Type I Error: False Positives
You'll learn what a Type I error is and how it's controlled by the significance level α.
What Is a Type I Error?
In hypothesis testing, a Type I error occurs when we reject the null hypothesis even though it's actually true. In other words, we claim to have found a significant effect or difference when, in reality, none exists—a false positive.
Think of it like a fire alarm going off when there's no fire. The alarm system "rejects" the hypothesis that everything is normal, but it's wrong.
The Connection to Significance Level α
The significance level α (alpha) is the probability threshold we set before conducting a test. It directly represents the maximum probability of committing a Type I error that we're willing to tolerate.
When we say "we test at the α = 0.05 level," we're saying: "If the null hypothesis is actually true, we accept a 5% chance of mistakenly rejecting it."
Formula:
P(Type I Error) = P(Reject H₀ | H₀ is true) = α
Why It Matters
- Lower α = stricter test: Setting α = 0.01 instead of 0.05 reduces false positives but makes it harder to detect real effects.
- Trade-off: Reducing Type I errors often increases Type II errors (failing to detect real effects).
- Real consequences: In medicine, a false positive might mean prescribing unnecessary treatment; in quality control, it might mean discarding good products.
The critical region and p-value approach both fundamentally control the Type I error rate at your chosen α.
Key Takeaway: A Type I error is rejecting a true null hypothesis (false positive), and its probability equals the significance level α you choose for your test.