Finding Conditional Probability From a Two-Way Table

A two-way table organizes counts by two categories. Conditional probability asks about one category after you know the other. The word “given” changes the denominator: instead of using the whole table, use only the row or column representing the known condition.

Read the table and identify the condition

Suppose the question asks for the probability a student passed, given that the student studied with a group. The condition is “studied with a group,” so the denominator is 24, not 50. The overlap count for both “passed” and “studied with a group” is 18. Therefore, the conditional probability is 18/24 = 3/4.

A reliable setup

1. Circle the condition after the word “given.”
2. Find the row or column total for that condition.
3. Find the cell where the condition and event intersect.
4. Divide the intersection count by the condition total.
5. Reduce the fraction or convert to a decimal or percent if requested.

Why reversing the condition changes the answer

The probability of passing given group study is 18/24. The probability of group study given passing is 18/30. The numerator is the same overlap, but the denominator changes because the known information changes. These conditional probabilities are not generally equal.

Common errors

• Using the grand total instead of the condition’s row or column total.
• Choosing the wrong cell because the row and column categories are reversed.
• Confusing P(A given B) with P(B given A).
• Reporting the overlap count without converting it into a probability.

Worked example with all four cells

Suppose a table records 50 students: 18 passed after studying in a group, 6 did not pass after group study, 12 passed without group study, and 14 did not pass without group study. The row total for group study is 24; the column total for passing is 30. For the question “passed, given group study,” keep only the 24 students who studied in a group. Eighteen of those 24 passed, so P(pass | group study) = 18/24 = 3/4 = 75 percent. The whole-table total of 50 is not the denominator because the question supplies the group-study condition.

If the question reverses the condition—group study given that the student passed—the same intersection of 18 is divided by 30 passing students, giving 18/30 = 3/5 = 60 percent. The overlap does not change, but the restricted group does. That is why “A given B” and “B given A” should be read carefully rather than treated as interchangeable phrases.

Connect conditional, joint, and overall probability

The intersection count represents both conditions happening together. In the example, the joint probability of passing and group study is 18/50 = 36 percent. The probability of group study overall is 24/50 = 48 percent, while the probability of passing overall is 30/50 = 60 percent. These answer three different questions: both events, the condition alone, and the outcome alone. A table helps because each margin and interior cell has a different role.

The formula P(A | B) = P(A and B)/P(B) expresses the same relationship when probabilities rather than counts are used. The denominator must be greater than zero; a conditional probability is undefined if no observations meet the stated condition. For a table of counts, divide the cell count by the total count in the conditioned row or column. For a table of percentages, check whether the percentages are row-based, column-based, or whole-table values before dividing again.

Check whether the condition changes the likelihood

Compare the conditional probability with the overall probability. In this example, passing is 75 percent among group-study students but 60 percent among all students, so the observed groups differ. That comparison alone does not establish that group study caused the difference. Students may differ in preparation, prior achievement, or other factors. A two-way table describes association; causal conclusions need evidence about how the groups were formed and what else differed.

Read tables with awkward layouts

• If the condition appears as a column heading, use that column total; if it appears as a row heading, use that row total.
• If the prompt says “neither,” “only,” or “at least one,” identify the relevant union or exclusive region before choosing a cell.
• If the prompt gives a conditional percentage directly, check which group it is conditional on. A value such as 75 percent may already be a row percentage rather than a raw count.
• Keep counts and probabilities distinct: 18 students is a count, while 18/24 is a probability within a selected group.

A final denominator check

Before calculating, rewrite the phrase after “given” as the denominator group. Then point to its margin and locate the intersection that also meets the event. Reduce the resulting fraction and check that it lies from 0 to 1. If your answer uses the grand total despite a stated condition, revisit the setup.

Key takeaway

For P(A given B), use the B group as the denominator and the A-and-B intersection as the numerator. The condition after “given” sets the sample space.

Use the condition to choose the denominator

Conditional probability asks for the chance of event A among cases where event B is already known to have occurred. The notation is P(A given B) = P(A and B)/P(B), provided P(B) is not zero. In a two-way table, the condition determines the relevant row or column; divide the count in the intersection by the total for that condition, not by the grand total.

Suppose a table shows 30 students who study with a group, 18 of whom pass, and 20 students who study alone, 10 of whom pass. There are 50 students total and 28 pass. The probability a student passes given group study is 18/30 = 0.60. The probability of group study given passing is 18/28, a different question with a different denominator.

Read rows, columns, and totals precisely

Label which variable is in the rows and which is in the columns. The intersection cell counts both conditions, while a row or column total counts all cases in one category. If a problem asks “among those who passed,” use the pass total. If it asks “among group-study students,” use the group-study total. Translate the phrase into a denominator before calculating.

A two-way table can also test whether two categorical variables appear independent. Compare the conditional proportion within one group with the overall proportion or another group’s proportion. Similar proportions suggest independence in the data; substantial differences suggest association. A table alone does not establish that one category caused the other.

Distinguish a conditional proportion from joint probability

The joint probability of both events uses the intersection count divided by the grand total. The conditional probability uses the intersection count divided by the total in the given condition. For the example, P(group study and pass) = 18/50, while P(pass given group study) = 18/30. Confusing these quantities is a common denominator error.

• Translate the condition into the row or column that supplies the denominator.
• Use intersection count divided by condition total.
• Reverse the condition and the result only when the wording does.
• Joint probability uses the grand total; conditional probability uses a subgroup total.
• Describe association without inferring causation from the table alone.