Scribe logoScribeNotes & practice
Sign inStart free
  1. Notes
  2. ›AP Statistics
  3. ›4 Inference for Quantitative Data: Means
  4. ›4.2 Interval for a mean or mean difference

AP Statistics · Unit 4 · Inference for Quantitative Data: Means

4.2 Interval for a mean or mean difference

Calculate and interpret a t confidence interval for a population mean, and identify when a t-distribution is the right model.

  • 4 min read
  • 5 parts
  • 10 questions
  • Draft
Start a Quick 5 on this topic→Jump to the 10 questions
1/5A confidence interval estimates a mean with a margin of error
  1. 1A confidence interval estimates a mean with a margin of error
  2. 2A t-distribution is used when the population standard deviation is unknown
  3. 3Three conditions must hold before a t interval is trustworthy
  4. 4A one-sample t interval combines the mean, critical value, and standard error
  5. 5A confidence level describes how reliably the method captures the mean
  6. Check your recall
  7. Practice: 10 questions
1

Part 1 of 5: A confidence interval estimates a mean with a margin of error

A random sample of commuters gives one average commute time. The true mean for all commuters is unknown. A different sample would give a different average.

The sample average is a point estimate. It is a single value used to estimate a parameter, and it is unlikely to equal μ\muμ exactly. So we add and subtract a cushion.

The margin of error is that cushion. A confidence interval is the point estimate plus and minus the margin of error, written xˉ±ME\bar{x}\pm\mathrm{ME}xˉ±ME. A wider cushion gives a wider range of values for μ\muμ.

Key terms · tap one for its definition

point estimate

A point estimate is a statistic, such as xˉ\bar{x}xˉ for μ\muμ, calculated from a sample to estimate a population parameter.

margin of error

The margin of error is the distance from the point estimate to either end of a confidence interval, in the units of the data.

confidence interval

A confidence interval is a range of values, built from sample data, that estimates a population parameter with a stated level of confidence.

Part 1 done

2

Part 2 of 5: A t-distribution is used when the population standard deviation is unknown

To find the margin of error, we need the spread of xˉ\bar{x}xˉ. That requires σ\sigmaσ, which is rarely known. We replace it with the sample standard deviation sss, and the result is the standard error sn\frac{s}{\sqrt{n}}n​s​.

Replacing σ\sigmaσ with sss adds extra uncertainty. So the standardized statistic follows a t-distribution, not a Normal one. It is symmetric and bell-shaped, with heavier tails.

Its exact shape depends on the degrees of freedom, df=n−1df=n-1df=n−1. Small samples give heavy tails. Large samples give a curve close to the Normal model.

Exam tip

Use sss and a t-distribution in nearly every real problem. Reserve σ\sigmaσ and the Normal model for the rare case where σ\sigmaσ is given.

Key terms · tap one for its definition

standard error

The standard error of the sample mean is sn\frac{s}{\sqrt{n}}n​s​, the estimated standard deviation of xˉ\bar{x}xˉ when σ\sigmaσ is unknown.

t-distribution

A t-distribution is a symmetric, bell-shaped distribution with heavier tails than the Normal model, determined by its degrees of freedom.

degrees of freedom

The degrees of freedom for a one-sample t procedure are df=n−1df=n-1df=n−1, which set the shape of the t-distribution.

Part 2 done

Quick check

One question on the part you just read

Easy

Identify the formula for a one-sample t confidence interval for a mean.
Show answer and explanation

Answer: A · Worked explanation

The interval is the point estimate plus and minus the critical value times the standard error, sn\frac{s}{\sqrt{n}}n​s​. Dividing sss by nnn instead of n\sqrt{n}n​ is a common slip. It makes the standard error far too small and the interval too narrow.
3

Part 3 of 5: Three conditions must hold before a t interval is trustworthy

A t interval works only when its model fits the data. Check three conditions first.

  • Random: the data come from a random sample or a randomized experiment.
  • Independent: sampling without replacement uses at most 10% of the population, so n≤0.10Nn\le0.10Nn≤0.10N, where NNN is the population size.
  • Normal: the population is Normal, n≥30n\ge30n≥30, or the sample data show no strong skew or outliers.

Paired data need one extra step. Subtract within each pair to get one difference per pair. Then treat the differences as a single sample, so xˉ\bar{x}xˉ is the mean of the differences and nnn counts pairs.

Common mistake

Two measurements on the same subjects are not independent groups. Use the differences, not two separate samples.

Part 3 done

4

Part 4 of 5: A one-sample t interval combines the mean, critical value, and standard error

The pieces now combine into one formula: xˉ±t∗sn\bar{x}\pm t^*\frac{s}{\sqrt{n}}xˉ±t∗n​s​. The critical value t∗t^*t∗ comes from a t table or calculator, using df=n−1df=n-1df=n−1 and the confidence level. Higher confidence gives a larger t∗t^*t∗. The product t∗snt^*\frac{s}{\sqrt{n}}t∗n​s​ is the margin of error.

Worked example

Build a 95% interval for a mean

A random sample of 16 commuters from a very large population has mean commute time xˉ=24.3\bar{x}=24.3xˉ=24.3 minutes and s=4.0s=4.0s=4.0 minutes. The dotplot shows no strong skew or outliers. Construct a 95% confidence interval for μ\muμ.
  1. 1
    Conditions: the sample is random, 16 is far below 10% of a very large population, and the data show no strong skew or outliers.
  2. 2
    Degrees of freedom: df=16−1=15df=16-1=15df=16−1=15, so t∗=2.131t^*=2.131t∗=2.131 for 95% confidence.
  3. 3
    Standard error: sn=4.016=1.0\frac{s}{\sqrt{n}}=\frac{4.0}{\sqrt{16}}=1.0n​s​=16​4.0​=1.0 minute.
  4. 4
    Interval: point estimate ±\pm± critical value ×\times× standard error, so 24.3±2.131×1.024.3\pm2.131\times1.024.3±2.131×1.0.
  5. 5
    Calculate: 24.3−2.131=22.16924.3-2.131=22.16924.3−2.131=22.169 and 24.3+2.131=26.43124.3+2.131=26.43124.3+2.131=26.431.

Answer: The 95% confidence interval is (22.2, 26.4)(22.2,\,26.4)(22.2,26.4) minutes, rounded.

Always check the conditions before computing. The numbers mean little if the model does not fit. Keep extra digits in the working, then round the final interval to one decimal place, matching the data.

Key term

critical value

The critical value t∗t^*t∗ is the number of standard errors, from a t-distribution with df=n−1df=n-1df=n−1, that captures the chosen central area.

Part 4 done

Quick check

One question on the part you just read

Easy

Identify the degrees of freedom for a one-sample t interval with n=20n=20n=20.
Show answer and explanation

Answer: C · Worked explanation

For a one-sample t procedure, the degrees of freedom are df=n−1=20−1=19df=n-1=20-1=19df=n−1=20−1=19. Using 20 forgets to subtract one. Using 18 subtracts too much, as if two parameters had been estimated from the single sample of 20 values, which is not the case here.
5

Part 5 of 5: A confidence level describes how reliably the method captures the mean

The interval above came from a method, so we judge the method. The confidence level tells us how often it succeeds. If we repeated the sampling many times, about 95% of the resulting 95% intervals would capture μ\muμ.

For the commute interval, write one sentence. We are 95% confident the interval from 22.2 to 26.4 minutes captures the mean commute time of all commuters.

Common mistake

A 95% level does not mean 95% of commutes fall inside the interval. It also does not mean μ\muμ has a 95% chance of lying in this one interval.

Two choices change the width. Higher confidence gives a larger t∗t^*t∗ and a wider interval. A larger nnn shrinks the standard error and narrows it.

Key term

confidence level

The confidence level is the long-run percentage of random samples of size nnn whose intervals capture the population mean.

Note read

Check your recall[12 marks]

5 written questions. Answer each one in your head or on paper, then open the mark scheme.

  1. 1.Calculate[3 marks]

    Calculate a 95% confidence interval for μ\muμ using n=25n=25n=25, xˉ=50.0\bar{x}=50.0xˉ=50.0, s=10.0s=10.0s=10.0, and t∗=2.064t^*=2.064t∗=2.064. Assume the conditions are met.

    Show mark schemeHide mark scheme

    Model answer

    Standard error =10.025=2.0=\frac{10.0}{\sqrt{25}}=2.0=25​10.0​=2.0. Margin of error =2.064×2.0=4.128=2.064\times2.0=4.128=2.064×2.0=4.128. Interval: 50.0±4.12850.0\pm4.12850.0±4.128, which is about (45.9, 54.1)(45.9,\,54.1)(45.9,54.1).

    • 1 mark for calculating the standard error as 2.0
    • 1 mark for multiplying by t∗t^*t∗ to get a margin of error of about 4.13
    • 1 mark for giving the interval as about 45.9 to 54.1
  2. 2.Identify[2 marks]

    Identify the degrees of freedom and the type of distribution used for a one-sample interval when n=25n=25n=25 and σ\sigmaσ is unknown.

    Show mark schemeHide mark scheme

    Model answer

    The distribution is a t-distribution with df=25−1=24df=25-1=24df=25−1=24.

    • 1 mark for identifying a t-distribution
    • 1 mark for giving df=24df=24df=24
  3. 3.Describe[2 marks]

    A 95% confidence interval for the mean commute time is 22.2 to 26.4 minutes. Describe what the 95% confidence level means.

    Show mark schemeHide mark scheme

    Model answer

    The method used to build the interval captures the true mean in about 95% of random samples. Repeating the sampling many times would give intervals that capture the mean about 95% of the time.

    • 1 mark for linking 95% to the long-run success rate of the method
    • 1 mark for referring to repeated random samples of the same size
  4. 4.Explain[2 marks]

    Explain why raising the confidence level from 95% to 99% makes an interval wider.

    Show mark schemeHide mark scheme

    Model answer

    A higher confidence level requires a larger critical value t∗t^*t∗. This increases the margin of error, so the interval is wider.

    • 1 mark for stating that a higher level needs a larger t∗t^*t∗
    • 1 mark for linking the larger t∗t^*t∗ to a larger margin of error and wider interval
  5. 5.Justify[3 marks]

    Justify using a t interval for a mean when a random sample of n=15n=15n=15 is taken from a population of 400 and its dotplot is roughly symmetric with no outliers.

    Show mark schemeHide mark scheme

    Model answer

    The sample is random. Fifteen is under 10% of 400, which is 40. The sample data show no strong skew or outliers, so the Normal condition is met.

    • 1 mark for stating that the sample is random
    • 1 mark for showing n=15n=15n=15 is at most 10% of 400
    • 1 mark for noting no strong skew or outliers for the Normal condition

Practice this topic

10 quick-fire multiple-choice questions, easy to hard. Tap an answer to see if you're right and the worked explanation. (The written questions with mark schemes are under “Check your recall” above.)

0 right0 of 10 answered

Includes the 2 quick checks in the note above.

3.

Easy·Identify[1 mark]

Identify when a t-distribution is used for an interval for a mean.
Show answer and explanation

Answer: C · Worked explanation

Replacing σ\sigmaσ with sss adds uncertainty, so the standardized statistic follows a t-distribution. This is true for any sample size, though large samples give a curve close to the Normal model. Sample size and skew affect whether the conditions are met, not whether the t-distribution is the correct model.
4.

Medium·Calculate[2 marks]

A random sample of n=36n=36n=36 has xˉ=80\bar{x}=80xˉ=80 and s=12s=12s=12, and the conditions are met. Using t∗=2.030t^*=2.030t∗=2.030, calculate the 95% confidence interval for μ\muμ.
Show answer and explanation

Answer: B · Worked explanation

The standard error is 1236=2.0\frac{12}{\sqrt{36}}=2.036​12​=2.0, so the margin of error is 2.030×2.0=4.062.030\times2.0=4.062.030×2.0=4.06. The interval is 80±4.0680\pm4.0680±4.06, which is 75.94 to 84.06. The interval 68.0 to 92.0 uses sss as the margin, and 78.0 to 82.0 omits t∗t^*t∗. The interval 79.3 to 80.7 divides sss by nnn.
5.

Medium·Determine[1 mark]

A t interval has standard error 2.0 and critical value t∗=2.262t^*=2.262t∗=2.262. Determine the margin of error.
Show answer and explanation

Answer: C · Worked explanation

The margin of error is the critical value times the standard error: 2.262×2.0=4.524≈4.522.262\times2.0=4.524\approx4.522.262×2.0=4.524≈4.52. Adding the two numbers gives 4.26, which is not how the formula works. The value 9.05 is the full width of the interval, double the margin of error.
6.

Medium·Identify[1 mark]

A 90% confidence interval for the mean daily screen time of all students at a school is 3.1 to 4.5 hours. Identify the correct interpretation.
Show answer and explanation

Answer: B · Worked explanation

The confidence level describes the success rate of the method at capturing the population mean μ\muμ. The interval is about μ\muμ, not about individual students. The sample mean is known and sits at the center, so it does not have a 90% chance of anything.
7.

Medium·Predict[1 mark]

A researcher quadruples the sample size, keeping the confidence level and the spread sss roughly the same. Predict the effect on the margin of error.
Show answer and explanation

Answer: A · Worked explanation

The standard error contains 1n\frac{1}{\sqrt{n}}n​1​, and 4=2\sqrt{4}=24​=2, so the margin of error roughly halves. It is not cut to a quarter, because the effect is the square root of the change in nnn. Larger samples do change the margin of error.
8.

Hard·Determine[2 marks]

Eight randomly chosen pairs give differences with mean 2.5 and standard deviation 1.2, with no strong skew or outliers. Using t∗=2.365t^*=2.365t∗=2.365, determine the 95% interval for the population mean difference.
Show answer and explanation

Answer: D · Worked explanation

Treat the 8 differences as one sample: 1.28≈0.424\frac{1.2}{\sqrt{8}}\approx0.4248​1.2​≈0.424 and 2.365×0.424≈1.002.365\times0.424\approx1.002.365×0.424≈1.00, giving 2.5±1.02.5\pm1.02.5±1.0. The interval 2.1 to 2.9 divides sss by nnn rather than n\sqrt{n}n​. The interval 1.7 to 3.3 uses a Normal critical value of 1.96. The interval -0.3 to 5.3 uses sss itself as the standard error.
9.

Hard·Justify[2 marks]

A random sample of n=12n=12n=12 from a very large population has a dotplot that is strongly right-skewed with one clear outlier. Justify whether a t interval for the mean is appropriate.
Show answer and explanation

Answer: D · Worked explanation

The Normal condition fails: nnn is below 30 and the sample data show strong skew and an outlier. Randomness does not fix shape. An unknown σ\sigmaσ is exactly why a t-distribution is used, so it is never a reason to avoid one.
10.

Hard·Determine[2 marks]

A 95% interval for μ\muμ from some data is 10.2 to 13.8. Determine which interval could be the 99% interval from the same sample.
Show answer and explanation

Answer: B · Worked explanation

The 95% interval has center 12.0 and margin of error 1.8. A 99% interval uses a larger t∗t^*t∗, so it has the same center and a larger margin of error. Only 9.5 to 14.5 is centered at 12.0 and wider. The interval 10.8 to 13.2 is narrower.

Keep going

Next topic4.3 Claims from a mean intervalPrevious topic4.1 Sample means

More in Unit 4 · Inference for Quantitative Data: Means

  1. 4.1Sample meansNotes · 10 questions
  2. 4.2Interval for a mean or mean differenceNotes · 10 questions
  3. 4.3Claims from a mean intervalNotes · 10 questions
  4. 4.4Setting up a test for a mean or mean differenceNotes · 10 questions
  5. 4.5Carrying out a test for a mean or mean differenceNotes · 10 questions
  6. 4.6Differences in sample meansNotes · 10 questions
  7. 4.7Interval for a difference of two meansNotes · 10 questions
  8. 4.8Claims about a difference of meansNotes · 10 questions
  9. 4.9Setting up a two-mean testNotes · 10 questions

All AP Statistics topics →

Draft: written by Scribe and checked automatically, not yet by a subject reviewer.

Practice questions are original, written by Scribe and checked automatically for correctness.

Topic codes follow the official numbering. Not endorsed by College Board or any exam board.

On this page

0/5 read

  1. 1A confidence interval estimates a mean with a margin of error
  2. 2A t-distribution is used when the population standard deviation is unknown
  3. 3Three conditions must hold before a t interval is trustworthy
  4. 4A one-sample t interval combines the mean, critical value, and standard error
  5. 5A confidence level describes how reliably the method captures the mean
  6. Check your recall
  7. Practice: 10 questions
All coursesScribeHow it worksPricingPrivacyTerms

Notes and questions are written by Scribe in our own words. Topic codes follow the official course numbering so you can find your place. Not produced or endorsed by the IB, College Board, AQA, OCR, Pearson Edexcel or any exam board.

© 2026 Scribe