Type II error in AP Statistics

In AP Statistics, a Type II error happens when a significance test does not find convincing evidence for the alternative hypothesis (Hₐ) even though Hₐ is actually true. In other words, you fail to reject a false null hypothesis. Its probability equals 1 − power.

What is Type II error?

A Type II error is a missed detection. The CED (3.8.A.2) defines it this way: there is not convincing evidence that Hₐ is true, but it is. Something real is going on, like a new drug that actually works or a process that actually drifted off spec. Your sample just didn't give you strong enough evidence to say so, so you failed to reject H₀.

Think of a smoke detector that stays quiet during a real fire. That's a Type II error. (A detector that goes off when you burn toast is the Type I error.) The probability of a Type II error is often written as β. The CED connects it directly to power, which is the probability that a test correctly rejects a false null hypothesis (3.8.A.3). So P(Type II error) = 1 − power (3.8.B.2). High power means a low chance of missing a real effect.

Why Type II error matters in AP® Statistics

Type II error lives in Unit 3: Inference for Categorical Data: Proportions, in Topic 3.8 Potential Errors When Performing Tests. It supports all four learning objectives in that topic. 3.8.A asks you to identify Type I and Type II errors. 3.8.B asks you to calculate their probabilities, where P(Type II) = 1 − power. 3.8.C asks you to identify what changes that probability: larger sample size, smaller standard error, a true parameter farther from the null, or a larger α all lower it. 3.8.D asks you to interpret the errors and weigh their consequences. The idea doesn't stay in Unit 3, though. Every significance test you run afterward, including tests for means in Unit 4, can make a Type II error. So the reasoning you learn here applies to the rest of the inference half of the course.

How Type II error connects across the course

Type I Error (Unit 3)

Type I and Type II errors are mirror images. A Type I error is a false alarm, where you reject H₀ when it's true. A Type II error is a missed alarm, where you fail to reject H₀ when it's false. You can only make one of them in any single test. If you rejected H₀, the only possible mistake is Type I. If you didn't, the only possible mistake is Type II.

Power of a Test (Units 3-4)

Power and Type II error are two sides of the same coin. When H₀ is false, the test either catches it (power) or misses it (Type II error). That's why P(Type II error) = 1 − power. Anything that boosts power automatically shrinks your chance of a Type II error.

Significance Level and the Tradeoff with α (Unit 3)

P(Type I error) = α, and raising α makes it easier to reject H₀, which increases power and lowers P(Type II error). So the two errors pull against each other when sample size is fixed. That's why 3.8.D says the consequences of a Type I error should drive your choice of α.

Complement Rule and Probability (Unit 2)

The formula P(Type II error) = 1 − power is just the complement rule from Unit 2. Given that H₀ is false, there are only two outcomes, correctly rejecting or failing to reject. Their probabilities have to add to 1.

Is Type II error on the AP® Statistics exam?

Type II error shows up in both multiple choice and free response, and context is everything.

Multiple choice usually asks you to do one of three things. First, pick which scenario is a Type II error. For example, with H₀: μ = 100 and Hₐ: μ > 100, the Type II error is concluding there's not convincing evidence that μ > 100 when the true mean really is greater than 100. Second, reason about how α relates to each error. Choosing α = 0.01 lowers the chance of a Type I error but raises the chance of a Type II error. Third, calculate β, as in a question where H₀: μ = 50, Hₐ: μ > 50, α = 0.05, and the true mean is actually 53.

Free response questions that used this term include 2018 FRQ Q6 (systolic blood pressure) and 2021 FRQ Q4 (a pet supply company offering $10 off repeat purchases). The skill being tested is writing the error in context. Name the parameter, say there was not convincing evidence for Hₐ, and say Hₐ is actually true. A generic line like "failing to reject a false null" usually won't earn full credit. Be ready to explain a real-world consequence too, like the manager not adopting a discount that actually would have increased repeat purchases.

Type II error vs Type I Error

Ask yourself what the test decided. A Type I error means you rejected H₀ but H₀ was actually true. That's a false positive, and its probability is α. A Type II error means you failed to reject H₀ but Hₐ was actually true. That's a false negative, and its probability is 1 − power. A quick memory trick is that Type I is "crying wolf" when there's no wolf, and Type II is missing the wolf that's really there.

Key things to remember about Type II error

• A Type II error happens when a test does not find convincing evidence for Hₐ, but Hₐ is actually true.

• The probability of a Type II error equals 1 minus the power of the test.

• You lower the chance of a Type II error by increasing sample size, decreasing standard error, increasing α, or having a true parameter farther from the null value.

• Lowering α to protect against a Type I error makes a Type II error more likely if the sample size stays the same.

• On FRQs, describe a Type II error in the context of the problem, naming the parameter and what conclusion was wrongly missed.

• The consequences of a Type II error help decide how large a sample a study needs, because sample size controls power.

Frequently asked questions about Type II error

What is a Type II error in AP Statistics?

A Type II error occurs when a significance test fails to find convincing evidence for the alternative hypothesis even though the alternative is actually true. You failed to reject a false H₀, and the probability of this happening is 1 − power.

How is a Type II error different from a Type I error?

A Type I error is rejecting H₀ when it's true, a false positive with probability α. A Type II error is failing to reject H₀ when it's false, a false negative with probability 1 − power. Which one you could have made depends on what your test decided.

Does a smaller significance level reduce Type II errors?

No, it does the opposite. Dropping α from 0.05 to 0.01 makes it harder to reject H₀, so power goes down and the chance of a Type II error goes up, unless you also increase the sample size.

How do you calculate the probability of a Type II error?

Use P(Type II error) = 1 − power. To find power, you need a specific true parameter value, such as a true mean of 53 when H₀ says μ = 50. Then you find the probability that the test rejects H₀ when that value is true.

Is Type II error on the AP Stats exam?

Yes. It's part of Topic 3.8 in Unit 3 and appears in multiple choice and FRQs, including 2018 FRQ Q6 and 2021 FRQ Q4. You'll need to identify it, describe it in context, and explain what factors make it more or less likely.