AP Statistics units 2027: all five, weighted

The five units at a glance

This is the course structure effective Fall 2026, first examined May 2027. Titles, weightings, topic codes and suggested pacing all come from the official Course and Exam Description, which this site transcribes rather than paraphrases.

Fifty-five topics in total, across about 107 suggested class periods of 45 minutes. The weightings apply to Section I, the 42 multiple-choice questions; Section II is four free-response questions that cut across units, and the exam format page has that structure in full.

The previous course had nine units. If a syllabus, video or book in front of you counts nine, it describes the course this one replaced. What changed in AP Statistics for 2027 covers the redesign as a whole.

Unit 1: Exploring One-Variable Data and Collecting Data

20-30% of the multiple-choice section, roughly 26 class periods, 13 topics. The heaviest unit on the exam.

1. 1.1 Introducing Statistics: What Can We Learn from Data?
2. 1.2 Variables
3. 1.3 Tabular Representation and Summary Statistics for One Categorical Variable
4. 1.4 Graphical Representations for One Categorical Variable
5. 1.5 Graphical Representations for One Quantitative Variable
6. 1.6 Descriptions for One Quantitative Variable Distributions
7. 1.7 Summary Statistics for One Quantitative Variable
8. 1.8 Graphical Representations of Summary Statistics for One Quantitative Variable
9. 1.9 Comparisons of the Distributions for One Quantitative Variable
10. 1.10 The Investigative Question Revisited and Data Collection
11. 1.11 Random Sampling
12. 1.12 Potential Problems with Sampling
13. 1.13 Experimental Design

Notice the shape of it. Topics 1.1 to 1.9 are descriptive statistics for one variable, and 1.10 to 1.13 are data collection: sampling, bias and experimental design. Two large areas share one unit, which is part of why this unit carries the largest weight in the course.

It also means the unit spans two quite different skills. The first half is computation and description; the second half is almost entirely writing. Work describing distributions practice for the first and experimental design practice for the second, and see Unit 1 in full.

Unit 2: Probability, Random Variables, and Probability Distributions

15-25% of the multiple-choice section, roughly 24 class periods, 12 topics.

1. 2.1 Tabular and Graphical Representations for the Distributions of Two Categorical Variables
2. 2.2 Summary Statistics for Two Categorical Variables
3. 2.3 Estimating Probabilities Using Simulation
4. 2.4 Introduction to Probability
5. 2.5 Mutually Exclusive Events
6. 2.6 Conditional Probability
7. 2.7 Independent Events and Unions of Events
8. 2.8 Introduction to Random Variables and Probability Distributions
9. 2.9 Parameters of Random Variables
10. 2.10 The Binomial Distribution
11. 2.11 The Normal Distribution
12. 2.12 Sampling Distributions and the Central Limit Theorem

Two things are worth flagging here. Two-way tables open the unit at 2.1 and 2.2, before probability begins, which is a sensible order: a two-way table is where conditional probability becomes concrete. And the unit ends at 2.12 with sampling distributions and the central limit theorem, which is the hinge of the entire course. Everything in Units 3 and 4 is built on it.

The geometric distribution is not in this list, and that is not an omission on our part: it was removed from the course. See what was removed from AP Statistics. Work probability rules practice and sampling distribution CLT practice, and see Unit 2 in full.

Unit 3: Inference for Categorical Data: Proportions

15-25% of the multiple-choice section, roughly 30 class periods, 15 topics. The most class periods of any unit.

1. 3.1 Estimators
2. 3.2 Sampling Distributions for Sample Proportions
3. 3.3 Constructing a Confidence Interval for a Population Proportion
4. 3.4 Justifying a Claim Based on a Confidence Interval for a Population Proportion
5. 3.5 Setting Up a Test for a Population Proportion
6. 3.6 p-Values
7. 3.7 Carrying Out a Test for a Population Proportion
8. 3.8 Potential Errors When Performing Tests
9. 3.9 Sampling Distributions for the Difference Between Sample Proportions
10. 3.10 Constructing a Confidence Interval for the Difference Between Two Population Proportions
11. 3.11 Justifying a Claim Based on a Confidence Interval for the Difference Between Two Population Proportions
12. 3.12 Setting Up a Test for the Difference Between Two Population Proportions
13. 3.13 Carrying Out a Test for the Difference Between Two Population Proportions
14. 3.14 Setting Up a Chi-Square Test for Homogeneity or Independence
15. 3.15 Carrying Out a Chi-Square Test for Homogeneity or Independence

Read the titles as a pattern and the unit gets much smaller than 15 topics. It runs the same five-step cycle twice: sampling distribution, confidence interval, justifying a claim from that interval, setting up a test, carrying out a test. Once for one proportion, at 3.2, 3.3, 3.4, 3.5 and 3.7, and once for the difference between two proportions, at 3.9 to 3.13. Chi-square then adds a shorter pass, set-up and carry-out only, at 3.14 and 3.15, with no confidence interval of its own. Three topics stand outside the cycle: 3.1 Estimators, 3.6 p-Values and 3.8 Potential Errors When Performing Tests. Five plus five plus two plus three is the 15.

That repetition is the whole reason inference is learnable. The arithmetic changes and the logic does not. Note also that chi-square lives here, in the categorical unit, rather than in a unit of its own, and that goodness of fit is not among the two chi-square topics because it was removed. Work proportion z-test practice and chi-square two-way practice, and see Unit 3 in full.

Unit 4: Inference for Quantitative Data: Means

10-20% of the multiple-choice section, roughly 18 class periods, 10 topics.

1. 4.1 Sampling Distributions for Sample Means
2. 4.2 Constructing a Confidence Interval for a Population Mean or Population Mean Difference
3. 4.3 Justifying a Claim Based on a Confidence Interval for a Population Mean or Population Mean Difference
4. 4.4 Setting Up a Test for a Population Mean or Population Mean Difference
5. 4.5 Carrying Out a Test for a Population Mean or Population Mean Difference
6. 4.6 Sampling Distributions for the Difference Between Two Sample Means
7. 4.7 Constructing a Confidence Interval for the Difference Between Two Population Means
8. 4.8 Justifying a Claim Based on a Confidence Interval for the Difference Between Two Population Means
9. 4.9 Setting Up a Test for the Difference Between Two Population Means
10. 4.10 Carrying Out a Test for the Difference Between Two Population Means

Same cycle as Unit 3, now with $t$ procedures. The phrase to read closely is population mean difference in topics 4.2 to 4.5: that is the paired setting, folded into the one-sample topics rather than given its own. Paired data are analyzed as one sample of differences, which is exactly why they sit there. See when to use a paired t-test and paired t practice.

Topic 4.9 in this course is Setting Up a Test for the Difference Between Two Population Means. It is not the removed topic 4.9 from the old course, which was combining random variables. This is the sort of collision that makes topic numbers unreliable across the redesign. Work t-test practice and two-sample t practice, and see Unit 4 in full.

Unit 5: Regression Analysis

10-20% of the multiple-choice section, roughly 9 class periods, 5 topics. The smallest unit in the course by both topic count and pacing.

1. 5.1 Graphical Representations Between Two Quantitative Variables
2. 5.2 Correlation
3. 5.3 Linear Regression Models
4. 5.4 Residuals
5. 5.5 Least-Squares Regression

Five topics for the same weight band as Unit 4's ten, on roughly half the class periods. Per topic, this is the densest unit on the exam, and per class period it is the unit most likely to have been rushed at the end of a school year. If you are triaging revision, that combination makes Unit 5 unusually cheap to repair.

The unit is descriptive and diagnostic rather than inferential: describing a scatterplot, interpreting $r$ and $r^2$, writing and using the least-squares line, and reading residuals. Inference for the slope was a whole unit in the old course and was removed, so a $t$ test for a slope is not examinable here.

Topic 5.4, Residuals, carries more than its share of the difficulty, because it is where you decide whether the model was appropriate at all. Work residual plots practice and regression practice, read how to interpret residual plots, and see Unit 5 in full.

How the old nine units relate to these five

This is where most pages about the redesign start inventing, so here is exactly what can be said.

What is sourced. The course went from nine units to five, and everything above is what the current framework contains, topic by topic. College Board's revisions page also names five entries that were removed outright, so a chapter with no match above may simply be gone rather than moved; what was removed from AP Statistics lists all five with their old codes translated. One of the five is routinely over-read: what went is transforming data to straighten a curve, and spotting the curve is topic 5.4 above, which is still assessed.

What is not sourced, and therefore is not published here. A per-unit crosswalk saying old Unit 3 became new Unit 2, and so on. The old Course and Exam Description is not a source in hand for this site, and a map written from memory would be exactly the kind of confident error this redesign is generating everywhere else. If you need the old structure, get it from the old document rather than from anyone's recollection of it.

What you can safely do instead is match on titles. Take any chapter of an older resource, find the closest title in the five lists above, and use that as the anchor. If nothing matches, check the removal list before assuming the topic moved. Should I use old AP Statistics review books? turns that matching exercise into a 20-minute audit of a book you already own.

Reading the weightings without misreading them

Three cautions before you build a revision plan on the table at the top.

The weightings describe Section I only. They are the share of the 42 multiple-choice questions. Section II is four free-response questions organized around the four course practices rather than around units, so a unit's weight does not tell you its share of the whole exam.

They are ranges, and they do not sum to 100 at the midpoints. The five midpoints add to 95, so if you are converting weights into hours or questions, scale rather than treating each midpoint as a final percentage. The worked examples below do it both ways.

Class periods are pacing, not importance. Unit 3 gets the most periods because inference takes the longest to teach, not because it is worth the most. Unit 5 gets the fewest for the same weight band as Unit 4, which is the opposite signal.

A reasonable reading: revise in proportion to the weightings, then add time wherever your practice scores are worst, then add a little more to Unit 5 because the pacing suggests it was taught fast. How to study for AP Statistics turns that into a week-by-week plan, and the cram sheet condenses the whole course for the last week.