Worksheets · AP Statistics · Unit 4

AP Statistics Unit 4 FRQ Practice: Inference for Means

Unit 4: Inference for Quantitative Data: Means

Three exam-style free-response questions on inference for means, each scored out of 10 points with a scoring guide and a printable PDF with room to write, free to use in class. Good practice before the Unit 4 Progress Check in AP Classroom. Multiple choice for this unit: Unit 4 worksheet.

Sigma Prep · sigmaprep.io/worksheets/ap-statistics

Name

AP Statistics Unit 4 Free Response: Inference for Means

Date

Directions: Show all your work. Indicate clearly the methods you use, because you will be scored on the correctness of your methods as well as on the accuracy and completeness of your results and explanations. Answer in the context of the question. Each question is worth 10 points, and each point is earned on its own. A graphing calculator with statistical capabilities may be used, and the exam’s formula sheet and tables apply.

Questions 1, 2 and 3 practice the kinds of questions in Questions 2, 3 and 4 of the exam’s free-response section. The exam’s Question 1 (data collection) carries little of Unit 4’s official free-response points, so the set has no question in its style.

Free response

Questions 1–3
  1. 1.
    The distribution of the length of visits to a downtown parking garage is skewed to the right, with mean 140 minutes and standard deviation 50 minutes. The garage has about 300,000 visits a year.
    1. (A)
      Describe the sampling distribution of the sample mean xˉ\bar{x} for random samples of 36 visits: give its mean, its standard deviation and its shape, and justify the shape.
    2. (B)
      (i) Find the probability that the mean length of a random sample of 36 visits is greater than 148.8 minutes. (ii) Would a sample mean above 148.8 minutes be unusual? Explain.
    1. (C)
      How would the standard deviation of xˉ\bar{x} change if the samples had 144 visits instead? Explain.
    2. (D)
      Two independent random samples of visits will be selected: the first sample has 30 visits and the second has 64. Let xˉ1\bar{x}_1 and xˉ2\bar{x}_2 be their sample means. (i) Describe the sampling distribution of xˉ1−xˉ2\bar{x}_1 - \bar{x}_2: give its mean, its standard deviation and its shape. (ii) Find the probability that xˉ1\bar{x}_1 is greater than xˉ2\bar{x}_2 by more than 9 minutes.
  2. 2.
    A baker tested a new yeast that is sold as making dough rise faster than the yeast she uses now. A total of 70 batches of dough were randomly assigned to the new yeast (30) or the current yeast (40). The rising time, in minutes, of the batches with the new yeast had mean 55.9 and standard deviation 6.5; the batches with the current yeast had mean 59.8 and standard deviation 5.4. The baker wants to know whether the mean rising time of dough made with the new yeast is less than the mean rising time of dough made with the current yeast.
    1. (A)
      Identify the appropriate inference procedure for investigating the question.
    2. (B)
      (i) State the null and alternative hypotheses. (ii) Define the parameters.
    1. (C)
      Verify the conditions for the test.
    2. (D)
      Calculate the test statistic, with its degrees of freedom, and the P-value.
    1. (E)
      What conclusion should be made at the α=0.05\alpha = 0.05 level? Justify your answer.
  3. 3.
    In a study of a cooling vest, 15 cyclists volunteered. On two days a week apart, each cyclist rode a stationary bike in a hot room until exhausted, once wearing the vest and once without it, in a random order. The times to exhaustion, in minutes, are shown in the table. The computer output below is for a paired tt-test on the differences (vest minus no vest).
    {"cols":["No vest","Vest"],"pairs":[[43.7,47.8],[29.2,33.3],[34.9,37.7],[42.9,41.2],[41.6,40.5],[39.9,42.2],[30.7,31.6],[37.3,40],[36.6,38.4],[33.7,33.4],[31.2,30],[36.2,36.4],[33.6,33.5],[31.2,30.9],[31,30.5]],"diff":false}CyclistNo vestVest143.747.8229.233.3334.937.7442.941.2541.640.5639.942.2730.731.6837.340.0936.638.41033.733.41131.230.01236.236.41333.633.51431.230.91531.030.5{"kind":"pairt","cols":["No vest","Vest"],"alt":"greater"}Paired T-Test: Vest, No vestSampleNMeanStDevSE MeanVest1536.495.201.342No vest1535.584.711.216Difference150.911.920.496T-Test of mean difference = 0 (vs > 0)T-Value = 1.84 P-Value = 0.0433 DF = 14
    1. (A)
      (i) Explain why the study had each cyclist do both conditions rather than randomly assigning half of the cyclists to each condition. (ii) Explain why the order of the two conditions was randomized for each cyclist.
    2. (B)
      (i) For how many of the 15 cyclists was the time to exhaustion with the cooling vest greater than with no vest? (ii) The computer output gives a mean difference (vest minus no vest) of 0.91. Interpret this value in context.
    1. (C)
      The computer output is for a test of whether the data give convincing evidence that, for cyclists like these volunteers, the mean time to exhaustion is greater with the cooling vest than with no vest. (i) State the null and alternative hypotheses, defining the parameter. (ii) Report the test statistic, its degrees of freedom and the P-value. (iii) What conclusion should be made at the α=0.10\alpha = 0.10 level? Justify your answer.
    2. (D)
      (i) Can it be concluded that the cooling vest caused the difference? Explain. (ii) To which cyclists can the results of this study be generalized? Explain.

      End of the free-response questions

Free AP Statistics worksheets at sigmaprep.io/worksheets/ap-statisticsOriginal questions written by Sigma Prep. AP is a registered trademark of the College Board, which is not affiliated with this sheet.

Scoring guide

Each point is earned on its own: a response earns a point when it meets that point’s requirement, whatever it earned on the others. The model answers are one complete response; “Also acceptable” names other responses that earn the same point.

Question 1: Sampling distributions of sample meansLike exam Question 2 (Practices 3 and 4: analyzing data and drawing conclusions) · 10 points

(A) 3 points

  1. Point 1. Correct mean of the sampling distribution. (Practice 3)

    Model answer: The mean of xˉ\bar{x} is μ=140\mu = 140 minutes.

  2. Point 2. Correct standard deviation, σn\frac{\sigma}{\sqrt{n}}. (Practice 3)

    Model answer: Its standard deviation is 5036=8.33\frac{50}{\sqrt{36}} = 8.33 minutes (the sample of 36 is less than 10% of the 300,000 visits).

  3. Point 3. Approximately normal, justified by the central limit theorem (n≥30n \ge 30) although the population is skewed. (Practice 4)

    Model answer: The population is skewed to the right, but n=36n = 36 is at least 30, so by the central limit theorem xˉ\bar{x} is approximately normal.

    Also acceptable: Any other correct, complete justification in context earns the point.

(B) 3 points

  1. Point 4. Standardizes the cutoff with the SD of xˉ\bar{x} (not σ\sigma). (Practice 3)

    Model answer: (i) z=148.8−1408.33=1.06z = \frac{148.8 - 140}{8.33} = 1.06.

    Also acceptable: The calculator route with no zz: normalcdf(148.8,∞,140,8.33)\text{normalcdf}(148.8, \infty, 140, 8.33), with the mean and the SD of xˉ\bar{x} (not σ\sigma), earns this point.

    Also acceptable: Carried-forward credit: a response that uses its own value from (A) (here 8.33) earns this point when its answer here is correct for that value.

  2. Point 5. Correct probability from the normal model for xˉ\bar{x}. (Practice 3)

    Model answer: P(xˉ>148.8)=normalcdf(148.8,∞,140,8.33)=0.1454P(\bar{x} > 148.8) = \text{normalcdf}(148.8, \infty, 140, 8.33) = 0.1454 (0.1455 from unrounded values).

    Also acceptable: The table route: z=1.06z = 1.06 (rounded to two places) gives 0.1446 from the standard normal table.

    Also acceptable: Either 0.1454, from the rounded values shown, or 0.1455, from unrounded values, earns the point.

    Also acceptable: Carried-forward credit: a response that uses its own value from (A) (here 8.33) earns this point when its answer here is correct for that value.

  3. Point 6. Judges unusualness by the probability (about 0.05 as the usual cutoff). (Practice 4)

    Model answer: (ii) No. A probability of 0.1454 (0.1455 from unrounded values) is not small, so such a sample mean would not be unusual.

    Also acceptable: Any clearly stated cutoff, applied correctly.

    Also acceptable: Any other correct, complete justification in context earns the point.

    Also acceptable: Carried-forward credit: a response that uses its own values from (B)(i) (here 0.1454 and 0.1455) earns this point when its answer here is correct for those values.

(C) 1 point

  1. Point 7. Half as large, because the SD is proportional to 1n\frac{1}{\sqrt{n}}. (Practice 4)

    Model answer: It would be half as large: σ144=12⋅σ36\frac{\sigma}{\sqrt{144}} = \frac{1}{2}\cdot\frac{\sigma}{\sqrt{36}}, because four times the sample size doubles n\sqrt{n}.

    Also acceptable: Any other correct, complete justification in context earns the point.

(D) 3 points

  1. Point 8. Correct mean of the difference, 0, since both samples come from the same population. (Practice 3)

    Model answer: (i) The mean is 140−140=0140 - 140 = 0 minutes.

    Also acceptable: Any other correct, complete description in context earns the point.

  2. Point 9. Correct SD (variances add) and shape (approximately normal, since both samples have at least 30). (Practice 3)

    Model answer: Its standard deviation is 50230+50264=11.063\sqrt{\frac{50^2}{30} + \frac{50^2}{64}} = 11.063 minutes, and it is approximately normal because both samples have at least 30 visits (central limit theorem), although the population is skewed to the right.

    Also acceptable: Any other correct, complete description in context earns the point.

  3. Point 10. Correct probability. (Practice 3)

    Model answer: (ii) P(xˉ1−xˉ2>9)=normalcdf(9,∞,0,11.063)=0.2080P(\bar{x}_1 - \bar{x}_2 > 9) = \text{normalcdf}(9, \infty, 0, 11.063) = 0.2080.

    Also acceptable: The table route: z=0.81z = 0.81 (rounded to two places) gives 0.2090 from the standard normal table.

    Also acceptable: Carried-forward credit: a response that uses its own values from (D)(i) (here 0 and 11.063) earns this point when its answer here is correct for those values.

Question 2: Comparing two meansLike exam Question 3 (Inference) · 10 points

(A) 1 point

  1. Point 1. Names a two-sample tt-test for a difference between two population means. (Practice 2)

    Model answer: A two-sample tt-test for a difference between two population means.

    Also acceptable: “2-SampTTest”, with the parameters made clear.

(B) 2 points

  1. Point 2. Correct hypotheses about μ1−μ2\mu_1 - \mu_2. (Practice 2)

    Model answer: (i) H0:μ1−μ2=0H_0: \mu_1 - \mu_2 = 0 and Ha:μ1−μ2<0H_a: \mu_1 - \mu_2 < 0.

    Also acceptable: H0:μ1=μ2H_0: \mu_1 = \mu_2 with the matching HaH_a.

    Also acceptable: The other order of subtraction (μ2−μ1\mu_2 - \mu_1, with the direction of HaH_a reversed), used consistently with the parameters.

  2. Point 3. Defines both parameters in context. (Practice 2)

    Model answer: (ii) μ1\mu_1 is the mean rising time of dough made with the new yeast and μ2\mu_2 is the mean rising time of dough made with the current yeast.

(C) 2 points

  1. Point 4. Random assignment stated. (Practice 2)

    Model answer: Random: the batches of dough were randomly assigned to the two treatments.

  2. Point 5. Normal/large sample: both sample sizes are at least 30. (Practice 2)

    Model answer: Normal/large sample: n1=30n_1 = 30 and n2=40n_2 = 40 are both at least 30.

(D) 3 points

  1. Point 6. Correct test statistic, unpooled. (Practice 3)

    Model answer: SE=6.5230+5.4240=1.46SE = \sqrt{\frac{6.5^2}{30} + \frac{5.4^2}{40}} = 1.46, so t=−2.67t = -2.67.

  2. Point 7. Correct degrees of freedom. (Practice 3)

    Model answer: df=55.70df = 55.70 from technology.

    Also acceptable: The conservative df=29df = 29, the smaller sample size minus 1.

  3. Point 8. Correct P-value. (Practice 3)

    Model answer: P-value =0.0050= 0.0050.

    Also acceptable: With the conservative df=29df = 29, P-value =0.0062= 0.0062; either earns the point.

    Also acceptable: The table route: t=−2.67t = -2.67 with the conservative df=29df = 29 gives 0.005 < P-value < 0.01 from the tt table.

    Also acceptable: Carried-forward credit: a response that uses its own results from Points 6 and 7 earns this point when its answer here is correct for those results.

(E) 2 points

  1. Point 9. Compares the P-value with α\alpha and makes the correct decision. (Practice 4)

    Model answer: Because the P-value of 0.0050 is less than α=0.05\alpha = 0.05, reject H0H_0.

    Also acceptable: Any other correct, complete justification in context earns the point.

    Also acceptable: Carried-forward credit: a response that uses its own value from (D) (here 0.0050) earns this point when its answer here is correct for that value.

  2. Point 10. Conclusion in context about the population means, in terms of HaH_a. (Practice 4)

    Model answer: There is convincing statistical evidence that the mean rising time of dough made with the new yeast is less than the mean rising time of dough made with the current yeast.

    Also acceptable: Carried-forward credit: a response that uses its own result from Point 9 earns this point when its answer here is correct for that result.

Question 3: A crossover experimentLike exam Question 4 (Practices 2, 3 and 4) · 10 points

(A) 2 points

  1. Point 1. Each cyclist is measured under both conditions, so the differences between cyclists (such as fitness) do not add variability to the comparison. (Practice 2)

    Model answer: (i) Each cyclist is measured under both conditions, so the comparison is made within each cyclist, and the large differences between cyclists in fitness do not add to the variability of the differences. This makes a real effect of the cooling vest easier to detect.

    Also acceptable: Pairing removes the cyclist-to-cyclist variability, so the comparison is more precise.

    Also acceptable: Any other correct, complete justification in context earns the point.

  2. Point 2. A random order keeps the effect of going first or second (practice, tiredness) from being confounded with the conditions. (Practice 2)

    Model answer: (ii) A cyclist may ride longer the second time because the test has become familiar. Randomizing the order balances such order effects between the two conditions, so they are not confounded with the effect of the cooling vest.

    Also acceptable: Any other correct, complete justification in context earns the point.

(B) 2 points

  1. Point 3. Correct count from the table (the differences that are positive). (Practice 3)

    Model answer: (i) Subtracting the “No vest” value from the “Vest” value for each cyclist, 8 of the 15 differences are positive, so 8 cyclists had a greater time to exhaustion with the cooling vest.

  2. Point 4. The mean difference as an average for these cyclists, with its direction and units, in context. (Practice 4)

    Model answer: (ii) On average, the time to exhaustion of these 15 cyclists was 0.91 minutes longer with the cooling vest than with no vest.

    Also acceptable: Stating it as the mean of the differences (vest minus no vest) for the cyclists in the study, in context.

    Also acceptable: Any other correct, complete interpretation in context earns the point.

(C) 4 points

  1. Point 5. Correct hypotheses about μd\mu_d, with μd\mu_d defined in context and the order of subtraction. (Practice 2)

    Model answer: (i) H0:μd=0H_0: \mu_d = 0 and Ha:μd>0H_a: \mu_d > 0, where μd\mu_d is the true mean difference in time to exhaustion (vest minus no vest) for cyclists like these volunteers.

  2. Point 6. Reads tt, the degrees of freedom and the P-value from the output. (Practice 3)

    Model answer: (ii) t=1.84t = 1.84 with df=14df = 14, and the P-value is 0.0433.

  3. Point 7. Compares the P-value with α\alpha and makes the correct decision. (Practice 4)

    Model answer: (iii) Because the P-value of 0.0433 is less than α=0.10\alpha = 0.10, reject H0H_0.

    Also acceptable: Any other correct, complete justification in context earns the point.

    Also acceptable: Carried-forward credit: a response that uses its own value from (C)(ii) (here 0.0433) earns this point when its answer here is correct for that value.

  4. Point 8. Conclusion in context about μd\mu_d, in terms of HaH_a. (Practice 4)

    Model answer: There is convincing statistical evidence that the true mean difference in time to exhaustion (vest minus no vest) for cyclists like these volunteers is greater than 0.

    Also acceptable: Carried-forward credit: a response that uses its own result from Point 7 earns this point when its answer here is correct for that result.

(D) 2 points

  1. Point 9. Yes: the conditions were imposed, in a random order, so a significant difference can be attributed to the condition. (Practice 4)

    Model answer: (i) Yes. This is an experiment: the cyclists were given both conditions in a random order, so a significant difference could be attributed to the cooling vest.

    Also acceptable: Any other correct, complete justification in context earns the point.

  2. Point 10. Only to units like those in the study, because they were volunteers. (Practice 4)

    Model answer: (ii) Only to cyclists like these volunteers: the cyclists volunteered and were not a random sample from a larger population.

    Also acceptable: Any other correct, complete justification in context earns the point.

What does the AP Statistics free response ask on Unit 4?

Unit 4 carries about a tenth of the AP Statistics free response: 102 of the 1,040 points on the 124 official free-response questions we counted (9.8%), and 6 of the 40 points in the four 2026 sample questions.

  • Question 3 is often a t procedure. On the 2027 exam, Question 3 asks for a significance test or a confidence interval, from the conditions to the conclusion. Of the 17 official free-response questions from 2012 to 2026 that fit Question 3, 5 were about means: 2 matched-pairs t tests (most recently in 2023), 2 two-sample t tests (most recently in 2026) and 1 one-sample t interval (2013).
  • Question 2 can ask about sample means. A part on the sampling distribution of x̄ (its mean, standard deviation or shape, or a probability about it) was in 4 official free-response questions from 2012 to 2026, most recently in 2026.
  • Question 4 can build on Unit 4. 5 official free-response questions from 2012 to 2026 that fit the Question 4 slot had a Unit 4 part, and the 2026 sample Question 4 asks for a two-sample t interval.
  • Every point is earned on its own. From 2027 each free-response question is worth 10 points, scored analytically: each point has its own requirement, and a later point does not depend on an earlier one.

Unit 4 has no data-collection question, so this set is practice for Questions 2, 3 and 4 of the exam. Every question and scoring guide is original.

Counted from the official AP Statistics free-response questions: the 2012–2026 exams, the practice exams and the 2026 course-description samples. No official question is reproduced here.

How often does the AP Statistics free response use each Unit 4 setting?

Settings of the questions on this set, and how many official AP Statistics free-response questions used each, 2012–2026
SettingQuestions on this setOfficial FRQs, 2012–2026Of them since 2019Last on the exam
Sampling distributions of sample meansSampling distributions · exam Question 21 of 3102015
Comparing two meansInference for means · exam Question 31 of 3732026
A crossover experimentQuestion 4 with a Unit 4 part · exam Question 41 of 3532025

Counts are of the 84 operational free-response questions, 2012–2026 (no 2020 exam was released), by the setting of the whole question. The Question 4 row counts questions that fit the exam’s new Question 4 and have a Unit 4 part.

Questions about this free-response set

Unit 4, Inference for Quantitative Data: Means, has 10 topics in the College Board's 2026 course and exam description: the sampling distribution of a sample mean; constructing a confidence interval for a population mean or a mean difference (paired data), with the conditions and the t distribution; using the interval to judge a claim; setting up and carrying out a one-sample or matched-pairs t-test; and the same for the difference of two population means (the sampling distribution of the difference, the two-sample t-interval and t-test). The course description weights the unit at 10–20% of the multiple-choice section.

Each of the four free-response questions is worth 10 points, scored analytically: every point has its own requirement (a variable named and classified, a sampling method identified, a comparison made in context with values), and each is earned on its own. The scoring guide here lists all 10 points of every question, with a model answer for each and the other responses that also earn it. The free-response section is half of the exam score.

The 2027 exam's Section II has four questions: Question 1 on Practices 1 and 2 (the investigative question and collecting data), Question 2 on Practices 3 and 4 (analyzing data and drawing conclusions), Question 3 on inference, and Question 4 on Practices 2, 3 and 4. The exam's Question 1 (data collection) carries little of Unit 4's official free-response points, so this set has three questions, in the style of Questions 2, 3 and 4.

Yes. A graphing calculator with statistical capabilities is allowed throughout the exam, and the exam supplies a formula sheet and tables. From 2027 the free-response answers are typed in Bluebook; this set prints with room to write so it can be done on paper in class.

We tagged every official free-response question from 2012 to 2026 (and the four 2026 course-description samples) by its setting and by the exam question it would fill from 2027. Each question here is in a setting the exam uses for that question: the sampling distribution of a sample mean, a two-sample t-test, and a crossover experiment read from paired computer output.

No. Every question here is original, written by Sigma Prep in the style of the exam: the settings, data and wording are our own, and no official question or scoring guideline is copied or reworded. Every number and every claim a scoring point makes (the method, the bias, the direction, the scope of a conclusion) was checked by an independent computer check before the set was published.

AP is a registered trademark of the College Board, which was not involved in making these questions and does not endorse them. All questions and scoring guides are original to Sigma Prep. Frozen 2026-11-20.