)

Analysis: You can see there are TWO ways to get a “Positive” result.

1. Top path: $0.01 \times 0.9 = 0.009$ (True Positive)
2. Bottom path: $0.99 \times 0.1 = 0.099$ (False Positive)

Most positives are actually false alarms!

---

3. Sampling Variability (Confidence Intervals)

When we say “95% Confidence,” we don’t mean the true value has a 95% chance of being in our specific interval. We mean that if we repeated the experiment 100 times, 95 of our created intervals would capture the truth.

Look closely: The red intervals completely miss the dotted line (the Truth). This visualizes the risk inherent in any sampling method.