A two-sample t interval estimates the difference between two population means when you don't know the population standard deviations. You take the difference of sample means as your point estimate, then add and subtract a margin of error using the formula $(\bar{x}_1-\bar{x}_2) \pm t^* \sqrt{\frac{s_1^2}{n_1}+\frac{s_2^2}{n_2}}$ .

Why This Matters for the AP Statistics Exam

This topic shows up when a question gives you two independent samples and asks you to estimate how much two population means differ. You need to recognize that the right procedure is a two-sample t-interval (not a one-sample interval and not a paired interval), confirm the conditions, run the calculation, and explain what the interval means in context.

On the AP Statistics exam, free-response questions can ask you to choose the correct procedure, verify conditions, construct the interval, and interpret it. The reasoning here connects directly to later topics in Unit 7, where you justify claims from a two-mean interval and run two-sample t-tests. Clear notation and a correct interpretation are important for strong exam work.

more resources to help you study

practice multiple choice FRQ practice & scoring cheatsheets score calculator key terms

Key Takeaways

Use a two-sample t-interval when comparing two population means from two independent samples with unknown population standard deviations.
The point estimate is $\bar{x}_1 - \bar{x}_2$ , the difference of the two sample means.
The margin of error is $t^*\sqrt{\frac{s_1^2}{n_1}+\frac{s_2^2}{n_2}}$ , where $t^*$ comes from a t-distribution with degrees of freedom found using technology.
Check three conditions before calculating: independent random samples, the 10% condition when sampling without replacement, and approximate normality.
Choose "not pooled" on your calculator because you should not assume the two populations have equal variances.
A correct interpretation references both samples and describes the populations they came from.

Comparing Two Population Means

Sometimes you want to compare two different populations to see whether they differ. For example, you might compare the weights of two types of apples, green vs. red. Maybe you think one weighs more than the other, or maybe you just think they are different. One way to estimate that difference is to build a confidence interval for the difference of two population means.

This procedure is called a two-sample t-interval for a difference of means. It applies when you have one quantitative variable measured in two independent samples and you do not know the population standard deviations.

Conditions

As with any inference, check your conditions before doing any calculations.

(1) Random

Your samples must come from a randomized process, since you want to infer something about populations. With two populations, both samples must be random. If you are running an experiment to compare two treatments, verify that subjects were randomly assigned to treatments, not just randomly selected.

(2) Independent

Because you are usually sampling without replacement, check that the samples are independent. Use the 10% condition for both samples: $n_1 \leq 0.10 N_1$ and $n_2 \leq 0.10 N_2$ .

Note: For a randomized experiment, you do not need to check the 10% condition. Random assignment is enough to support independence.

(3) Normal

To treat the sampling distribution of $\bar{x}_1 - \bar{x}_2$ as approximately normal, you need both samples to support normality. You can do this by the Central Limit Theorem (both $n_1 > 30$ and $n_2 > 30$ ), by knowing both populations are normally distributed, or by checking that graphs of both samples (such as boxplots) show no strong skewness or obvious outliers.

Calculations

To build the interval, find the point estimate and the margin of error.

Point Estimate

The point estimate is your best guess for the difference between the two population means, based on the sample means. Subtract the two sample means:

$\bar{x}_{1}-\bar{x}_{2}$

Margin of Error

The margin of error is what you add and subtract from the point estimate to create the interval. The margin of error is the critical value times the standard error:

$t^*\sqrt{\frac{{s_{1}}^2}{n_{1}}+\frac{{s_{2}}^2}{n_{2}}}$

Here $\sqrt{\frac{s_1^2}{n_1}+\frac{s_2^2}{n_2}}$ is the standard error of the difference of two sample means, and $t^*$ is the critical value for the central C% of a t-distribution with degrees of freedom found using technology.

Full Interval

Putting it together, the confidence interval is:

$(\bar{x}_{1}-\bar{x}_{2}) \pm t^*\sqrt{\frac{{s_{1}}^2}{n_{1}}+\frac{{s_{2}}^2}{n_{2}}}$

These formulas are not printed directly on the formula sheet, but you can build them from the general structure of estimate ± (critical value)(standard error) using the standard error formulas that are provided.

Calculator Commands

A faster way to calculate the interval is to use technology such as a graphing calculator. On a TI-84, go into the STAT menu, scroll to TESTS, and select 2-SampTInt, then enter the given statistics to compute the interval.

Example

You have a bag of green apples and a bag of red apples and want to estimate the difference in the population mean weights. A sample of 30 green apples has a mean of 5 oz with a standard deviation of 0.2 oz. A sample of 30 red apples has a mean of 4.5 oz with a standard deviation of 0.15 oz. Construct and interpret a confidence interval for the difference in the two population mean weights.

Both samples are size 30, so the Central Limit Theorem supports approximate normality. Using technology gives the interval (0.408, 0.592).

Always select not pooled when doing two-sample intervals and tests. This is because you do not know whether the two populations have equal variances.

A solid interpretation might read: "We are 95% confident that the difference in the population mean weights (green minus red) is between 0.408 oz and 0.592 oz." Because both values are positive, the interval suggests green apples tend to weigh more than red apples in these populations.

How to Use This on the AP Statistics Exam

Free Response

If a question gives you two independent samples and asks for an estimate of the difference in means:

Name the procedure: a two-sample t-interval for a difference of means.
State and check the conditions: independent random samples (or random assignment), the 10% condition when sampling without replacement, and normality for both samples.
Show the structure of your calculation: point estimate $\bar{x}_1 - \bar{x}_2$ , then margin of error $t^*\sqrt{\frac{s_1^2}{n_1}+\frac{s_2^2}{n_2}}$ .
Report the interval and interpret it in context, referencing both samples and the populations.

Problem Solving

Be careful with the order of subtraction and stay consistent. The sign of the difference depends on which mean you call group 1.
Keep your units attached. An interval like (0.408, 0.592) is in ounces here.
Use technology for $t^*$ and the degrees of freedom. The degrees of freedom fall between the smaller of $n_1 - 1$ and $n_2 - 1$ and $n_1 + n_2 - 2$ .

Common Trap

If both endpoints of the interval are positive (or both negative), the data suggest a real difference in that direction. If the interval contains 0, then 0 is a plausible value for the difference, which means you do not have convincing evidence that the means differ.

Common Misconceptions

Mixing up two-sample and paired procedures. A two-sample t-interval is for two independent samples. If the data come in matched pairs, you analyze the differences within pairs instead, which is a different procedure.
Forgetting that both samples need checking. Conditions apply to each sample. For normality, both sample sizes should be over 30 if the distributions are skewed, not just one.
Pooling by default. On the AP exam, choose "not pooled" because you should not assume the populations have equal variances.
Misreading the confidence level. Being 95% confident is about the long-run success rate of the procedure across repeated samples. It does not mean there is a 95% probability that the specific calculated interval contains the true difference.
Interpreting the interval as being about sample means. The interval estimates the difference of the population means, not the difference of the sample means you already calculated.
Skipping the 10% condition logic. You only need the 10% condition when sampling without replacement. For a randomized experiment, random assignment handles independence.

Vocabulary

The following words are mentioned explicitly in the College Board Course and Exam Description for this topic.

Term	Definition
approximately normal	A distribution that closely follows the shape of a normal distribution, allowing for the use of normal probability methods.
confidence interval	A range of values, calculated from sample data, that is likely to contain the true population parameter with a specified level of confidence.
confidence interval procedure	A statistical method used to construct an interval estimate for a population parameter based on sample data.
critical value	A value from the standard normal distribution used to determine the margin of error for a given confidence level.
degrees of freedom	A parameter of the t-distribution that affects its shape; as degrees of freedom increase, the t-distribution approaches the normal distribution.
difference of population means	The difference between the mean values of two distinct populations, calculated as μ₁ - μ₂.
independence	The condition that observations in a sample are not influenced by each other, typically ensured through random sampling or randomized experiments.
independent samples	Two or more separate groups of data where the values in one group do not influence or depend on the values in another group.
margin of error	The amount by which a sample statistic is likely to vary from the corresponding population parameter, calculated as the critical value times the standard error.
normal distribution	A probability distribution that is mound-shaped and symmetric, characterized by a population mean (μ) and population standard deviation (σ).
population means	The average values of two distinct populations being compared, denoted as μ₁ and μ₂.
population standard deviations	The measure of spread in each of two populations; when unknown, sample standard deviations are used as estimates.
quantitative variable	A variable that is measured numerically and can take on a range of values, allowing for mathematical operations and statistical analysis.
randomized experiment	A study design where subjects are randomly assigned to treatment groups to establish cause-and-effect relationships.
sample mean	The average of all values in a sample, denoted as x̄, used as an estimate of the population mean.
sample standard deviations	The measures of variability within each of the two samples, denoted as s₁ and s₂.
sample statistic	A numerical value calculated from sample data that is used to estimate the corresponding population parameter.
sampling distribution	The probability distribution of a sample statistic (such as a sample proportion) obtained from repeated sampling of a population.
sampling without replacement	A sampling method in which an item selected from a population cannot be selected again in subsequent draws.
simple random sample	A sample selected from a population such that every possible sample of the same size has an equal chance of being chosen.
skewed distributions	Distributions that are not symmetric, with data concentrated on one side and a tail extending to the other side.
standard error	The standard deviation of a sampling distribution, which measures the variability of a sample statistic across repeated samples.
t-distribution	A probability distribution used when the population standard deviation is unknown and the sample standard deviation is used instead, characterized by heavier tails than the normal distribution.
two-sample t-interval	A confidence interval procedure used to estimate the difference between two population means using sample data from two independent samples.

Frequently Asked Questions

What is a confidence interval for the difference of two means?

It is an interval of plausible values for the difference between two population means, usually written as mu1 minus mu2. In AP Statistics, the procedure is a two-sample t-interval when the samples are independent and population standard deviations are unknown.

When do you use a two-sample t-interval?

Use a two-sample t-interval when you have one quantitative variable measured in two independent samples or groups and you want to estimate the difference between two population means.

What are the conditions for a two-sample t-interval?

Check that the two samples or groups are independent, that data come from random samples or a randomized experiment, that the 10% condition holds when sampling without replacement, and that the sampling distribution is approximately normal.

What is the formula for a two-sample t-interval?

The interval is point estimate plus or minus margin of error: (xbar1 - xbar2) plus or minus t* times sqrt(s1^2/n1 + s2^2/n2). Technology is usually used for degrees of freedom.

Should you pool standard deviations for AP Statistics two-sample intervals?

No. For AP Statistics, use the unpooled two-sample t-procedure unless a problem explicitly says otherwise. Calculator procedures usually label this as not pooled.

How do you interpret a confidence interval for two means?

State that you are confident the true difference in population means falls between the interval endpoints, and define the order of subtraction in context. The interpretation must reference the populations, not just the samples.