Σ
SigmaPrep
AP Statistics worksheetsBlog
← All articles
AP StatisticsAP Exams

AP Statistics Review: The Most-Tested Questions, Unit by Unit

Σ
Sigma Prep
Math Instructor · 10+ Years Experience
November 2026·15 min read
Key takeaways
  • Unit weight: Unit 1 (one-variable data and collecting data) is about a third of both sections: 32.5% of the official multiple choice and 31.3% of the free-response points we counted. Unit 5 (regression) is the smallest on both.
  • Top types: the scope of inference and summary statistics have the most official multiple-choice questions (11 each); describing a scatterplot and comparing distributions are the most common parts of the free response.
  • The 2027 exam: 42 multiple-choice questions with four choices, some in sets, and 4 free-response questions of 10 points each. The course goes from 9 units to 5, and the geometric distribution, combining random variables, the chi-square goodness-of-fit test and inference for the slope are gone.

We sorted every public official AP Statistics question by what it actually asks: 406 multiple-choice questions on today’s course from 13 College Board sources (1997–2026) and 124 free-response questions worth 1,040 points (2012–2026). Each one is tagged to its unit and topic in the 2026 Course and Exam Description (the CED) and to a more specific question type. This post puts the five units side by side: how much each carries, the question types asked most across the whole course, what the 2027 exam looks like, how the old unit numbers map to the new ones, and what has left the course. Use it as an AP Statistics study guide: review the types in order, then practice each unit with its own review post and practice test.

AP Statistics exam format (2027)

The CED that takes effect in fall 2026 changes the exam. It is still 3 hours, with two sections worth half the score each: 42 multiple-choice questions with four choices, some in sets, and 4 free-response questions of 10 points each. A graphing calculator is allowed throughout both sections, and you get a formula sheet and tables.

The AP Statistics exam from May 2027: sections, questions and time
SectionWhat it isQuestionsTime
I. Multiple choiceFour choices (A–D); single questions and sets that share one stimulus4290 min
II. Question 1Practices 1 and 2: the investigative question and collecting data190 min
II. Question 2Practices 3 and 4: analyzing data and drawing conclusions1
II. Question 3Inference: a significance test or a confidence interval1
II. Question 4Practices 2, 3 and 4 across the course1
  • Multiple choice: 42 questions instead of 40, with four choices instead of five. Some come in sets: 6 of the 24 sample questions in the 2026 CED are in 2 sets.
  • Free response: 4 questions instead of 6, each worth 10 points and scored analytically: each point has its own requirement, earned on its own. There is no Investigative Task any more.
  • Question 1 is a study-design question. Of the operational free-response questions we counted, 3 from 2012–2018 fit it and 7 from 2019–2026, one on every exam since 2021. Their settings were experiments (5), sample surveys (3) and observational studies (2); the 2026 CED sample Question 1 is a sample survey.
  • The course: 9 units become 5 and 80 topics become 55. Unit 1 merges one-variable data with collecting data, Unit 2 gathers probability and the normal and sampling distributions, Units 3 and 4 are inference for proportions and for means, and Unit 5 is descriptive regression.

Which units are tested most

The CED gives each unit a weight range for the multiple-choice section only. Here is how the official questions compare, in both sections.

AP Statistics units: the 2026 course description’s multiple-choice range, the share of 406 official multiple-choice questions and the share of 1,040 official free-response points
UnitCED weight (MC)Official MCFRQ points
1. Exploring One-Variable Data and Collecting Data20–30%32.5%31.3%
2. Probability, Random Variables, and Probability Distributions15–25%20.9%20.6%
3. Inference for Categorical Data: Proportions15–25%21.2%18.7%
4. Inference for Quantitative Data: Means10–20%18.7%9.8%
5. Regression Analysis10–20%6.7%9.1%

“Official MC” is the share of the 406 multiple-choice questions on today’s course; “FRQ points” the share of the 1,040 points on the 124 free-response questions. The free-response shares add up to less than 100% because 108.6 points were on topics that have left the course.

  • Unit 1 is the biggest unit in both sections. 132 official multiple-choice questions, above the top of its CED range, and 326 free-response points.
  • Units 2 and 3 are about a fifth each of the multiple choice and the free response.
  • Unit 4 is a multiple-choice unit: 18.7% of the multiple choice but 9.8% of the free-response points.
  • Unit 5 is the smallest on the old exams, but the CED expects more: 6.7% of the official multiple choice against a CED range of 10–20%, and 3 of the 24 2026 sample questions.

Old AP Statistics unit numbers and the new units

Most AP Statistics reviews online, and most old released material, still use the 9 units of the 2020 course description. From fall 2026 there are 5, so an old “Unit 6 review” is today’s Unit 3. Old Units 1 and 3 are the new Unit 1, old Unit 4 is the core of Unit 2, old Units 6 and 7 are Units 3 and 4, and old Unit 9 (inference for slopes) is gone.

The 9 units of the 2020 AP Statistics course description (exams 2020–2026) and where their content is in the 5 units of the 2026 course description (exams from 2027)
Old unit (2020–2026)New unitWhere it went
1. Exploring One-Variable Data1 and 2Unit 1; the normal distribution moved to Unit 2.
2. Exploring Two-Variable Data2 and 5Two-way tables in Unit 2, scatterplots and regression in Unit 5; transformations and influential points were dropped.
3. Collecting Data1Unit 1, with one-variable data.
4. Probability, Random Variables, and Probability Distributions2Unit 2; combining random variables and the geometric distribution were dropped.
5. Sampling Distributions2, 3, 4The idea and the central limit theorem in Unit 2, each statistic’s sampling distribution with its inference in Units 3 and 4.
6. Inference for Categorical Data: Proportions3Unit 3.
7. Inference for Quantitative Data: Means4Unit 4.
8. Inference for Categorical Data: Chi-Square3Tests for homogeneity and independence in Unit 3; the goodness-of-fit test was dropped.
9. Inference for Quantitative Data: SlopesDroppedDropped from the course.

The 15 most-tested question types

Ranked by the number of official multiple-choice questions on today’s course (ties by the number of sources, then the six 2012–2019 practice exams), with how many of the 13 sources had one, how many were on the practice exams, and how many of the 84 operational free-response questions since 2012 had a part on it. These are every type with 8 or more official questions: 7 from Unit 1, 5 from Unit 2 and 3 from Unit 3.

1. Scope of inference: generalizing and cause and effect

Random selection lets you generalize to the population; random assignment lets you conclude cause and effect.

Unit 111 official questions on 9 of 13 sources5 on the 2012–2019 practice exams7 FRQs since 2012, 5 of them since 2019
Try it
A city health office selected a random sample of 500 employees of large companies in the city. It recorded whether each employee bikes to work and then tracked the number of sick days each employee took over the next year. The results showed that the employees who bike to work took fewer sick days on average, and the difference was statistically significant. Which of the following is the most appropriate conclusion?
  1. The results can be generalized to all employees of large companies in the city, because the employees were randomly selected from that population, and it can be concluded that the difference in number of sick days was caused by whether the employees bike to work, because the employees were randomly selected.
  2. The results can be generalized to all employees of large companies in the city, because the employees were randomly selected from that population, and it can be concluded that the difference in number of sick days was caused by whether the employees bike to work, because the difference was statistically significant.
  3. The results should be generalized only to employees like those in the study, because the employees were not randomly selected, and it cannot be concluded that the difference in number of sick days was caused by whether the employees bike to work, because nothing was randomly assigned.
  4. The results can be generalized to all employees of large companies in the city, because the employees were randomly selected from that population, and it cannot be concluded that the difference in number of sick days was caused by whether the employees bike to work, because nothing was randomly assigned.
Show solutionHide solution
Solution

(D) The results can be generalized to all employees of large companies in the city, because the employees were randomly selected from that population, and it cannot be concluded that the difference in number of sick days was caused by whether the employees bike to work, because nothing was randomly assigned. The employees were randomly selected from all employees of large companies in the city, so the results generalize to that population. But nobody was assigned to bike or not; the employees chose. Other differences between bikers and non-bikers could explain the sick days, so the study cannot show cause and effect.

Common mistake
mixing up the two kinds of randomness. (A) uses random selection to justify cause and effect, and (B) uses statistical significance; only random assignment of the treatment supports a causal conclusion.

Practice: Studies and Experiments worksheet · AP Statistics Unit 1 worksheet

2. Computing or reading summary statistics

Median, quartiles, mean and standard deviation from a list, a dotplot or a frequency table.

Unit 111 official questions on 7 of 13 sources8 on the 2012–2019 practice exams4 FRQs since 2012, 3 of them since 2019
Try it
A fish hatchery measured the lengths, in centimeters, of 11 trout from one of its ponds. The length values, in centimeters, are listed below. 33, 35, 21, 33, 30, 33, 29, 30, 35, 36, 32. What is the interquartile range (IQR) of the length values, in centimeters?
  1. 4.04.04.0
  2. 5.05.05.0
  3. 15.015.015.0
  4. 35.035.035.0
Show solutionHide solution
Solution

(B) 5.05.05.0. In order the lengths are 21, 29, 30, 30, 32, 33, 33, 33, 35, 35, 36. The median is the 6th value, 33. The lower half (the 5 values below it) has median Q1=30Q_1 = 30Q1​=30 and the upper half has median Q3=35Q_3 = 35Q3​=35, so the IQR is 35−30=535 - 30 = 535−30=5 centimeters.

Common mistake
putting the median into both halves when nnn is odd, which gives quartiles of 30 and 34 and an IQR of 4 (A). (C) is the range, 36−21=1536 - 21 = 1536−21=15, and (D) is Q3Q_3Q3​ itself.

Practice: Mean, Median, Mode and Range worksheet · AP Statistics Unit 1 worksheet

3. Identifying the sampling method

Simple random, stratified, cluster, systematic, convenience or voluntary response, from how the sample was chosen.

Unit 110 official questions on 10 of 13 sources, one of them a 2026 CED sample5 on the 2012–2019 practice exams

Practice: Sampling Methods and Bias worksheet · AP Statistics Unit 1 worksheet

4. Changing units or adding a constant

Adding a constant shifts the center but not the spread; multiplying by a constant multiplies both.

Unit 19 official questions on 8 of 13 sources, one of them a 2026 CED sample5 on the 2012–2019 practice exams

Practice: Standard Deviation worksheet · AP Statistics Unit 1 worksheet

5. Chi-square expected counts, degrees of freedom and statistic

expected count=(row total)(column total)table total\text{expected count} = \frac{(\text{row total})(\text{column total})}{\text{table total}}expected count=table total(row total)(column total)​ and df=(r−1)(c−1)df = (r - 1)(c - 1)df=(r−1)(c−1).

Unit 39 official questions on 8 of 13 sources5 on the 2012–2019 practice exams3 FRQs since 2012, none since 2019

Practice: Chi-Square Goodness of Fit worksheet · AP Statistics Unit 3 worksheet

6. z-scores and relative position

z=x−meanSDz = \frac{x - \text{mean}}{\text{SD}}z=SDx−mean​: how many standard deviations a value is from the mean, to compare values on different scales.

Unit 19 official questions on 8 of 13 sources3 on the 2012–2019 practice exams

Practice: Normal Distributions worksheet · AP Statistics Unit 1 worksheet

7. Inverse normal: a value from an area

A percentile or cutoff from invNorm, or a mean or standard deviation from a known percentile.

Unit 29 official questions on 7 of 13 sources5 on the 2012–2019 practice exams2 FRQs since 2012, 1 of them since 2019

Practice: Normal Distributions worksheet · AP Statistics Unit 2 worksheet

8. A one-proportion z-interval

p^±z∗p^(1−p^)n\hat{p} \pm z^*\sqrt{\frac{\hat{p}(1 - \hat{p})}{n}}p^​±z∗np^​(1−p^​)​​, after checking random sampling, the 10% condition and large counts.

Unit 39 official questions on 6 of 13 sources8 on the 2012–2019 practice exams3 FRQs since 2012, 1 of them since 2019
Try it
To estimate the proportion of all customers of the online clothing store who returned at least one item last year, a researcher selected a random sample of 440 customers. In the sample, 193 of the customers returned at least one item last year. (The store has about 250,000 customers.) Which of the following is a 95% confidence interval for the proportion of all customers of the online clothing store who returned at least one item last year?
  1. (0.378,0.500)(0.378, 0.500)(0.378,0.500)
  2. (0.384,0.494)(0.384, 0.494)(0.384,0.494)
  3. (0.392,0.485)(0.392, 0.485)(0.392,0.485)
  4. (0.415,0.462)(0.415, 0.462)(0.415,0.462)
Show solutionHide solution
Solution

(C) (0.392,0.485)(0.392, 0.485)(0.392,0.485). The sample proportion is p^=193440=0.439\hat{p} = \frac{193}{440} = 0.439p^​=440193​=0.439, and z∗=1.960z^* = 1.960z∗=1.960 for 95%. The standard error is 0.439(0.561)440=0.0237\sqrt{\frac{0.439(0.561)}{440}} = 0.02374400.439(0.561)​​=0.0237, so the margin of error is 1.960(0.0237)=0.0461.960(0.0237) = 0.0461.960(0.0237)=0.046 and the interval is 0.439±0.0460.439 \pm 0.0460.439±0.046, from 0.392 to 0.485.

Common mistake
using the wrong critical value. (A) and (B) are the same interval with z∗z^*z∗ for other confidence levels, and (D) leaves z∗z^*z∗ out, so its “margin” is just the standard error.

Practice: Confidence Intervals worksheet · AP Statistics Unit 3 worksheet

9. Joint, marginal and conditional relative frequencies

Read a two-way table: a joint share uses the grand total, a marginal share a row or column total over the grand total, and a conditional share the total of the group it is conditioned on.

Unit 28 official questions on 8 of 13 sources, one of them a 2026 CED sample4 on the 2012–2019 practice exams8 FRQs since 2012, 6 of them since 2019
Try it
A plant nursery grew 331 pepper plants of three varieties and recorded whether each plant produced fruit within 70 days. The table below shows the results. Of the Cayenne plants, what proportion produced fruit within 70 days?
{"rowVar":"Variety","colVar":"Fruit within 70 days","rows":["Ancho","Banana","Cayenne"],"cols":["Yes","No"],"counts":[[80,50],[53,55],[76,17]]}Variety \ Fruit within 70 daysYesNoTotalAncho8050130Banana5355108Cayenne761793Total209122331
  1. 0.18280.18280.1828
  2. 0.22960.22960.2296
  3. 0.61540.61540.6154
  4. 0.81720.81720.8172
Show solutionHide solution
Solution

(D) 0.81720.81720.8172. The question is about the Cayenne plants only, so the denominator is their total, 93. Of those, 76 produced fruit within 70 days: 7693=0.8172\frac{76}{93} = 0.81729376​=0.8172.

Common mistake
dividing by the wrong total. 76331=0.2296\frac{76}{331} = 0.229633176​=0.2296 (B) is the joint share of all plants, and (C) is the share for the Ancho plants, 80130\frac{80}{130}13080​.

Practice: Two-Way Tables worksheet · AP Statistics Unit 2 worksheet

10. Power and what increases it

Power is the probability of rejecting H0H_0H0​ when a particular alternative is true. A larger sample, a larger α\alphaα or a true value farther from the null all increase it.

Unit 38 official questions on 8 of 13 sources4 on the 2012–2019 practice exams1 FRQ since 2012 (2018)

Practice: Hypothesis Tests worksheet · AP Statistics Unit 3 worksheet

11. Binomial probability

P(X=k)P(X = k)P(X=k) or a cumulative probability for a fixed number of independent trials with the same probability of success.

Unit 28 official questions on 8 of 13 sources, one of them a 2026 CED sample3 on the 2012–2019 practice exams6 FRQs since 2012, 5 of them since 2019

Practice: Probability with Combinatorics worksheet · AP Statistics Unit 2 worksheet

12. Normal probability: an area from a value

Standardize, then find the area on the correct side with normalcdf.

Unit 28 official questions on 7 of 13 sources, one of them a 2026 CED sample6 on the 2012–2019 practice exams8 FRQs since 2012, 4 of them since 2019
Try it
The lengths of songs on a large streaming playlist are approximately normal with mean 235 seconds and standard deviation 24 seconds. What is the probability that the length of a randomly selected song on the playlist is more than 213 seconds?
  1. 0.01090.01090.0109
  2. 0.17970.17970.1797
  3. 0.82030.82030.8203
  4. 0.91670.91670.9167
Show solutionHide solution
Solution

(C) 0.82030.82030.8203. z=213−23524=−0.92z = \frac{213 - 235}{24} = -0.92z=24213−235​=−0.92, and the area to the right of it is normalcdf(213,∞,235,24)=0.8203\text{normalcdf}(213, \infty, 235, 24) = 0.8203normalcdf(213,∞,235,24)=0.8203. 213 seconds is less than one standard deviation below the mean, so most songs are longer than that.

Common mistake
taking the wrong tail. The area to the left of 213 seconds is 0.1797 (B); “more than” is the area to the right.

Practice: Normal Distributions worksheet · AP Statistics Unit 2 worksheet

13. Comparing distributions from graphs

Compare center, spread and shape, and name both groups, in context.

Unit 18 official questions on 7 of 13 sources5 on the 2012–2019 practice exams10 FRQs since 2012, 5 of them since 2019

Practice: Dot Plots worksheet · AP Statistics Unit 1 worksheet

14. Conditional probability and the multiplication rule

P(A and B)=P(A)⋅P(B∣A)P(A \text{ and } B) = P(A) \cdot P(B \mid A)P(A and B)=P(A)⋅P(B∣A), and P(B∣A)=P(A and B)P(A)P(B \mid A) = \frac{P(A \text{ and } B)}{P(A)}P(B∣A)=P(A)P(A and B)​.

Unit 28 official questions on 7 of 13 sources3 on the 2012–2019 practice exams7 FRQs since 2012, 1 of them since 2019

Practice: Conditional Probability worksheet · AP Statistics Unit 2 worksheet

15. Identifying sampling bias

Name the bias (undercoverage, nonresponse, response or voluntary response) and the direction it pushes the estimate.

Unit 18 official questions on 6 of 13 sources5 on the 2012–2019 practice exams6 FRQs since 2012, 2 of them since 2019

Practice: Sampling Methods and Bias worksheet · AP Statistics Unit 1 worksheet

What the free response asks most

The free response is half the score, and its most common parts are not the multiple choice’s. The question types with a part in the most operational free-response questions, 2012–2026:

The question types with a part in the most operational free-response questions, 2012–2026, and their official multiple-choice count
Question typeUnitFRQs since 2012Since 2019Official MC
Describe a scatterplot51143
Compare distributions from graphs11058
Joint, marginal and conditional relative frequencies2868
Normal probability2848
Scope of inference17511
Mean and SD of a discrete random variable2755
Independent events2746
Conditional probability and the multiplication rule2718
  • Describing a scatterplot is the most common free-response part but has only 3 official multiple-choice questions. Direction, form, strength and unusual features, in context: see the Unit 5 review.
  • Describing a random assignment has a single official multiple-choice question but was in 6 operational free-response questions, 5 of them since 2019.
  • Probability is a free-response staple: five of these eight types are from Unit 2.

Rising and fading on the free response

We compare the 42 operational free-response questions of 2012–2018 with the 42 of 2019–2026 (both 7 exams; 2020 was never released), by each unit’s share of the points.

  • Rising: Units 1 and 2. Unit 1 went from 24.9% of the points to 32.5%, and Unit 2 from 22.0% to 28.8%.
  • Fading: Units 3 and 5. Unit 3 went from 24.5% to 14.3%, and Unit 5 from 15.1% to 7.3%.
  • Steady: Unit 4, 8.5% and then 9.4%.

Question 3 still guarantees one full inference question on every 2027 exam, so the drop in Unit 3’s share does not make inference optional.

What has left the course

The 2026 CED drops several topics older exams asked about. Of the 445 unique official multiple-choice questions we found, 39 are on content that is no longer in the course, and so are 108.6 of the 1,040 free-response points.

  • Combining random variables (means and variances of sums and differences): 16 official multiple-choice questions, and a part of 6 free-response questions.
  • Inference for the slope of a regression line: 14 multiple-choice questions and 4 free-response questions.
  • The chi-square goodness-of-fit test: 5 multiple-choice questions and 2 free-response questions. Chi-square tests for homogeneity and independence stay, in Unit 3.
  • The geometric distribution: 1 multiple-choice question and 2 free-response questions.
  • Transformations and influential points in regression: 3 multiple-choice questions.
  • The Investigative Task, the old sixth free-response question: 9 of the questions we counted had a part asking for a new statistic or method.

Old released and practice exams are still good practice for everything else; skip the questions on these topics, and remember that the old multiple choice had five choices.

What students over- and under-practice

We tagged 1,130 multiple-choice questions from teacher tests, review packets, AP Classroom exports and prep books (reposted official questions removed) and compared each unit’s share of that practice with its share of the 406 official questions.

Each unit’s share of 1,130 practice multiple-choice questions against its share of the 406 official ones
UnitPractice vs official shareRatio
1. One-variable data and collecting data28.6% vs 32.5%0.9×
2. Probability and distributions30.7% vs 20.9%1.5× over
3. Inference for proportions15.3% vs 21.2%0.7× under
4. Inference for means15.0% vs 18.7%0.8× under
5. Regression10.4% vs 6.7%1.6× over
  • Probability is over-practiced in multiple choice, though Unit 2 is also rising on the free response.
  • Inference is under-practiced: Units 3 and 4 together are 39.9% of the official multiple choice but 30.3% of the practice.
  • Each unit review lists the types inside the unit that are over- and under-practiced.

How we counted

  • Official multiple choice. 406 questions on today’s course from 13 College Board sources: the 1997, 2002 and 2007 released exams, the 2008, 2012–2015, 2018 and 2019 practice exams, and the sample questions in the 2005, 2020 and 2026 course descriptions. The 24 samples in the 2026 CED are the only official questions in the new format.
  • Official free response. 124 questions worth 1,040 points: the 2012–2026 exams except 2020, the practice exams and the four 2026 samples. Points before 2027 are split from each question’s holistic score in proportion to its scoring sections.
  • Matching. Each question is tagged to one of the CED’s 5 units and 55 topics and to a question type inside the topic.
  • What gets practiced. 1,130 multiple-choice questions from teacher and prep material, for comparison only.

No College Board question is reproduced here. The stats are counts of public College Board material, and every example problem is our own.

Practice by unit

Every unit has a free practice test of 20 original multiple-choice questions in the 2027 format, with sets, an answer key, worked solutions and a PDF, and a free-response set scored out of 10 points per question, with a scoring guide. Each unit also has its own review post with its most-tested question types, worked examples and the most common mistakes.

  • Unit 1: Exploring One-Variable Data and Collecting DataPractice test · Free response · Review
  • Unit 2: Probability, Random Variables, and Probability DistributionsPractice test · Free response · Review
  • Unit 3: Inference for Categorical Data: ProportionsPractice test · Free response · Review
  • Unit 4: Inference for Quantitative Data: MeansPractice test · Free response · Review
  • Unit 5: Regression AnalysisPractice test · Free response · Review

All five units are on the AP Statistics practice tests page. Every formula of the course, with a worked example and the calculator command for each, is on our free AP Statistics formula sheet.

Common Questions

From May 2027 the exam is 3 hours: 42 multiple-choice questions with four choices in 90 minutes, some in sets that share a stimulus, and 4 free-response questions of 10 points each in 90 minutes. Question 1 is on the investigative question and collecting data, Question 2 on analyzing data, Question 3 is a full inference question and Question 4 mixes those skills. Each section is half the score.

Unit 1, Exploring One-Variable Data and Collecting Data. It was 132 of the 406 official multiple-choice questions on today’s course (32.5%) and 31.3% of the 1,040 free-response points we counted, the most of any unit in both sections. The CED gives it 20–30% of the multiple choice.

The math is light: the 2026 course description asks only for first-year algebra, you get a formula sheet and tables, and a graphing calculator is allowed throughout. What makes it hard is the writing. Half the score is free response, 40 points that each need a method, a condition or a conclusion in context, and its most common parts are not what the multiple choice asks most: describing a scatterplot was in 11 operational free-response questions since 2012 but only 3 official multiple-choice questions. Inference is the other hurdle: Units 3 and 4 are 39.9% of the official multiple choice but only 30.3% of the practice questions we tagged.

The 2026 course and exam description drops combining random variables, the geometric distribution, the chi-square goodness-of-fit test, inference for the slope of a regression line, and transformations and influential points in regression. The Investigative Task is gone from the free response, and the course goes from 9 units to 5.

Yes, for most questions. 39 of the 445 unique official multiple-choice questions we found are on topics that have left the course, mostly combining random variables (16) and inference for the regression slope (14), so skip those. Old exams also have five answer choices and 40 multiple-choice questions, against four choices and 42 questions from 2027.

Practice by unit: free AP Statistics worksheets

Exam-style multiple choice and free response with answer keys. No sign-up needed.

Get the AP Statistics worksheets →

Keep reading

AP Statistics

AP Statistics Unit 1 Review: The Most-Tested Data and Study Design Questions

AP Statistics Unit 1 review for the 2027 exam: the data and study design questions the official exams ask most, with examples and a practice test.

AP Statistics

AP Statistics Unit 2 Review: The Most-Tested Probability and Distributions Questions

AP Statistics Unit 2 review for the 2027 exam: the probability and distribution questions the official exams ask most, with examples and a practice test.

AP Statistics

AP Statistics Unit 3 Review: The Most-Tested Inference for Proportions Questions

AP Statistics Unit 3 review for the 2027 exam: the inference for proportions questions the official exams ask most, with examples and a practice test.

Σ
SigmaPrep

Sigma Prep (sigmaprep.io) makes free AP Statistics practice: exam-style multiple choice and free response for all five units, with answer keys and scoring guides, plus a formula sheet.

AP Statistics

Practice tests by unitFormula sheetUnit reviewsStatistics worksheets

Sigma Prep

Math worksheetsAboutContact

Legal

TermsPrivacy
© 2026 Sigma Prep. All rights reserved.AP® is a trademark registered by the College Board, which is not affiliated with, and does not endorse, this site.