Sampling Distribution of the Test Statistic
You'll learn: How test statistics behave predictably under the null hypothesis, letting us calculate the probability of seeing our result by chance.
The Core Idea
When you calculate a test statistic (like a Z or t-value) from your sample, you're asking: "If the null hypothesis were actually true, how unusual is this number?"
Here's the key insight: if the null hypothesis is true, your test statistic isn't just a random number—it follows a known probability distribution. This distribution tells you what values are common versus rare when nothing special is happening in your population.
Think of it like a carnival game. If the game is fair (null hypothesis true), you know the distribution of possible scores. When someone gets an extremely high score, you can calculate: "What's the probability of scoring this high in a fair game?" A tiny probability suggests the game might not be fair after all.
Why This Matters
You've already learned that test statistics like Z and t summarize how far your sample result is from what the null hypothesis claims. Now you understand where the magic happens: these statistics follow distributions we can look up (standard normal for Z, t-distribution for t, etc.).
This is what makes hypothesis testing work! You can calculate a test statistic from your data, then use its sampling distribution to find the p-value—the probability of getting a result at least as extreme as yours if H₀ were true.
Key Takeaway: Under the null hypothesis, test statistics follow known distributions (Z ~ N(0,1), t ~ t-distribution, etc.), which allows us to calculate probabilities and make decisions about whether our data contradicts H₀.