AP Statistics pacing guide: 32 weeks, unit by unit

How this pacing guide is built

The official framework assigns each unit a suggested number of 45 minute class periods: about 26 for Unit 1, 24 for Unit 2, 30 for Unit 3, 18 for Unit 4, and 9 for Unit 5, a total of 107 periods. At roughly five periods a week, that comes to about 22 weeks of new content. The remaining 10 weeks of a 32 week year go to cumulative review, free response training, and full length practice exams, which is where most of the score growth in the second half of the course actually happens.

This is a full year teaching plan, unit by unit, week by week. It is not the same document as how to study for AP Statistics, which is a student facing final review sequence for the last eight weeks before the exam, or the cram sheet, which condenses the whole course onto one page for the final days. Use this guide to lay out the calendar; use those two once review season starts.

Each unit hub page lists every topic with its own worked page: Unit 1, Unit 2, Unit 3, Unit 4, Unit 5. A block schedule or a shorter year will not match five periods a week; scale every week count in this guide by your own periods per week instead of trying to force five.

Weeks 1 to 5: Unit 1, exploring one-variable data and collecting data

Unit 1 is the heaviest single unit on the multiple-choice section and also the widest: nine topics of one-variable description followed by four topics of data collection that read almost like a different subject.

Budget week 3 with room to spare. Comparing distributions asks students to write full sentences about shape, center, spread, and outliers side by side, and that is the first place in the course where description turns into writing, which slows a class down more than the mechanics suggest it should.

Weeks 6 to 10: Unit 2, probability, random variables, and probability distributions

Unit 2 opens with two-way tables, spends the middle third on the probability rules, and closes with random variables and the two named distributions that carry into every later inference unit.

Topic 2.12 is worth teaching slowly even though the calendar only gives it half a week. Every confidence interval and every significance test in Units 3 and 4 depends on the sampling distribution idea students meet here for the first time.

Weeks 11 to 16: Unit 3, inference for categorical data, proportions

Unit 3 is the longest unit on the calendar and it runs the same five-step inference cycle twice, once for one proportion and once for the difference between two, before a shorter chi-square pass.

Week 12 is the week to slow down for. Setting up a test looks like the same conditions students already checked for a confidence interval, but the null value replaces the sample proportion inside the standard error, and that single substitution is a common point-loser straight through to test day.

Weeks 17 to 20: Unit 4, inference for quantitative data, means

Unit 4 repeats the Unit 3 cycle with t procedures in place of z, plus paired data folded into the one-sample topics rather than given a topic of its own.

Students who mastered the proportion cycle in Unit 3 usually move faster here, with one exception: identifying whether a two-sample design is actually paired data in disguise. Build week 18 around that identification skill, not just the paired-t arithmetic.

Weeks 21 to 22: Unit 5, regression analysis

Five topics, the smallest unit on the calendar, but the densest per topic and the one most often rushed because it lands at the point in the year when everyone is tired.

On the current exam, regression is descriptive rather than inferential: students interpret the slope, the correlation, r-squared, and residual plots, but there is no significance test for a slope to teach here. That is one topic you can safely drop from an older syllabus without losing anything students will be tested on.

Weeks 23 to 32: cumulative review, free response, and full length practice

The last ten weeks are not more content. They are the stretch where a course that has covered every topic turns into a course that can answer any question about any topic, in any order, against a clock.

Once week 26 starts, hand the remaining sequence to how to study for AP Statistics for the day by day detail; it picks up exactly where this table leaves off, aimed straight at students rather than at a teaching calendar.

Where classes fall behind, and what is safe to cut

The same three places eat a calendar every year, in roughly this order.

Unit 1's second half. The first nine topics are computation and description, which move at a predictable pace. Topics 1.10 through 1.13, sampling and experimental design, are almost entirely writing, and classes that budgeted computation time for them run short. If week 4 or 5 is slipping, do not compress the writing topics; compress week 2 instead, since 1.5 through 1.7 repeat skills from topics 1.1 through 1.4 in slightly more detail.

Unit 3's length. At 30 suggested class periods, Unit 3 is the longest unit and the two-cycle structure, one proportion then two proportions, tempts a class to teach the second cycle at full speed because it feels repetitive. It is repetitive, and that is exactly where a week can be safely trimmed: if you are behind, teach 3.9 through 3.13 as a single compressed week built entirely around what changed from the one-proportion version, rather than re-deriving every step from scratch.

Unit 5, every year. Nine suggested class periods for five topics, sitting at the point in the calendar when momentum is lowest, is a combination that produces a rushed unit almost by design. If a class is going to lose a topic to time pressure, residuals, 5.4, is the one that cannot be that topic: it is where students decide whether a linear model was even appropriate, and it appears inside free response questions built around Unit 1 and Unit 2 skills as often as it does inside dedicated regression questions.

One cut is already made for you and is worth stating plainly so nobody re-teaches it by habit from an older textbook: chi-square goodness of fit, the geometric distribution, and a significance test for a regression slope are not on the current exam. Time spent on any of them is time a class does not have. See what was removed from AP Statistics for the complete list with the old topic codes attached, so a syllabus built from an older book can be checked chapter by chapter rather than from memory.

If a class loses an entire week to snow days, testing, or an assembly schedule, the least damaging place to absorb it is the review stretch, not the new content stretch: drop one of the three full length practice exams in weeks 27, 29, and 31 to two, and use the recovered week for whichever unit's cumulative review, week 23, 24, or 25, covers the material the lost week was supposed to teach.