AP Statistics · Topic 4.9 · Unit 4

AP Stats 4.9: Setting Up a Two-Sample t-Test

By Jude Wallis · Published July 28, 2026

Topic 4.9 sets up a two-sample t-test for a difference of means. The null hypothesis is mu-1 equals mu-2, equivalently mu-1 minus mu-2 equals 0; the alternative is one-sided or two-sided. Then verify the random, 10 percent, and normality conditions for both samples.

AP Statistics: Unit 4 (topics 4.9). Topic 4.9 (Setting Up a Test for the Difference Between Two Population Means) in the Fall 2026 AP Statistics course.

What topic 4.9 covers

Topic 4.9 is the setup stage of a significance test comparing two population means. You identify the method, name the parameters, write the hypotheses, and check conditions, but you do not compute the statistic yet; that is topic 4.10.

The correct method for two independent samples is a two-sample t-test for a difference between two population means. Name both populations and the response variable in context so the parameters $\mu_1$ and $\mu_2$ are unambiguous.

Define those parameters fully: reference the two population means, the response variable, and both populations in context. On the free-response section, parameters given only as bare symbols, or hypotheses written about sample means, lose credit because they describe the samples rather than the populations you want to compare.

Writing the hypotheses

The null hypothesis says the two population means are equal. It can be written two equivalent ways:

H_0: \mu_1 = \mu_2 \quad \text{or} \quad H_0: \mu_1 - \mu_2 = 0.

The alternative reflects the question and is also written either way:

$H_a: \mu_1 < \mu_2$ (equivalently $\mu_1 - \mu_2 < 0$ ),
$H_a: \mu_1 > \mu_2$ (equivalently $\mu_1 - \mu_2 > 0$ ),
$H_a: \mu_1 \ne \mu_2$ (equivalently $\mu_1 - \mu_2 \ne 0$ ).

Choose the direction from the research question before seeing the data, and keep the order of subtraction consistent with any interval. For more on picking the alternative, see null vs alternative hypothesis.

Verifying the conditions

A two-sample t-test requires the same three conditions as the two-sample interval, each in context:

Randomization: two independent random samples, or a randomized experiment.
10% condition: when sampling without replacement, $n_1 \le 0.10N_1$ and $n_2 \le 0.10N_2$ . This is unnecessary for a randomized experiment.
Sample data (normality): both sample sizes are at least 30, or both populations are stated to be approximately normal. If either sample size is under 30, both sample distributions should be free from strong skewness and outliers.

When the data come from a randomized experiment rather than two random samples, randomization still holds through the random assignment of treatments to experimental units, and the 10% condition is not needed. If you are choosing among procedures, which statistical test to use walks through the decision, and the two-sample t-test calculator runs the test once you carry it out.

Setting up a two-sample t-test

A teacher wants to know whether a new lesson method raises test scores compared with the standard method. Students are randomly assigned to the new method (group 1) or the standard method (group 2), with 36 students in each group. Identify the procedure, state the hypotheses, and verify the conditions.

The response variable, test score, is quantitative and two independent groups are compared, so use a two-sample t-test for a difference between two population means.
Define the parameters: let $\mu_1$ be the mean score for the new method and $\mu_2$ the mean score for the standard method.
State the hypotheses. The question asks whether the new method raises scores, so $H_0: \mu_1 = \mu_2$ and $H_a: \mu_1 > \mu_2$ .
Randomization: students were randomly assigned to the two methods, so this holds and the 10% condition is unnecessary for a randomized experiment.
Sample data: both groups have $n = 36 \ge 30$ , so the normality condition is satisfied.

Use a two-sample t-test for the difference in mean scores with $H_0: \mu_1 = \mu_2$ and $H_a: \mu_1 > \mu_2$ . Randomization holds and both samples exceed 30, so carrying out the test in topic 4.10 is justified.

Frequently asked questions

Are the two ways of writing the null hypothesis really the same?

Yes. Writing mu-1 equals mu-2 and writing mu-1 minus mu-2 equals 0 say the identical thing: the two population means are equal. Use whichever matches how you state the alternative and the parameter, and keep the order of subtraction consistent throughout the problem.

When would I use a matched-pairs test instead?

Use a matched-pairs test, which is a one-sample t-test on differences, when the two measurements are linked, such as before and after on the same subject or naturally paired units. Use a two-sample t-test when the two groups are independent, with no pairing between an observation in one group and one in the other.