Quick Summary
This guide focuses on the final and most critical step of a significance test: making a conclusion. You will learn how to interpret a P-value by comparing it to a significance level (α) to make a formal decision about the null hypothesis. By the end of this lesson, you will be able to write a complete, context-rich conclusion for a one-proportion z-test and correctly describe the potential errors associated with your decision.
Key Concepts
The "Conclude" step of a hypothesis test synthesizes all your work into a final, meaningful statement. It's where you answer the original research question using statistical evidence.
The P-value and the Significance Level (α)
The entire decision-making process hinges on comparing two numbers: the P-value and the significance level (α).
P-value: The probability of getting a sample result as extreme or more extreme than the one you observed, assuming the null hypothesis (H₀) is true. Think of it as the "probability of the weirdness" of your data if the null hypothesis were correct. A small P-value means your observed data is very surprising (unlikely) under the null hypothesis.
Significance Level (α): A pre-determined threshold for "weirdness." Before conducting the test, we decide how much evidence we require to be convinced. This value, α (alpha), is the probability of a Type I error we are willing to risk. Common values for α are 0.05, 0.01, and 0.10. If α is not stated in a problem, you should assume α = 0.05.
The Decision Rule
The rule for making a decision is simple and absolute. It is the core of this topic.
If the P-value < α: The observed result is statistically rare or "weird" enough to doubt the null hypothesis. The evidence is strong.
- Decision:Reject the null hypothesis (H₀).
If the P-value \ge α: The observed result is not surprising enough to doubt the null hypothesis. The evidence is not strong enough.
- Decision:Fail to reject the null hypothesis (H₀).
A helpful mnemonic: "If the P is low, the null must go. If the P is high, the null will fly."
Writing a Complete Conclusion
A high-scoring conclusion on the AP exam always has two parts, written in the context of the problem:
The Statistical Justification: Explicitly compare the P-value to the significance level (α) to justify your decision to either reject or fail to reject the null hypothesis.
The Contextual Conclusion: State what your decision means in relation to the alternative hypothesis (Hₐ) and the original question. Do you have convincing evidence for Hₐ, or not?
Example Structure:
"Because our P-value of [value] is [less than / greater than or equal to] our significance level of α = [value], we [reject / fail to reject] the null hypothesis. We [have / do not have] convincing evidence to conclude that [state the alternative hypothesis in words]."
"Failing to Reject" vs. "Accepting" H₀
This is a critical distinction. We never "accept" the null hypothesis. Statistical tests are designed to measure the strength of evidence against H₀.
Failing to reject H₀ means we did not find sufficient evidence to conclude Hₐ is true. It's like a jury delivering a "not guilty" verdict. This doesn't mean the defendant is proven innocent; it just means the prosecution didn't provide enough evidence to convict "beyond a reasonable doubt."
"Accepting H₀" would imply we have proven H₀ is true, which a significance test cannot do. The absence of evidence is not evidence of absence.
Statistical Significance
When we get a P-value less than our significance level (α) and reject the null hypothesis, we say the results are statistically significant. This simply means that the observed difference is too large to be plausibly explained by random chance alone.
Type I and Type II Errors
Every decision we make comes with the risk of being wrong. In hypothesis testing, there are two specific types of errors we can make.
[Image: A 2x2 grid showing the two possible truths (H₀ true, Hₐ true) vs. the two possible decisions (Reject H₀, Fail to Reject H₀), with cells labeled Correct Decision, Type I Error, Type II Error, and Correct Decision (Power).]
Type I Error:
Definition: We reject H₀ when H₀ was actually true.
Analogy: A false alarm. The fire alarm goes off, but there is no fire. You conclude a new drug is effective when it actually isn't.
Probability: The probability of a Type I error is equal to the significance level (α). If α = 0.05, you are accepting a 5% chance of making this type of error.
Type II Error:
Definition: We fail to reject H₀ when H₀ was actually false (meaning Hₐ was true).
Analogy: A missed opportunity. There really is a fire, but the alarm fails to go off. You fail to find evidence that a new drug is effective when it actually is.
Probability: The probability of a Type II error is denoted by β (beta). Calculating β is beyond the scope of the AP exam, but you must understand the concept.
Key Vocabulary
P-value: The probability of observing a sample statistic as extreme or more extreme than the one obtained, given that the null hypothesis is true.
Significance Level (α): The pre-determined threshold for rejecting the null hypothesis. It represents the maximum acceptable probability of making a Type I error.
Statistically Significant: A result of a study is deemed statistically significant if the calculated P-value is less than the chosen significance level (α), leading to the rejection of the null hypothesis.
Type I Error: The error made when a true null hypothesis is rejected. The probability of this error is α.
Type II Error: The error made when a false null hypothesis is not rejected. The probability of this error is β.
Fail to Reject the Null Hypothesis: The decision reached when the P-value is greater than or equal to the significance level (α). It indicates that there is not enough statistical evidence to support the alternative hypothesis.
Calculator Tech (TI-84)
No major calculator functions are required for this specific topic. The P-value would have been calculated in the previous step (the "Do" phase) using a function like . This topic focuses on the interpretation of that P-value.
How to Show Work on the FRQ
This topic is the "Conclude" part of the four-step State-Plan-Do-Conclude inference process. To earn full credit on an FRQ, your conclusion must be thorough and well-structured. Use the following template.
The "Conclude" Template
(Decision & Justification): "Because our P-value of [insert P-value] is [less than / greater than or equal to] our significance level of α = [insert alpha], we [reject / fail to reject] the null hypothesis (H₀)."
(Conclusion in Context): "We [have / do not have] convincing statistical evidence to conclude that [re-state the alternative hypothesis, Hₐ, in the words of the problem]."
Template for Describing a Potential Error
When asked to describe a Type I or Type II error, you must do so in the context of the problem.
Identify the Error Type: Based on your decision (Reject H₀ or Fail to Reject H₀), identify which error was possible.
If you rejected H₀, you could have made a Type I Error.
If you failed to reject H₀, you could have made a Type II Error.
Describe the Error in Context:
Type I Error Template: "A Type I error in this context would be concluding that [state Hₐ in words] when, in reality, [state H₀ in words]."
Type II Error Template: "A Type II error in this context would be concluding that we don't have enough evidence for [state Hₐ in words] when, in reality, [state Hₐ in words] is true."
Practice Problems
Problem 1:
A national polling organization claims that 40% of U.S. adults believe the country is on the right track. A local politician is concerned that the sentiment in her district is lower. She takes a random sample of 150 adults in her district and finds that 51 of them (34%) believe the country is on the right track. A significance test is conducted, yielding a P-value of 0.084. Using a significance level of α = 0.05, what conclusion should the politician make?
Solution:
Here, we apply the "Conclude" template.
H₀: p = 0.40 (The proportion of adults in the district who believe the country is on the right track is 40%).
Hₐ: p < 0.40 (The proportion... is less than 40%).
Conclusion:
Because our P-value of 0.084 is greater than our significance level of α = 0.05, we fail to reject the null hypothesis. We do not have convincing statistical evidence to conclude that the proportion of adults in this district who believe the country is on the right track is less than 40%.
Problem 2:
A pharmaceutical company has developed a new allergy medication. They claim that it is effective for more than 75% of patients. In a clinical trial, 185 out of 230 randomly selected patients reported relief. The researchers performed a significance test and calculated a P-value of 0.021.
(a) Using a significance level of α = 0.05, state the conclusion of the test.
(b) Based on your conclusion in part (a), what type of error could have been made? Describe this error in the context of the problem.
Solution:
(a) Conclusion:
H₀: p = 0.75 (The proportion of patients for whom the medication is effective is 75%).
Hₐ: p > 0.75 (The proportion... is greater than 75%).
Because our P-value of 0.021 is less than our significance level of α = 0.05, we reject the null hypothesis. We have convincing statistical evidence to conclude that the proportion of patients for whom the new allergy medication is effective is greater than 75%.
(b) Potential Error:
Since our decision was to reject the null hypothesis, the only possible error we could have made is a Type I Error.
A Type I error in this context would be concluding that the medication is effective for more than 75% of patients when, in reality, it is only effective for 75% (or less). A potential consequence is that the company markets and sells a drug that is not as effective as they claim.
Common Mistakes to Avoid
"Accepting the Null Hypothesis." This is the most common and critical error. You must always say "fail to reject H₀." Failing to find evidence against a claim is not the same as proving the claim is true.
Confusing the P-value and Alpha (α). The decision is always based on comparing the P-value to α. Do not compare the sample proportion (p̂) to α, or the test statistic (z) to α.
Writing a Conclusion Without Context. Simply stating "Reject H₀" is not enough. You must explain what rejecting H₀ means for the specific scenario you are investigating (e.g., what it means for the politician or the allergy medication). Always link your conclusion back to the alternative hypothesis.
Misinterpreting the P-value. Do not define the P-value as "the probability that H₀ is true." This is incorrect. The P-value is calculated assuming H₀ is true. It's the probability of your data (or more extreme data), not the probability of the hypothesis.
Stating a Definitive Conclusion. Avoid absolute language like "we have proven that..." Statistical conclusions are about the strength of evidence, not absolute proof. Use phrases like "we have convincing evidence to conclude..." or "we do not have convincing evidence to conclude..."