How many topics are in AP Statistics? 55

55 topics, and how they split

The AP Statistics course effective Fall 2026, first examined May 2027, lists 55 numbered topics. Each one has a code such as 1.7 or 3.14, where the first number is the unit. Count them straight off the framework's topic list and you get $13 + 12 + 15 + 10 + 5 = 55$.

Unit 3 holds the most topics and Unit 5 the fewest, a spread of three to one. Every topic in a unit is listed with its code on that unit's page:

• Unit 1: Exploring One-Variable Data and Collecting Data
• Unit 2: Probability, Random Variables, and Probability Distributions
• Unit 3: Inference for Categorical Data: Proportions
• Unit 4: Inference for Quantitative Data: Means
• Unit 5: Regression Analysis

If you want the units themselves rather than the topic count, how many units are in AP Statistics answers that one, and the full unit and topic list prints all 55 codes with their titles.

Class time per topic is almost identical everywhere

Divide each unit's suggested class periods by its topic count and the numbers barely move.

Three units land on 2.0 periods per topic and two on 1.8, on the framework's approximate pacing. Across the whole course it is $107 \div 55 \approx 1.95$ periods per topic. Correlate the five topic counts against the five class-period counts and $r \approx 0.996$, which is about as close to a straight line as five real data points get.

So the schedule is built the way you would guess: a topic is roughly a topic's worth of teaching time, and a unit with more topics gets proportionally more of the year. Topic count is an excellent proxy for how long a unit takes to teach.

Two cautions on that. The class periods are the framework's suggested pacing, written with a tilde because they are approximate, and no real classroom follows them to the period. And a period of teaching time is not a period of revision time, because the topics differ enormously in how much of them you have to reconstruct from scratch later.

Exam weight per topic need not follow the topic count

Now do the same division with the published weightings. Each unit's weighting is a range, so a range divided by a topic count is a range, not a single number.

The cleanest comparisons are the two pairs of units that share a band, because the band cancels out.

Units 2 and 3 are both 15-25%, with 12 topics and 15 topics. If the two units land at the same point in that band, each Unit 2 topic carries $15 \div 12 = 1.25$ times what a Unit 3 topic carries.

Units 4 and 5 are both 10-20%, with 10 topics and 5 topics. Same reasoning, and at a matched point the factor is exactly 2: five topics hold what ten hold. Nothing forces them to land together, though. Unit 4 at 20% against Unit 5 at 10% sits inside both bands, and there a topic from each unit is worth the same 2.00%.

Across the whole table the only two bands that do not overlap at all are Unit 5's 2.00% to 4.00% and Unit 3's 1.00% to 1.67%. The average Unit 5 topic is worth more of Section I than the average Unit 3 topic, whatever the exam does inside those ranges. Elsewhere you cannot rank so cleanly: Unit 1's per-topic band runs up to 2.31%, which is above Unit 5's floor of 2.00%, so a heavy Unit 1 and a light Unit 5 would put the average Unit 1 topic ahead. Read every figure in the weight-per-topic column as a unit average. College Board weights units and not topics, so nothing published here ranks one individual topic against another.

What that means for revision

Put the two tables next to each other and the finding is a looseness. Teaching time is handed out per topic at a nearly constant rate, locked between 1.8 and 2.0 periods. Exam weight per topic is only bounded: it has to vary by at least 20%, because Unit 5's floor of 2.00% sits above Unit 3's ceiling of 1.67%, and the bands allow it to vary by a factor of 4. So the number of topics you have covered tells you how much of the year has gone, and it does not tell you how much of the exam you are holding.

Three things follow.

Do not budget revision by counting topics. A student who splits time evenly across all 55 gives Unit 3 27.3% of their revision and Unit 5 9.1%. Set those against the same units' shares of Section I and both fall outside the published band: Unit 3 tops out at 25% and Unit 5 starts at 10%. The other three units do land inside theirs, at 23.6% against 20-30% for Unit 1, 21.8% against 15-25% for Unit 2 and 18.2% against 10-20% for Unit 4, so the mismatch is concentrated at the two ends of the course.

Revise Unit 5 topic by topic. Five topics at 2.00% to 4.00% of Section I each, roughly 1% to 2% of the whole exam, means a single one of them is worth its own revision session rather than a paragraph inside a chapter. Correlation, Linear Regression Models and Residuals are three of the five. How to interpret residual plots is the drill for the last of them.

Revise Unit 3 by its pattern, not by its list. Fifteen topics at 1.00% to 1.67% each gives Unit 3 the lowest per-topic ceiling on the course, though Unit 4 shares its 1.00% floor and can land below it. The unit runs one five-step cycle twice, once for a single proportion and once for the difference between two, then a shorter two-topic chi-square pass 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. Learning the cycle once and then learning those three beats learning fifteen separate things, and the unit breakdown sets out how the fifteen decompose.

One limit on all of it. These weightings describe Section I only, the 42 multiple-choice questions, which are 50% of the exam. Section II is four free-response questions: three multi-focus questions named for the course practices they target, and Question 3, which is always an inference question and is 12.5% of the exam on its own. So no unit weighting and no per-topic figure derived from one describes your whole score, though that guaranteed inference question is one more reason Units 3 and 4 repay the time. The four course practices is the other half of the picture, and the exam format page has both sections in full.

Nearly half the topics are inference

Units 3 and 4 hold 25 of the 55 topics between them, which is 45.5% of the list, and 48 of the roughly 107 class periods, which is 44.9% of the year. Their weightings add to a band of 25% to 45% of Section I.

That is worth seeing as one number rather than two units. Almost half of AP Statistics by count, and almost half of it by scheduled time, is sampling distributions and the inference built on them: confidence intervals, significance tests and the two chi-square tests. The old course spread inference over more units, and the redesign cut one of them outright, because the old Unit 9, inference for regression slopes, is gone. Inference for proportions and for means was not cut, and it is still nearly half the topic list, which is the short answer to is inference still on the AP Statistics exam.

The other 30 topics are Unit 1's data description and data collection, Unit 2's probability and distributions, and Unit 5's regression. Unit 2 ends at topic 2.12, Sampling Distributions and the Central Limit Theorem, which is the topic almost all of those 25 inference topics stand on. By count it is one topic out of 55. By consequence it is the hinge of the course, which is the sharpest single reminder that these counts are a planning tool and not a ranking.

Topic numbers do not carry across the redesign

A topic code is only meaningful against the course it came from, and the redesign reused the numbering space. Two live codes collide with numbers on College Board's removal list:

• 2.9 on that list was analyzing departures from linearity. Topic 2.9 in the current course is Parameters of Random Variables, which is examinable.
• 4.9 on that list was combining random variables. Topic 4.9 in the current course is Setting Up a Test for the Difference Between Two Population Means, which is examinable.

The other entries on that list have no current counterpart at all: 4.12 does not exist, because Unit 4 stops at 4.10; 8.2 and 8.3 do not exist, because the course has five units; and the fifth entry is not a topic code at all but a whole unit, the old Unit 9, Inference for Quantitative Data: Slopes, so there is no Unit 9 either. So the safe rule is never to quote a bare number from the removal list against the current course. Match on titles instead.

One of the removals is over-read constantly. What went is the topic on transforming data to straighten a curve, and the course now has no topic on that. Reading curvature in a residual plot is a different thing and is still assessed, under topic 5.4 Residuals. What was removed from AP Statistics works through all five entries with the same care.

The practical version: if a book, video or worksheet cites a topic number, check it against a current unit list before you trust it. If it counts nine units, its numbers describe the course this one replaced. What changed for 2027 covers the rest of the redesign.