Tabular and Graphical Representations for Two Categorical Variables
Build and interpret two-way tables, side-by-side bar charts, segmented bar charts, and mosaic plots to compare distributions of two categorical variables.
Unit 2 teaches you how to quantify uncertainty using probability rules and models, and how those models describe the behavior of statistics computed from random samples.
Fiveable's AP Statistics Unit 2 review covers 12 topics, with guided practice questions for the unit.
Build and interpret two-way tables, side-by-side bar charts, segmented bar charts, and mosaic plots to compare distributions of two categorical variables.
Calculate joint, marginal, and conditional relative frequencies from two-way tables and use them to evaluate association between two categorical variables.
Design simulations to estimate probabilities, interpret relative frequency as an estimate of true probability, and apply the Law of Large Numbers.
Define sample spaces, assign probabilities using equally likely outcomes, and apply the complement rule P(not E) = 1 - P(E).
Identify disjoint events using joint probability, explain why P(A and B) = 0 for mutually exclusive events, and distinguish this from independence.
Calculate P(A|B) using the formula P(A and B) / P(B), apply the general multiplication rule, and use two-way tables or tree diagrams to organize conditional probability problems.
Test independence using conditional probability, apply the multiplication rule for independent events, and use the addition rule P(A or B) = P(A) + P(B) - P(A and B).
Define discrete random variables, construct valid probability distributions as tables or graphs, and read cumulative probability distributions.
Calculate and interpret the expected value mu_X and standard deviation sigma_X of a discrete random variable using the weighted sum formulas.
Verify BINS conditions, calculate binomial probabilities using C(n,x) p^x (1-p)^(n-x), and find the mean np and standard deviation sqrt(np(1-p)).
Apply the Empirical Rule, convert values to z-scores, find probabilities as areas under the normal curve, and determine percentile cutoffs using inverse normal calculations.
Describe sampling distributions through simulation, state and apply the CLT for sample means, and explain how sample size affects the shape and spread of a sampling distribution.
Try AP-style questions on unit 2, then use the review notes below for anything you miss.
A timed, AP-style test built only from unit 2 questions — multiple choice plus free response, scored with AI feedback. Use the standard format or build your own.
A two-way table (contingency table) cross-classifies two categorical variables. Cells hold counts or relative frequencies. Three types of relative frequency matter for comparing variables and justifying claims about association.
Given a two-way table, can you calculate a joint, marginal, and conditional relative frequency and use the conditional frequencies to argue whether the two variables appear associated?
| Frequency type | Denominator | What it describes |
|---|---|---|
| Joint | Table total | Proportion in one specific cell |
| Marginal | Table total | Overall distribution of one variable |
| Conditional (row) | Row total | Distribution of column variable within one row category |
| Conditional (column) | Column total | Distribution of row variable within one column category |
Simulation estimates probability by running a random process many times and recording relative frequencies. Formal probability assigns numbers to events using a sample space and fixed rules.
Can you design a simulation to estimate a probability, explain why more trials improve the estimate, and apply the complement rule to a basic event?
| Approach | Source | When to use |
|---|---|---|
| Simulation / empirical | Observed trials or technology | True probability unknown or complex |
| Theoretical | Sample space structure | Equally likely outcomes, simple events |
These three topics give you the full toolkit for combining and conditioning probabilities. Confusing mutually exclusive with independent is the most common error in this section.
Can you distinguish mutually exclusive from independent events, apply the general multiplication rule, and use the addition rule correctly when events overlap?
| Property | Definition | Key formula |
|---|---|---|
| Mutually exclusive | Cannot occur together | P(A and B) = 0 |
| Independent | Knowing one does not change the other | P(A and B) = P(A) x P(B) |
| Neither | Can occur together and are related | Use P(A|B) = P(A and B) / P(B) |
A discrete random variable X assigns a numerical value to each outcome of a random process. Its probability distribution lists every possible value and the probability of each. Parameters describe the center and spread of that distribution.
Given a probability distribution table, can you verify it is valid, compute the expected value and standard deviation, and interpret both in context?
The binomial model applies when a fixed number of independent trials each have exactly two outcomes (success or failure) with the same probability of success. The acronym BINS summarizes the four conditions: Binary outcomes, Independent trials, fixed Number of trials, Same probability of success.
Can you verify all four BINS conditions, calculate a binomial probability using the formula, and compute and interpret the mean and standard deviation in context?
The normal distribution is a continuous, symmetric, bell-shaped model fully described by its mean mu and standard deviation sigma. Probability equals area under the curve. You find areas using z-scores with a standard normal table or technology.
Can you apply the Empirical Rule, convert a value to a z-score, find the probability that X falls in an interval, and find the value corresponding to a given percentile?
A sampling distribution describes how a sample statistic varies across all possible samples of size n from a population. The Central Limit Theorem (CLT) states that for large enough n, the sampling distribution of the sample mean is approximately normal regardless of the population shape.
Can you describe what a sampling distribution represents, state the CLT conditions and conclusion, and explain how sample size affects the shape and spread of the sampling distribution of x-bar?
| Term | Definition |
|---|---|
| Law of Large Numbers | As the number of independent trials increases, the observed relative frequency of an outcome converges to its true probability. |
| Empirical Rule | For a normal distribution, approximately 68% of values fall within 1 standard deviation of the mean, 95% within 2, and 99.7% within 3. Also called the 68-95-99.7 rule. |
| Simpson's Paradox | A trend seen in aggregated data reverses or disappears when the data are broken into subgroups; a reminder to examine conditional distributions rather than only marginal totals. |
| Bayes' theorem | A formula for updating a conditional probability with new information: P(A|B) = P(B|A) × P(A) / P(B), applied when reversing the direction of conditioning. |
| law of total probability | The total probability of an event equals the sum of its probabilities across each partition of the sample space, weighted by the probability of each partition. |
Mutually exclusive events have P(A and B) = 0, which actually means they are dependent (knowing A occurred tells you B cannot occur). Independent events can both happen; they just do not influence each other's probability. These two properties are almost never both true at the same time for events with positive probability.
P(A|B) divides by P(B), not by the table total. When working from a two-way table, restrict your denominator to the row or column total for the given condition, not the grand total.
The binomial formula only applies when all four conditions hold: binary outcomes, independent trials, fixed n, and constant p. Sampling without replacement from a small population violates independence, so the binomial is only approximately valid when the sample is less than 10% of the population.
The normal model is a continuous approximation. When a problem asks for the probability of an exact value from a continuous distribution, the answer is 0. Probabilities correspond to intervals, not single points.
The expected value mu_X is the long-run average over many repetitions, not a prediction for any single trial. For a binomial with n = 10 and p = 0.3, the expected value is 3, but 3 successes is not guaranteed on any given set of 10 trials.
Free-response questions frequently present a two-way table and ask you to determine whether two categorical variables are associated. The expected task is to compute and compare conditional relative frequencies across groups, then write a conclusion in context. Saying the distributions look different is not enough; you must reference specific numerical evidence from the conditional distributions.
Multiple-choice and free-response items test whether you can select and correctly apply the right probability rule. Common task patterns include identifying whether events are independent or mutually exclusive, setting up a conditional probability from a two-way table, and using the addition rule without double-counting. Writing probability notation such as P(A|B) and showing the formula before substituting numbers earns credit on free-response items.
The binomial, normal, and sampling distribution topics in Unit 2 reappear directly in Units 3 and 4 as the basis for confidence intervals and hypothesis tests. Exam questions may ask you to verify conditions for using a normal approximation, interpret a standard error, or explain what the Central Limit Theorem guarantees about the shape of a sampling distribution. Precise language about shape, center, and spread of a sampling distribution is expected.

Unit 2 printables
Print the unit cheatsheets ($3) or full-length exam ($12) for Unit 2. Yours to keep, no subscription needed.
get printablesBy the end of Unit 2, you should be able to translate a context into events, use probability rules without mixing up conditional and joint probability, calculate binomial or normal probabilities, and describe the shape, center, and variability of a sampling distribution. The exam rewards the method and the interpretation, so practice writing what a result means in the problem's context instead of stopping at a calculator output.
Common Unit 2 mistakes include assuming mutually exclusive events are independent, reversing a conditional probability, using a binomial model without checking its conditions, or forgetting that a sampling-distribution standard deviation depends on sample size. Slow down long enough to identify the variables, population, and requested conclusion before calculating.
Start with the vocabulary and conditions, then mix short calculations with full-sentence interpretations. For each missed question, label whether the problem was choosing a method, doing the calculation, or interpreting the result. Finish with timed mixed practice so you have to recognize the method without a topic label.
Sampling distributions and the central limit theorem provide the probability foundation for the confidence intervals and hypothesis tests in Units 3 and 4.