4.5 Conditional Probability

What is conditional probability in AP Statistics?

Conditional probability is the chance that one event happens given that another event has already happened, written P(A | B) = P(A ∩ B) / P(B). When you see the word "given," restrict your attention to just the group named after "given" and find what fraction of that group fits the event you care about.

Why This Matters for the AP Statistics Exam

Conditional probability shows up across multiple-choice and free-response questions, especially with two-way tables and tree diagrams. You will be asked to calculate probabilities from data, recognize when one event depends on another, and explain your reasoning clearly. On free-response questions, showing the structure of your work matters: write the expression you are using, substitute the correct values, and report a final answer. This skill also sets up later topics like independence, random variables, and statistical inference, so getting comfortable with it now pays off all year.

Key Takeaways

• The conditional probability formula is P(A | B) = P(A ∩ B) / P(B). You divide the joint probability by the probability of the condition.
• The word "given" is your signal to use conditional probability and to restrict the sample space to the condition.
• The general multiplication rule, P(A ∩ B) = P(A) · P(B | A), comes directly from rearranging the conditional probability formula.
• In a two-way table, condition on a row or column total, then divide the matching cell by that total.
• Conditional probability is not the same as multiplying two marginal probabilities, unless the events are independent.
• Show your expression, your substituted values, and your final answer on free-response work.

Representations of Probability

Several visuals help you set up and calculate probabilities:

• Probability histograms: bar graphs where the x-axis shows possible outcomes and the y-axis shows the probability of each.
• Tree diagrams: diagrams that show outcomes across a series of stages, like rolling a die and then drawing a card. The probabilities on later branches are usually conditional probabilities based on what happened earlier.
• Venn diagrams: overlapping circles inside a rectangle that show how sets of outcomes relate, which makes joint and conditional probabilities easier to see.
• Two-way tables: tables that organize counts for two categorical variables, which makes joint, marginal, and conditional probabilities quick to read off.

For conditional probability, two-way tables and tree diagrams are the workhorses. Two-way tables and Venn diagrams help with joint probabilities (the probability of two events happening together) and conditional probabilities. Tree diagrams help you track sequences of events and multiply along branches.

Conditional Probability: Multiply and Divide

Conditional probability is the probability of an event occurring given that another event has already occurred. It is written P(A | B), the probability of A given B.

The formula is:

$P(A \mid B) = \frac{P(A \cap B)}{P(B)}$

And the other direction:

$P(B \mid A) = \frac{P(A \cap B)}{P(A)}$

Rearranging gives the general multiplication rule:

$P(A \cap B) = P(A) \cdot P(B \mid A)$

This says the probability that both A and B happen equals the probability of the first event times the probability of the second event given the first has happened.

Just like "OR" points to unions and "AND" points to intersections, the word "given" is your cue to use conditional probability.

Practice Problem 1

A biotechnology company is developing a new gene therapy to treat a rare genetic disease. The clinical trial included 100 patients with the disease, and the results were as follows:

Find (A) the probability that a patient responds to the therapy, given that they experienced side effects, and (B) the probability that a patient does not respond to the therapy, given that they did not experience side effects.

Answer

Use the conditional probability formula:

$P(A \mid B) = \frac{P(A \cap B)}{P(B)}$

In a two-way table, this means dividing the relevant cell count by the total for the given condition.

(A) Among the 50 patients who experienced side effects, 40 responded:

$P(\text{response} \mid \text{side effects}) = \frac{40}{50} = 0.80$

So among patients who experienced side effects, 80% responded to the therapy.

(B) Among the 50 patients who did not experience side effects, 15 did not respond:

$P(\text{no response} \mid \text{no side effects}) = \frac{15}{50} = 0.30$

So among patients who did not experience side effects, 30% did not respond to the therapy.

The key idea: conditional probabilities are calculated within the group named after "given," not by multiplying two marginal probabilities.

Practice Problem 2

A consumer research company surveys 1000 customers who bought clothes from a particular brand in the past year. The data are organized in this two-way table:

Find (A) the probability that a customer is satisfied, given that they would definitely purchase again, and (B) the probability that a customer is not satisfied, given that they would consider purchasing again.

Answer

(A) Restrict to the 600 customers who would definitely purchase again. Of those, 350 are satisfied:

$P(\text{satisfied} \mid \text{definitely purchase}) = \frac{350}{600} \approx 0.583$

So about 58.3% of customers who would definitely purchase again are satisfied.

(B) Restrict to the 400 customers who would consider purchasing again. Of those, 250 are not satisfied:

$P(\text{not satisfied} \mid \text{consider purchase}) = \frac{250}{400} = 0.625$

So 62.5% of customers who would consider purchasing again are not satisfied.

Notice that you condition on a row total and divide the matching cell by it. You cannot find these answers by multiplying the two separate (marginal) probabilities, because satisfaction and purchase intention are not assumed to be independent here.

How to Use This on the AP Statistics Exam

MCQ

• When you see "given," restrict the sample space to the condition before you compute anything.
• For a two-way table, find the row or column total that matches the condition, then divide the matching cell by that total.
• Watch for answer choices that come from multiplying two marginal probabilities. That only works when the events are independent.

Free Response

• Write the conditional probability expression first, for example P(A | B) = P(A ∩ B) / P(B), so your structure is clear.
• Substitute the actual values you pulled from the problem, then give a final numerical answer.
• If you use a tree diagram, label each branch with the correct probability and make sure later branches use conditional probabilities.
• Interpret your answer in context when the question asks for it, using the right group and units.

Common Trap

• Reversing the condition. P(A | B) and P(B | A) are usually different. Read carefully to see which event is given.

Common Misconceptions

• Multiplying marginals instead of conditioning. P(A | B) is not P(A) · P(B). You divide the joint probability by the probability of the condition. Multiplying marginals only gives the joint probability when the events are independent.
• Confusing P(A | B) with P(B | A). These answer different questions and usually have different values. The event after the bar is the one that has already happened.
• Using the grand total as the denominator. For a conditional probability, the denominator is the total for the given group, not the total sample size.
• Assuming independence automatically. Do not assume two events are independent unless the problem states it or you verify it. Conditional probabilities often change depending on the condition.
• Skipping the setup on free response. An answer with no expression or substituted values does not show your reasoning. Write the formula, plug in, and report the result.

Related AP Statistics Guides

• Unit 4 Overview: Probability, Random Variables, and Probability Distributions
• 4.1 Introducing Statistics: Random and Non-Random Patterns?
• 4.2 Estimating Probabilities Using Simulation
• 4.3 Introduction to Probability
• 4.4 Mutually Exclusive Events
• 4.9 Combining Random Variables