p-Values

AP Statistics· 12 min read

1. Formal Definition of the p-value★★☆☆☆⏱ 3 min

[Block type renderer not yet implemented · content will appear once shipped]

📘 Definition

p-value

P

The probability of observing a test statistic as extreme, or more extreme, than the one calculated from your sample data, if the null hypothesis is exactly true.

Example:

If your sample gives a z-score of +2.1 for a right-tailed test, the p-value is the area to the right of z=2.1 under the standard normal curve.

✓ Quick check

Test your understanding of the definition before moving on

Which of the following is a correct description of a p-value?
- The probability the null hypothesis is true
- The probability of getting a test statistic at least as extreme as observed, assuming H0 is true
- The probability the alternative hypothesis is true
- The probability your sample result will replicate
Reveal answer
The probability of getting a test statistic at least as extreme as observed, assuming H0 is true —
This matches the formal definition, and is the exact phrasing College Board accepts for full points.

2. Calculating p-values for Proportion Z-Tests★★★☆☆⏱ 4 min

[Block type renderer not yet implemented · content will appear once shipped]

z = \frac{\hat{p} - p_0}{\sqrt{\frac{p_0(1-p_0)}{n}}}

For a right-tailed alternative $H_{a} : p > p_{0}$ , the p-value is the area to the right of your calculated z-score
For a left-tailed alternative $H_{a} : p < p_{0}$ , the p-value is the area to the left of your calculated z-score
For a two-tailed alternative $H_{a} : p \neq = p_{0}$ , the p-value is double the tail area beyond the absolute value of your z-score

📐 Worked Example

A researcher tests whether the proportion of high school students who own a smartphone is different from 0.75, using a sample of 200 students where 162 own smartphones. Calculate the appropriate p-value.

1
First, identify the null proportion, sample proportion, and sample size:
2
$p_0 = 0.75, \hat{p} = 162/200 = 0.81, n = 200$
3
Calculate the z test statistic:
4
$z = \frac{0.81 - 0.75}{\sqrt{\frac{0.75 * 0.25}{200}}} \approx 1.96$
5
This is a two-tailed test, so find the area to the right of z=1.96 and double it:
6
$p-value = 2 * P(Z > 1.96) \approx 2 * 0.025 = 0.05$

Exam tip:

You can use your TI-84 normalcdf function to calculate p-values directly, but you must show your z-score calculation on FRQs to earn full credit.

3. Interpreting p-values in Context★★★☆☆⏱ 3 min

[Block type renderer not yet implemented · content will appear once shipped]

4. Drawing Conclusions Using p-values★★★☆☆⏱ 3 min

[Block type renderer not yet implemented · content will appear once shipped]

If p-value < $α$ : Reject the null hypothesis. There is statistically significant evidence to support the alternative hypothesis in context.
If p-value >= $α$ : Fail to reject the null hypothesis. There is not statistically significant evidence to support the alternative hypothesis in context.

📐 Worked Example

A test for a proportion has a p-value of 0.027, using a significance level of $α = 0.05$ . Write a valid conclusion for a test of whether more than 60% of voters support a ballot measure.

1
Compare the p-value to alpha:
2
$0.027 < 0.05$
3
State the formal decision about the null hypothesis:
4
We reject the null hypothesis that the true proportion of voters supporting the measure is 0.60.
5
State the contextual conclusion tied to the research question:
6
There is statistically significant evidence at the 0.05 level to conclude that more than 60% of voters support the ballot measure.

5. Common Pitfalls

Wrong move:

Claiming the p-value is the probability the null hypothesis is true

Why:

The p-value is calculated conditional on H0 being true, so it cannot quantify the probability H0 is true

Correct move:

Always state that you assume H0 is true when describing the p-value

Wrong move:

Forgetting to double the tail area for a two-tailed test

Why:

This will give you a p-value half the correct size, leading you to incorrectly reject H0

Correct move:

Check the direction of the alternative hypothesis before calculating the p-value

Wrong move:

Writing a conclusion that says 'we accept the null hypothesis'

Why:

Failing to reject H0 does not prove H0 is true, it only means you do not have enough evidence to reject it

Correct move:

Always use the exact phrasing 'fail to reject the null hypothesis'

Wrong move:

Interpreting a p-value without referencing the problem's specific context

Why:

AP rubrics deduct 100% of points for generic interpretations that do not name the parameter

Correct move:

Include the population, parameter, and sample result in every interpretation

Wrong move:

Using the sample proportion instead of the null hypothesis proportion when calculating standard error

Why:

This gives an incorrect z-score and wrong p-value that will not earn credit

Correct move:

Always use $p_{0}$ , the null value, for standard error calculation for proportion tests

6. Quick Reference Cheatsheet

Test Type	p-value Calculation	Decision Rule at α=0.05
Right-tailed $H_{a} : p > p_{0}$	$P (Z > z_{o b s})$	Reject H0 if p < 0.05
Left-tailed $H_{a} : p < p_{0}$	$P (Z < z_{o b s})$	Reject H0 if p < 0.05
Two-tailed $H_{a} : p \neq = p_{0}$	$2 * P (Z > ∣ z_{o b s} ∣)$	Reject H0 if p < 0.05

When this came up on past exams

AI-estimated based on syllabus patterns — cross-check with official past papers for accuracy. Use only as revision-focus signals.

2023 · FRQ 2
Interpret p-value for proportion test
2022 · MCQ Set 4
Calculate two-tailed p-value
2021 · FRQ 1
Use p-value to draw conclusion

What's Next

Mastering p-values is the foundation for all inference you will encounter for the rest of your AP Stats course, from two-proportion tests to t-tests for means, chi-square tests, and linear regression inference. The College Board weights p-value understanding extremely heavily, with at least one p-value question appearing on nearly every recent AP exam. Next, you will learn how to pair p-values with confidence intervals to draw consistent two-sided conclusions, before moving on to power and type I/type II error calculations that build directly on your understanding of hypothesis testing logic. Prioritize practicing full contextual interpretations to avoid losing easy points on FRQs.