Introduction to Statistical Power
You'll learn what statistical power is and why it measures our ability to detect real effects.
What Is Statistical Power?
Statistical power is the probability that your hypothesis test will correctly reject a false null hypothesis. In other words, it's the likelihood that you'll detect an effect when one truly exists.
Remember from previous lessons that a Type II error (β) occurs when you fail to reject a false null hypothesis—essentially missing a real effect. Power is defined as:
Power = 1 - β
If β = 0.20 (a 20% chance of missing a true effect), then your power is 0.80 or 80%.
Understanding Power Through an Analogy
Think of statistical power like the sensitivity of a metal detector. A powerful detector (high power) reliably beeps when metal is actually present underground. A weak detector (low power) often fails to beep even when treasure is buried right beneath it. The detector's sensitivity is its "power"—its ability to correctly identify when something real exists.
Why Power Matters
When designing a study, you want high power—typically 80% or higher. This means:
- If there's a real difference or effect, you have an 80% chance of detecting it
- You're unlikely to waste resources on a study that misses important findings
- Your research is more credible and reproducible
Low power means you might conclude "no effect" when an effect actually exists—a costly mistake in medical trials, quality control, or policy decisions.
Key Takeaway: Statistical power (1-β) is the probability of correctly rejecting a false null hypothesis—your test's ability to detect real effects when they exist.