Conditional Relative Frequency in AP Statistics

A conditional relative frequency is a relative frequency for a specific part of a contingency table, found by dividing a cell count by its row total or column total instead of the grand total. In AP Stats Topic 2.3, comparing these proportions across groups is how you check for association.

What is Conditional Relative Frequency?

A conditional relative frequency answers the question "out of just this group, what proportion falls in this category?" You take a cell count from a two-way (contingency) table and divide it by the total for that row or that column, not by the grand total. The word "conditional" means you've restricted your attention to one slice of the data, like only the Republicans, or only the patients on Treatment A.

Here's the intuition. A two-way table has three kinds of percentages hiding in it. Divide a cell by the grand total and you get a joint relative frequency. Divide a row or column total by the grand total and you get a marginal relative frequency. Divide a cell by its own row or column total and you get a conditional relative frequency. That last one is the workhorse, because comparing conditional relative frequencies across groups (say, the percent who support a policy among Democrats vs. among Republicans) is how you decide whether two categorical variables are associated. If the conditional distributions look different across groups, the variables are associated. If they're roughly the same, there's no evidence of association.

Why Conditional Relative Frequency matters in AP Statistics

This term lives in Unit 2 (Exploring Two-Variable Data), Topic 2.3 (Statistics for Two Categorical Variables). It directly supports learning objective 2.3.A, which has you calculate statistics from a two-way table, and 2.3.B, which has you compare those statistics to judge whether two categorical variables are associated. Conditional relative frequencies are the bridge between those two skills. Calculating them is arithmetic; comparing them across groups is statistical reasoning, and that comparison is exactly what "checking for association" means in this unit. The same logic resurfaces later as conditional probability and independence in Unit 4, so the better you understand it here, the easier probability gets.

How Conditional Relative Frequency connects across the course

Marginal Frequency (Unit 2)

Marginal relative frequencies use the grand total as the denominator; conditional relative frequencies use a row or column total. Same table, same cells, different denominator. The marginal distribution tells you about one variable alone, while the conditional distribution tells you how that variable behaves within a specific group.

Contingency Table (Unit 2)

Conditional relative frequencies don't exist without a two-way table to compute them from. The table is the raw material; conditioning on a row or column is what turns those counts into a comparison you can actually interpret.

Joint Frequency (Unit 2)

A joint frequency counts people in a single cell (e.g., Republicans who support the policy). Divide that by the grand total and it's a joint relative frequency. Divide it by the row total instead and it becomes conditional. One cell, two very different stories depending on the denominator.

Conditional Probability (Unit 4)

Conditional relative frequency is conditional probability in disguise. P(Support | Republican) is just the conditional relative frequency of supporters among Republicans. Unit 4's independence check, comparing P(A | B) to P(A), is the probability version of comparing a conditional distribution to a marginal one.

Is Conditional Relative Frequency on the AP Statistics exam?

On multiple choice, expect a filled-in two-way table with a stem like "if a person is randomly selected from those who prefer Brand B, what proportion..." That "from those who" phrasing is your cue to condition, meaning your denominator is the Brand B total, not the grand total. Other stems hand you percentages within a group ("40% of Republicans support the policy") and ask you to reason backward to counts or to compare groups. The most common trap is dividing by the wrong total, so always ask yourself "out of whom?" before you divide. On FRQs, the comparison version shows up. You may be asked whether two categorical variables appear associated, and the expected answer compares conditional relative frequencies across groups (with actual numbers) and states whether the distributions differ.

Conditional Relative Frequency vs Marginal Relative Frequency

Both come from the same two-way table, and both are proportions, which is why they get mixed up. A marginal relative frequency divides a row or column total by the grand total and describes one variable by itself (e.g., 40% of all patients got Treatment A). A conditional relative frequency divides a cell by its row or column total and describes one variable within a slice of the other (e.g., 40% of Treatment A patients improved). Quick test: if the denominator is the whole table, it's marginal; if the denominator is one group, it's conditional.

Key things to remember about Conditional Relative Frequency

• A conditional relative frequency is a cell count divided by its row total or column total, not the grand total.

• The phrase "of those who" or "among" in a question is your signal to condition, and it tells you which group's total goes in the denominator.

• Comparing conditional relative frequencies across groups is how you determine whether two categorical variables are associated (LO 2.3.B).

• If the conditional distributions are roughly the same for every group, the variables show no evidence of association; if they differ noticeably, the variables are associated.

• Joint, marginal, and conditional relative frequencies all come from the same table; only the denominator changes.

• This idea returns in Unit 4 as conditional probability, where P(A | B) is computed exactly the same way.

Frequently asked questions about Conditional Relative Frequency

What is conditional relative frequency in AP Stats?

It's a proportion computed within one group of a two-way table, found by dividing a cell count by that cell's row or column total. For example, if 64 of 160 Treatment A patients improved, the conditional relative frequency of improvement given Treatment A is 64/160 = 0.40.

What's the difference between conditional and marginal relative frequency?

Marginal relative frequencies divide row or column totals by the grand total and describe one variable alone. Conditional relative frequencies divide a cell by its row or column total and describe one variable within a specific group of the other variable.

Do equal conditional relative frequencies mean the variables are independent?

For sample data, the careful wording is that the variables show no evidence of association. If the conditional distributions of one variable are identical across all groups of the other, that's the Unit 2 picture of "not associated," and it matches the formal independence definition you'll see in Unit 4.

How do I know which total to divide by in a two-way table?

Read the condition. "Of those who prefer Brand B" means you divide by the Brand B total. The group named after "of" or "given" or "among" always supplies the denominator; the category you're asked about supplies the numerator.

Is conditional relative frequency the same as conditional probability?

Essentially yes, just in different units. In Unit 2 it's a descriptive proportion from a table; in Unit 4 the same calculation becomes P(A | B) = P(A and B) / P(B). If you can read a two-way table conditionally, you already know how to compute conditional probability.