AP Statistics measures how well you interpret statistical data and make reasoned decisions under uncertainty. The course and exam were recently redesigned, so before preparing it matters which format you are preparing for.
The exam changed: be careful with old materials
The format College Board publishes for the 2027 exam is this: the exam is fully digital and taken in the Bluebook app. Section 1 has 42 multiple choice questions and Section 2 has four free response questions; both sections last 90 minutes and carry half the score.
The old format had 40 multiple choice questions and six free response questions, the last one a long investigative task. Most prep books on the market and most past exams are still in the old format. Old free response questions remain valuable practice for writing justifications, but the timing and the number of questions are now different.
Five units and their weights
Unit 1, exploring one-variable data and collecting data: 20 to 30 percent. Sampling and experimental design are in this unit too.
Unit 2, probability, random variables and probability distributions: 15 to 25 percent. The binomial and normal distributions are here.
Unit 3, inference for categorical data (proportions): 15 to 25 percent. Confidence intervals, the difference of two proportions and the chi-square test are here.
Unit 4, inference for quantitative data (means): 10 to 20 percent.
Unit 5, regression analysis: 10 to 20 percent. Correlation and the least squares line are here.
The heaviest unit is the first, and many students treat it as easy and move through it quickly. The condition checks in the inference units rest directly on the sampling and experimental design from Unit 1.
Four free response questions
Each question is worth 10 points and they always come in the same order. College Board ties the questions to the four statistical practices of the course: formulating questions, collecting data, analyzing data and interpreting results.
Question 1 is multi-part and mainly assesses the first two: framing the investigative question, and choosing and justifying a sampling or experimental design.
Question 2 is multi-part and mainly assesses the last two: analyzing data with calculations and graphs, and interpreting the result with a justification.
Question 3 is a pure inference question: a confidence interval or significance test built from start to finish.
Question 4 combines several units and is the least predictable question.
Calculator and formula sheet
A calculator is allowed on the whole exam, and a graphing calculator with statistical functions is expected on the free response section. A formula sheet and tables are provided.
The number the calculator returns earns nothing on its own. The name of the procedure, the check of its conditions and the interpretation of the result in context must be written down.
Four steps of an inference answer
A pattern widely used in classrooms splits an inference question into four steps: state the hypotheses or the parameter in words, name the procedure and check its conditions with numbers, do the calculation, and write the conclusion in context, linked to the P-value or the interval. Scoring guidelines look for each of these four parts separately.
The three most common mistakes
Writing a conclusion without context. Reject H0 is incomplete; what has been shown must be written in the words of the problem.
Accepting H0. A significance test can only reject H0 or fail to reject it; saying we accept it loses credit.
Passing the conditions without numbers. The sentence conditions are met earns nothing on its own; each condition is written with its name and its numbers.
How long it takes
That depends on the starting point rather than on a fixed number of months. For a student with a solid probability foundation, the real work is building the habit of writing justifications; that is settled with a diagnostic rather than an estimate.