What Does a Union Probability Say About Students?
In the previous tutorial, Checking Whether Categories Are Truly Disjoint, you checked whether one respondent could belong to both categories. Now use that distinction to interpret a probability about two school activities. Suppose \(A\) is the event that a randomly selected student participates in one activity, and \(B\) is the event that the student participates in another.
The notation \(P(A\text{ or }B)\), also written \(P(A\cup B)\), refers to students who are in \(A\), in \(B\), or in both. Here, “or” is inclusive: a student who participates in both activities is included. The union does not mean “one activity but not the other.”
So, if a school activity dataset gives \(P(A\text{ or }B)=0.72\), a careful interpretation is: “If one student is selected at random from the students represented in this dataset, the probability that the student participates in activity A, activity B, or both is 0.72.” Because \(0.72=72\%\), you can also say that 72% of the students in the dataset participate in at least one of the activities.
The phrase at least one is a useful check on the meaning. It includes the students in the first activity only, the second activity only, and both activities. It excludes students in neither. This connects to the regions of a two-event Venn diagram in Venn Diagrams for Disjoint and Overlapping Events.
Build a Complete Interpretation
A strong interpretation names the chance process, the group being considered, and the event represented by the probability. For a school dataset, the chance process is usually selecting one student at random from a stated group. The event is participation in at least one of the two named activities.
Say that one student is selected at random from the students represented in the dataset.
Say “participates in activity A, activity B, or both,” or use “participates in at least one of the two activities.”
Give 0.72 as a probability, or convert it to 72% of the represented students.
The subject of the sentence matters. A probability is attached to the random selection, not to a named student whose activity status is already known. For example, “There is a 72% chance that Maya participates in one of the activities” is not an appropriate interpretation if Maya is a particular student in the dataset. Instead, identify the randomly selected student.
Also keep the group in view. If the dataset records 200 students from one school, the direct interpretation is about a random selection from those 200 students. It does not automatically describe every student in the district, students at other schools, or students in a later year. A broader claim would need an appropriate basis beyond this probability statement.
Worked Example: Robotics or Debate
An invented dataset records participation in robotics and debate for 200 students at a school. Of the students, 52 participate in robotics only, 36 participate in both activities, 56 participate in debate only, and 56 participate in neither. Interpret \(P(R\text{ or }D)=0.72\), where \(R\) means “participates in robotics” and \(D\) means “participates in debate.”
State: The chance process is selecting one of the 200 students at random. The event \(R\text{ or }D\) includes students in robotics, students in debate, and students in both.
Plan: Since the probability refers to the union, include the both-activities group once. Check the number of students in the union by combining robotics only, both, and debate only. Then describe the result for a random selection from this dataset.
Do: The number of students in the union is:
The four response groups account for all 200 students: \(52+36+56+56=200\). The union probability is therefore \(144/200=0.72\), which is 72%. The complement provides a check: 56 students are in neither activity, and \(1-56/200=0.72\).
Conclude: If one of the 200 students is selected at random, the probability that the student participates in robotics, debate, or both is 0.72. Equivalently, 72% of the students in this dataset participate in at least one of these activities.
“Or” Does Not Mean “Exactly One”
One common source of confusion is treating “A or B” as “A only or B only.” In ordinary probability notation, the union \(A\cup B\) includes the overlap. If the question means that exactly one activity is involved, it must exclude students who participate in both. The earlier tutorial Probability of Exactly One of Two Events treats that different event.
Likewise, “both” describes only the overlap \(A\cap B\), not the whole union. If you are asked to interpret \(P(A\text{ or }B)=0.72\), do not say that 72% participate in both activities. The probability gives the share in at least one, and the overlap may be only part of that share.
When a dataset gives activity counts, the categories may overlap. In Union Probabilities from a Two-Way Table and The General Addition Rule, you saw ways to find a union probability without counting the overlap twice. For this tutorial, the main task is to give the resulting probability a precise meaning.
Worked Example: Orchestra or School Newspaper
An invented survey records whether 150 students participate in orchestra and whether they work on the school newspaper. The dataset shows 84 orchestra participants, 66 newspaper participants, and 42 students who do both. Let \(O\) mean “participates in orchestra” and \(N\) mean “works on the school newspaper.” Interpret \(P(O\text{ or }N)\).
State: The selection is one student chosen at random from these 150 students. The requested event is participating in orchestra, working on the newspaper, or doing both.
Plan: The categories overlap, so the 42 students in both must be counted once in the union. Use the general addition rule from earlier in the course to find the union count, then express it as a proportion of the 150 students.
Do: Using the activity totals and the overlap, the union count is:
As a check, the counts are 42 orchestra only \((84-42)\), 24 newspaper only \((66-42)\), and 42 in both. Thus the union count is \(42+24+42=108\), the same result. Dividing by the dataset total gives:
Conclude: If one of the 150 students is selected at random, the probability that the student participates in orchestra, works on the school newspaper, or does both is 0.72. In this dataset, 72% of the students participate in at least one of those activities.
Probability, Percentage, and Count
The value 0.72 can be stated as a decimal probability, as 72%, or—when the dataset size is known—as a count. These forms communicate related information, but the wording should make clear which one you are giving. For a dataset of 250 students, for example, a union probability of 0.72 corresponds to \(0.72(250)=180\) students in the union.
A count is meaningful only alongside its group size. “180 students participate in at least one activity” gives a count, while “72% of the 250 students in the dataset participate in at least one activity” states a proportion. In either case, name the two activities and include students who do both.
This interpretation describes the probability model for selecting a student at random from the dataset. In this setting, each listed student is treated as equally likely to be selected. As discussed in Interpreting Probability as Long-Run Relative Frequency, probability also has a long-run relative-frequency interpretation across comparable repetitions. You do not need to imply that the one selection itself will include a fraction of a student: one selected student either belongs to the union or does not.
Worked Example: Environmental Club or Peer Tutoring
An invented dataset includes 250 students. Of these, 110 participate in the environmental club, 120 participate in peer tutoring, and 50 participate in both. Let \(E\) mean “participates in the environmental club” and \(T\) mean “participates in peer tutoring.” Find and interpret \(P(E\text{ or }T)\).
State: Select one of the 250 students at random. The event \(E\text{ or }T\) means that the student participates in the environmental club, peer tutoring, or both.
Plan: The reported totals overlap, so subtract the both-activities count once when finding the union. Then divide by 250 and interpret the result for these students.
Do: The union count is:
The exclusive-region check gives 60 environmental-club-only students, 70 peer-tutoring-only students, and 50 students in both. Their total is \(60+70+50=180\). The probability is:
Conclude: If one of these 250 students is selected at random, the probability that the student participates in the environmental club, peer tutoring, or both is 0.72. In other words, 72% of the students in this dataset participate in at least one of the two activities.
Common Mistakes and AP Exam Tips
- Leaving out the overlap in words. Saying “activity A or activity B” can sound exclusive in everyday conversation. Make the probability meaning explicit by adding “or both” or saying “at least one of the two.”
- Calling the union “both.” The union includes A only, B only, and both. The word “both” refers only to students in the intersection.
- Saying “exactly one” when the event is a union. Exactly one excludes the overlap; the union includes it. Use exactly one only when the question explicitly asks for that event.
- Forgetting the selection mechanism. A complete interpretation says one student is selected at random from the students represented in the dataset. Without that context, the sentence may sound like an unsupported claim about a different group.
- Reporting a percentage without naming the group. “72% participate” is incomplete if the reader cannot tell which students or which activities are meant. Name both activities and the dataset’s students.
- Turning a dataset probability into a claim about all students. Unless the dataset represents the wider population in an appropriate way, keep the conclusion limited to the students it contains.
A full-credit interpretation connects the probability to the random selection and states the event in context. For \(P(A\text{ or }B)=0.72\), a reliable answer says that a randomly selected student from the specified dataset has probability 0.72 of participating in A, B, or both; equivalently, 72% of those students participate in at least one.
Check Your Understanding
Write or assess an interpretation in context. Be sure to distinguish the union from “both” and “exactly one.”
- A dataset contains 100 students, and \(P(\text{choir or soccer})=0.72\). Write a sentence interpreting this probability for a student selected at random from the dataset.
- Does \(P(A\text{ or }B)=0.72\) mean that 72% participate in both activities? Explain.
- In a dataset of 200 students, 144 participate in at least one of two activities. State the union probability and interpret it.
- A student writes, “There is a 72% chance that every student at the school joins one of the clubs.” Identify what is wrong with this interpretation.
- Explain why “participates in activity A or activity B, or both” is a clearer interpretation of a union than “participates in exactly one.”