Exam format reminder: 40 MCQs (90 min, 50% of score) plus 6 FRQs (90 min, 50% of score). The FRQ section includes 5 short questions and 1 investigative task worth more than the others. No formula sheet is provided — you are expected to know the formulas. A table of standard normal probabilities (z-table) and t-distribution critical values are included.
The 8-week schedule
| Week | Focus | What to actually do |
|---|---|---|
| Week 1 Mar 9 | Units 1 & 2: Exploring One- & Two-Variable Data | Graphical displays (dotplots, histograms, boxplots, stemplots) and their shape/center/spread descriptions. Calculating and interpreting mean, median, IQR, standard deviation, and outlier rules. Normal distribution, z-scores, and percentile calculations using the z-table. Scatterplots, correlation, least-squares regression line (LSRL) equation and interpretation, residual plots, and coefficient of determination (r²). 20 released MCQs. Practice writing interpretation sentences for every graph — always name the variable and the direction. |
| Week 2 Mar 16 | Unit 3: Collecting Data + Unit 4: Probability | Sampling methods (SRS, stratified, cluster, systematic) and their advantages and sources of bias. Experimental design: control groups, random assignment, blinding, blocking, placebo effect. Observational studies vs. experiments and why only experiments establish causation. Then shift to probability: basic rules, complement, addition rule, multiplication rule, conditional probability, and independence. Construct two-way tables and tree diagrams for conditional probability problems. 20 MCQs across both units. 1 FRQ on experimental design — practice writing a complete design description that names the treatments, the random assignment method, and the response variable. |
| Week 3 Mar 23 | Unit 5: Sampling Distributions | The Central Limit Theorem and when it kicks in (n ≥ 30, or smaller if population is normal). Sampling distribution of the sample proportion p̂: mean = p, standard deviation = √(p(1−p)/n), conditions for approximate normality (np ≥ 10 and n(1−p) ≥ 10). Sampling distribution of the sample mean x̄: mean = μ, standard deviation = σ/√n. Calculating probabilities for x̄ and p̂ using z-scores. This unit is the bridge between descriptive statistics and inference — every test you do in Units 6–9 rests on these distributions. 15 calculation problems. Write out the full sampling distribution (shape, center, spread) for at least 5 different scenarios before moving on. |
| Week 4 Apr 13 | Unit 6: Inference for Proportions | One-sample z-test for a proportion and one-sample z-interval for a proportion. Two-sample z-test and z-interval for the difference of two proportions. For every procedure, practice checking all three conditions: Random (was the data collected randomly?), Independent (10% condition: n ≤ 10% of population), Large Counts (np̂ ≥ 10 and n(1−p̂) ≥ 10). Interpreting confidence intervals: "We are 95% confident that the true proportion of [variable] is between [lower] and [upper]." Interpreting p-values in context. 15 problems, 2 complete 4-step FRQs. This unit appears on almost every released exam in some form. |
| Week 5 Apr 20 | Unit 7: Inference for Means | One-sample t-test and t-interval. Paired t-test for matched pairs data (recognize when data is paired vs. two independent groups). Two-sample t-test and t-interval for the difference of two means. For every procedure: Random, Independent (10% condition), and Normal/Large Sample (either n ≥ 30, or population is stated as normal, or the sample data shows no strong skew or outliers). Know when to use a t-distribution vs. z: always t for means unless the population standard deviation σ is given (rare). Degrees of freedom for one-sample t: df = n−1. Take Practice Exam 1 at end of week. |
| Week 6 Apr 27 | Units 8 & 9: Chi-Square + Inference for Regression | Chi-square goodness-of-fit test (one variable, comparing to claimed distribution), chi-square test for independence (one sample, two variables), and chi-square test for homogeneity (two or more populations). Expected counts formula: (row total × column total) / table total. Chi-square conditions: random, independent, all expected counts ≥ 5. Inference for slope: t-test for slope of the LSRL, t-interval for slope, interpreting computer output (coefficient, SE, t-statistic, p-value). Redo all missed problems from Practice Exam 1. 2 timed FRQs on chi-square. |
| Week 7 May 4 | Full AP FRQ Practice | Dedicated FRQ week. Write 5 complete 4-step procedures each day from memory — no notes. Monday–Tuesday: inference for proportions and means. Wednesday: chi-square and regression. Thursday: Investigative Task format. The Investigative Task is the last FRQ and is worth roughly double a short question — it often introduces an unfamiliar procedure or context and tests whether you can reason statistically, not just recall steps. Friday: timed 90-minute FRQ session (6 questions) using a released exam FRQ set. Review scoring guidelines and identify every point lost to missing context sentences. |
| Week 8 May 11 | Final review & rest | Write all inference procedures from memory in one sitting: the name, hypotheses, conditions, test statistic formula, and conclusion sentence structure for every test (z for proportions, t for means, paired t, two-sample t, chi-square GOF, chi-square independence/homogeneity, t for slope). Mixed MCQ practice Mon–Tue targeting your two weakest content areas from Practice Exam 1. Take Practice Exam 2 on Wednesday with full timing. Thu: error analysis only — no new content. Fri–Sat: review the z-table and t-table so you are not slow looking up critical values. Rest the day before the exam. May exam day is the second week of May. |
Get an AP Statistics schedule that adapts to your exam date
The AI generator takes your exam date and current unit progress and rebuilds when you miss a session. Tracks which inference procedures you have practiced so you drill the right tests.
Generate My Schedule FreeContent area weights on the AP Statistics exam
The College Board breaks AP Statistics into four content areas with published exam weights. Statistical Inference and Probability together make up about 65% of your score, so they deserve the majority of your prep time.
- Exploring Data (Units 1–2): ~23% (graphical displays, summary statistics, normal distributions, LSRL, residuals)
- Sampling and Experimentation (Unit 3): ~16% (sampling methods, experimental design, sources of bias)
- Probability and Simulation (Units 4–5): ~30% (probability rules, conditional probability, sampling distributions — the CLT lives here)
- Statistical Inference (Units 6–9): ~35% (highest weight; inference for proportions, means, chi-square, and regression slope)
Within inference, Units 6 and 7 (proportions and means) are tested most heavily. If you have limited time, those two units are where every extra hour pays off.
What most AP Statistics students get wrong
They write math answers instead of statistics answers. This is the single biggest source of lost points on AP Statistics FRQs. A p-value of 0.04 is not an answer — "because p = 0.04 < α = 0.05, we reject the null hypothesis. There is convincing evidence that the true proportion of students who prefer the new format is greater than 0.5" is an answer. The examiners explicitly dock points for responses that omit the context (the variable name, the direction, the population). Every conclusion you write must reference what is actually being studied.
They skip checking conditions. Checking conditions is not optional and not a formality. It is a scored part of every inference FRQ. Students who jump straight to the test statistic lose those points on every single inference question. Practice writing the conditions check before you touch your calculator: name the condition, state what the condition requires, and confirm it is met using numbers from the problem.
They confuse the paired t-test with the two-sample t-test. If the data comes in natural pairs (before/after measurements on the same subject, left hand vs. right hand, twins in a study), you compute differences and run a one-sample t-test on those differences. If the two groups are independent, you run a two-sample t-test. Choosing the wrong procedure on an FRQ costs points even if your arithmetic is correct.
They do not practice the Investigative Task. The final FRQ is worth more than a standard short question and consistently features a context or procedure students have not seen before. It is designed to test statistical reasoning, not recall. Students who have only practiced routine 4-step procedures are unprepared. Practice reading unfamiliar scenarios and identifying which statistical tools apply, then explaining your reasoning in plain language.
They misinterpret confidence intervals. "There is a 95% probability that the true proportion is in this interval" is wrong — the interval either contains the true proportion or it does not. The correct interpretation is about the process: "if we repeated this procedure many times, 95% of the intervals constructed this way would contain the true proportion." The AP exam tests this distinction directly.
The 4-week compressed version
- Week 1: Units 1, 2, 3, and 4. Two sessions on graphical displays and summary statistics, one session on LSRL and residuals, one session on experimental design, one session on probability rules and conditional probability. 15 MCQs per unit area.
- Week 2: Unit 5 (sampling distributions) and Units 6–7 (inference for proportions and means). Practice checking conditions for every problem. Practice Exam 1 at end of week.
- Week 3: Units 8 and 9 (chi-square and regression inference). Three complete 4-step FRQs per session. Practice Exam 2 at end of week. Focus on Investigative Task format for at least one session.
- Week 4: Weak-area targeting Mon–Wed using both practice exam error logs. Write all inference procedures from memory Thu. Rest Fri–Sun before the exam.