Sampling Distribution of a Proportion
What you'll learn: How sample proportions cluster around the true population proportion and when they follow a normal pattern.
The Core Idea
When you repeatedly draw random samples from a population and calculate the proportion each time (like the percentage of "yes" responses), those sample proportions won't all be identical—they'll vary. The sampling distribution of a proportion describes the pattern of this variation.
Just as the Central Limit Theorem tells us that sample means tend toward normality, sample proportions also tend to follow a normal distribution under the right conditions. This happens when your sample is large enough and the true proportion isn't too close to 0 or 1.
How It Works
Imagine a population where 60% support a policy (p = 0.60). If you poll 100 random people repeatedly:
- Sample 1 might show 58% support
- Sample 2 might show 63% support
- Sample 3 might show 61% support
These sample proportions (p̂) will cluster around the true value (0.60) in a bell-shaped pattern. The sampling distribution centers at the population proportion p, with spread determined by the standard error: √[p(1-p)/n].
Normal Approximation Rule of Thumb
The normal approximation works well when both np ≥ 10 and n(1-p) ≥ 10. This ensures you have enough "successes" and "failures" in your expected sample.
Key Takeaway: Sample proportions vary predictably around the true population proportion, and with sufficient sample size, this variation follows a normal distribution—letting us make precise probability statements about how close our sample proportion likely is to the truth.