POLS 3316: CLT & Correlation Interactive

POLS 3316: Statistics for Political Scientists

Interactive Lecture Module: Lecture 8

1. The Law of Large Numbers (LLN)

The LLN states that as the number of trials increases, the observed proportion of outcomes will converge on the theoretical probability. In a coin flip, we expect 50% Heads. In the short run (e.g., 5 flips), we might get 80% Heads. Watch how the bars level out as you increase the flips.

Total Flips: 0
Heads: 0
Tails: 0

2. The Central Limit Theorem (CLT)

This is the magic of statistics. Even if the population is uniform (like a single die roll, which is flat), the distribution of sample means becomes a Bell Curve (Normal) as the sample size increases.

Try increasing the Dice per Roll to see the shape change from flat to mound-shaped.

n=1 (Uniform) 1 n=30 (Normal)

If n=1, we plot the value of the die. If n=5, we roll 5 dice, average them, and plot the average.

100 500 5000

3. Correlation vs. Covariance

Covariance measures the direction of a relationship but is sensitive to scale (units). Correlation (Pearson’s r) is standardized between -1 and 1.

Adjust the “Relationship Strength” to tighten the dots (affecting Correlation). Adjust “Spread/Scale” to change the range of data (affecting Covariance).

Negative Zero Positive
Compact Spread Out
Pearson’s Correlation (r)
0.00
Bounded: [-1 to +1]
Covariance
0.00
Unbounded: depends on units

© 2026 POLS 3316 Course Materials. Created for Lecture 8.