Interpreting P-Values Correctly
What you'll learn: You'll understand what p-values actually mean and avoid the most common mistakes people make when reading them.
What a P-Value Really Tells You
A p-value answers this question: "If the null hypothesis were true, how surprising would my data be?"
It measures how compatible your observed data is with the null hypothesis, nothing more. Think of it like this: imagine you're claiming a coin is fair (that's your null hypothesis). You flip it 10 times and get 9 heads. The p-value tells you how often you'd see results this extreme if the coin really were fair.
What P-Values Do NOT Tell You
Here's where most people go wrong:
- A p-value is NOT the probability that the null hypothesis is true. You can't say "there's a 5% chance H₀ is correct." The null hypothesis either is or isn't true—we just don't know which.
- A p-value is NOT the probability your alternative hypothesis is true. P-values say nothing about the probability of hypotheses being true or false.
- A low p-value doesn't measure effect size or importance. You can have p = 0.001 for a tiny, practically meaningless difference if your sample is large enough.
The Right Interpretation
When you see p = 0.03, say: "If the null hypothesis were true, we'd see data this extreme about 3% of the time. This is sufficiently rare that we have evidence against H₀."
Don't say: "There's a 97% chance the alternative hypothesis is true." That's a fundamental misinterpretation.
Key Takeaway: P-values measure how surprising your data is under the null hypothesis—they don't tell you the probability that any hypothesis is true or false.