Significance Levels and Critical Regions
You'll learn how to set thresholds for deciding when evidence is strong enough to reject a null hypothesis.
The Core Idea
When you perform a hypothesis test, you need a clear rule for deciding whether your test statistic is "extreme enough" to reject the null hypothesis. This is where significance levels and critical regions come in.
The significance level, denoted α (alpha), is the probability threshold you set before collecting data. It represents how much risk you're willing to take of rejecting a true null hypothesis (a Type I error). Common choices are α = 0.05, 0.01, or 0.10.
The critical region (or rejection region) is the set of test statistic values that would lead you to reject the null hypothesis. It's determined by your α level and the distribution of your test statistic under the null hypothesis.
How They Work Together
Think of a courtroom analogy: α is like setting the standard of proof ("beyond reasonable doubt"). The critical region is the range of evidence so extreme that it meets that standard.
For example, if you're testing a population mean with α = 0.05 using a two-tailed test:
- You find the critical values (like ±1.96 for a standard normal) that cut off the most extreme 5% of the distribution (2.5% in each tail)
- If your test statistic falls in these tails (the critical region), you reject H₀
- Statistical significance means your result landed in the critical region—it's unlikely enough under H₀ to warrant rejection
The p-value and α work hand-in-hand: reject H₀ when p-value < α, which is equivalent to the test statistic falling in the critical region.
Key Takeaway: The significance level α sets your tolerance for false positives, while the critical region translates that into specific test statistic values that trigger rejection of the null hypothesis.