Sampling Distribution of the Sample Proportion
What you'll learn: How sample proportions behave when you repeatedly sample from a population, and why this pattern is essential for understanding survey results and A/B tests.
The Core Idea
You've already learned about the sampling distribution of the sample mean. Now we'll look at proportions instead—like the percentage of voters who support a candidate, or the fraction of users who click an ad.
When you take a sample and calculate a proportion (like "40% prefer option A"), that's just one possible result. If you took another sample, you'd likely get a slightly different proportion—maybe 38% or 43%. The sampling distribution of the sample proportion describes the pattern of all possible sample proportions you could get from repeated sampling.
How It Works
Imagine a population where exactly 50% of people prefer tea over coffee. If you:
- Take a sample of 30 people and count the proportion who prefer tea
- Repeat this process hundreds of times
- Plot all those sample proportions
You'd see a distribution centered around 0.50 (the true population proportion), with variation around it. Smaller samples produce more spread-out distributions; larger samples cluster tighter around the true value.
Shape and Behavior
Just like sample means, sample proportions follow predictable patterns:
- The distribution centers on the true population proportion
- Larger sample sizes create narrower, more precise distributions
- Under certain conditions (sufficient sample size), this distribution approximates a Normal distribution
Key Takeaway: Sample proportions vary from sample to sample, but they follow a predictable pattern centered on the true population proportion—larger samples give more reliable estimates with less variability.