Sampling Distributions
What you'll learn: How statistics calculated from samples themselves form predictable probability distributions.
The Core Idea
Imagine you're studying the average height of all adults in a city (the population). You can't measure everyone, so you take a random sample of 50 people and calculate their average height. If you repeat this process—take another random sample of 50, then another, and another—each sample will give you a slightly different average.
A sampling distribution is the probability distribution of these sample statistics (like the mean, variance, or standard deviation) across many possible samples.
Why This Matters
When you compute a statistic from one sample (say, sample mean), you need to know: "How reliable is this number? How much would it vary if I sampled differently?"
The sampling distribution answers this. It tells you:
- The expected value of your statistic across all possible samples
- The spread (how much the statistic varies between samples)
- The shape of the distribution (often normal, thanks to the Central Limit Theorem)
Example Without Code
Suppose the true population mean height is 170 cm. You take 1,000 different samples (each with 50 people) and compute 1,000 sample means. These means might range from 168 cm to 172 cm, clustering around 170 cm. The distribution of these 1,000 means is the sampling distribution of the sample mean.
The standard error (the standard deviation of the sampling distribution) tells you how precise your sample mean is as an estimate of the population mean.
Key Takeaway: A sampling distribution describes how a statistic (like the mean) varies across repeated random samples, allowing you to quantify uncertainty in your estimates.