The four AP Statistics statistical practices run through every unit and every exam question: formulating questions, collecting data, analyzing data, and interpreting results. This page collects the four topic guides so you can review each practice, see how they connect, and know exactly what the exam expects from each one.
Fiveable's AP Statistics Statistical Practices review covers 5 topics, with guided practice questions for the unit.
Use these guides when you want to strengthen a specific practice skill, when you are losing points on interpretation or setup questions, or when you are doing a final pass before exam day.
AP Statistics is organized around content units, but it is also organized around four statistical practices that describe what statisticians actually do. Every free-response question and most multiple-choice questions ask you to perform one or more of these practices. Knowing the practices by name and by what they demand helps you recognize what a question is really asking.
The four statistical practices are Formulate Questions (Skill 1), Collect Data (Skill 2), Analyze Data (Skill 3), and Interpret Results (Skill 4). They form a cycle: you write a question, plan how to gather data, produce calculations and graphs, and then explain what the results mean in context.
Free-response scoring rubrics are built around these practices. A correct calculation with a missing or context-free interpretation loses points because Interpret Results is a scored skill, not a bonus. Knowing which practice a question targets tells you what the rubric is looking for.
Collect Data shows up in Unit 1 study design questions and again when you name hypotheses in Units 3 and 4. Interpret Results is required any time you write a conclusion for a confidence interval or significance test. No practice is confined to one unit.
The practices form a logical sequence. Formulate Questions defines what you are trying to learn. Collect Data determines how you gather evidence and which inference procedure fits. Analyze Data produces the numbers and graphs. Interpret Results turns those outputs into a real-world conclusion.
When an AP Statistics free-response question says 'describe,' 'justify,' 'identify,' or 'interpret,' it is signaling which practice is being scored. Formulate Questions asks you to write a valid investigative question. Collect Data asks you to choose a procedure or identify a study design. Analyze Data asks you to compute or construct. Interpret Results asks you to explain what a number, interval, or decision means in the context of the problem. Matching your response to the right practice is how you earn full credit.
Learn what makes an investigative question statistically valid and how this skill sets up every study and inference task on the exam.
Review study design, bias, randomization, hypothesis setup, and how to match an inference procedure to the data structure.
Work through the computational and graphical outputs: graphs, summary statistics, probabilities, and the mechanics of inference procedures.
Practice writing complete, in-context interpretations of p-values, confidence intervals, regression outputs, and test conclusions.
Estimate your AP score based on your performance across units and practices to guide where to focus your remaining review time.
Skill 1.A requires you to write a question that can only be answered by collecting and analyzing data. A valid investigative question names a variable of interest, acknowledges variability, and is answerable through a statistical study such as a survey, observational study, or experiment. Vague questions like 'Do people like exercise?' are not valid because they do not specify a measurable variable or a population.
Can you write a valid investigative question for a given scenario and explain why it requires statistical analysis rather than a simple lookup?
| Not a valid investigative question | Valid investigative question |
|---|---|
| Do students sleep enough? | What is the mean number of hours of sleep per night for students at this school, and does it differ by grade level? |
| Is the new drug better? | Is the mean recovery time for patients taking the new drug lower than for patients taking the placebo? |
Skill 2 covers study design, ethical data collection, and matching an inference procedure to the data structure. You need to distinguish observational studies from experiments, identify sources of bias, describe randomization, and name the correct test or interval. This practice also includes naming hypotheses and identifying Type I and Type II errors, which appear in Units 3 and 4.
Given a study description, can you identify whether it is an experiment or observational study, name a potential source of bias, and select the correct inference procedure?
| Feature | Observational Study | Experiment |
|---|---|---|
| Treatment assignment | Not assigned by researcher | Randomly assigned by researcher |
| Can establish causation | No | Yes |
| Example | Survey on sleep habits | Random assignment to new drug vs. placebo |
Skill 3 is where you do the computational and graphical work. This includes constructing dotplots, histograms, boxplots, and scatterplots; computing means, medians, standard deviations, and IQR; calculating probabilities using normal, binomial, and geometric distributions; and running the mechanics of confidence intervals and significance tests. The guide for this practice covers all of these output types.
Can you compute a standardized test statistic and a confidence interval for a proportion or mean, and construct an appropriate graph for a given data set?
| Output type | What you produce | Where it appears |
|---|---|---|
| Graph | Histogram, boxplot, scatterplot, residual plot | Units 1, 5 |
| Summary statistic | Mean, standard deviation, IQR, correlation r | Units 1, 5 |
| Probability | P(X = k), P(a < X < b) using z-scores or tables | Unit 2 |
| Inference result | Test statistic, p-value, confidence interval | Units 3, 4 |
Skill 4 is the most heavily penalized practice when students skip it or write it vaguely. Every interpretation must reference the context of the problem. Interpreting a p-value means stating the probability of getting results at least as extreme as observed, assuming the null is true, not just comparing it to alpha. Interpreting a confidence interval means describing what the interval says about the parameter, not the statistic.
Can you write a complete interpretation of a p-value, a confidence interval, and a regression slope, each in the context of a specific problem?
| Output | Incomplete interpretation | Complete in-context interpretation |
|---|---|---|
| p-value of 0.03 | The p-value is less than 0.05, so reject H0. | A p-value of 0.03 means that if the true mean recovery time were equal for both treatments, there is a 3% chance of seeing a difference at least this large by chance alone. We reject H0 and conclude the new drug reduces recovery time. |
| 95% CI for proportion | We are 95% confident the proportion is between 0.42 and 0.58. | We are 95% confident that the true proportion of students at this school who walk to school is between 0.42 and 0.58. |
Once an interval is computed, the parameter is either in it or not. The correct interpretation refers to the method: intervals built this way capture the true parameter 95% of the time in repeated sampling. Do not say 'there is a 95% chance the parameter is in this interval.'
Writing 'the p-value is 0.04, so we reject H0' earns partial credit at best. Every interpretation must name the variable, population, and real-world meaning. Examiners look for context in every sentence of Skill 4 responses.
An observational study cannot establish causation no matter how strong the association. If researchers did not randomly assign treatments, the study is observational. Confounding variables are always a concern in observational studies.
Collect Data includes verifying that conditions for a procedure are met. Stating the correct test name without checking randomness, independence, and the appropriate sample size condition will cost points on free-response questions.
The p-value is computed assuming the null is true. It is not the probability that H0 is true or false. A small p-value means the observed result is unlikely under H0, which is evidence against H0, not proof that H0 is false.
A single multi-part free-response question can ask you to formulate a question, describe a data collection method, compute a test statistic, and interpret the result. Each part targets a different practice, and each is scored separately. Missing one practice in your response means losing those points even if the rest is correct.
Many multiple-choice questions test one practice in isolation. A question might give you a completed calculation and ask only for the correct interpretation, targeting Skill 4 without any computation. Recognizing which practice a question targets helps you focus your response and avoid overthinking.
Students who can calculate correctly but write vague or context-free interpretations consistently lose points. Phrases like 'in the context of this problem' in a rubric mean the interpretation must name the variable and population. Practicing complete interpretation sentences is one of the highest-return exam preparation moves.
Given a scenario, produce a question that names a measurable variable, acknowledges variability, and requires data collection to answer. Check that it is not answerable by a simple fact lookup.
Distinguish experiments from observational studies, explain why randomization matters, and name at least one source of bias in a described study. Know that only experiments can establish causation.
Given a problem setup, name the correct test or interval, state the conditions required, and verify those conditions using the given information. Know when to use z versus t procedures.
Compute a standardized test statistic, find a p-value, construct a confidence interval, and build or describe an appropriate graph. Show all formula steps when the exam asks you to calculate.
For every p-value, confidence interval, slope, correlation, or test conclusion, write a sentence that includes the context of the problem. Avoid generic statements that could apply to any problem.
Given a hypothesis testing scenario, describe what a Type I error and a Type II error would mean in real-world terms, not just in abstract statistical language.
Read the Formulate Questions guide first if you find yourself unsure what a question is asking or if you struggle to write investigative questions on free-response prompts. This practice frames everything else.
If you frequently choose the wrong test or interval, or if study design questions trip you up, work through the Collect Data guide. Pay close attention to the conditions for each procedure and the distinction between experiments and observational studies.
If your calculations are losing points, the Analyze Data guide covers every output type you need: graphs, summary statistics, probabilities, and inference mechanics. Practice computing test statistics and confidence intervals by hand.
Interpretation is the most common source of lost points on free-response questions. Read the Interpret Results guide and practice writing complete, in-context sentences for p-values, confidence intervals, slopes, and test conclusions.
After reviewing the guides, use the AP Statistics score calculator to estimate where you stand and decide whether to spend remaining time on a specific practice or a specific content unit.
Open the individual guides for Statistical Practices when you want a closer review of one topic.
Use AP-style practice after you review the notes so you can check what you understand.
Practice free-response reasoning and compare your answer with scoring guidance.
Open the Fiveable one-page unit review, then explore visual cheatsheets for a quick refresher.
Estimate your broader AP score goal after you review the course and exam format.
Print the cheatsheet bundle ($14) or full-length exam ($12) for this subject. Yours to keep, no subscription needed.
get printables