Conditional probability asks a specific question: once you know that one event has happened, what is the probability of another event? The additional information changes which outcomes remain possible. Your denominator must describe that restricted set of outcomes.
This is why “the probability of A given B” is usually different from “the probability of B given A.” Both calculations may use the same overlapping group, but they compare that group with different totals.
The most useful first step is to translate the condition into ordinary language. Who or what is still eligible to be selected? Which outcomes have been ruled out? Only after answering those questions should you select a formula. For broader topic support, see AS Level Math S1 at Abeel Eschool.
What “given” changes in a probability question
Imagine a class of forty students. Eighteen study a particular optional subject. Of those eighteen, twelve also belong to a study club. If a student is selected from the whole class, the probability of studying the subject and belonging to the club is 12/40.
Now you are told that the selected student studies that subject. Your possible group has narrowed from forty students to eighteen. The probability that the student belongs to the club, given that they study the subject, is 12/18 = 2/3.
The twelve students in the overlap have not changed. What changed is the reference group. You are no longer comparing them with everyone in the class.
Underline the condition during revision. Then write “among these outcomes” before naming the group. This makes the denominator a meaning you can explain rather than a number selected from memory.
Use the conditional-probability formula carefully
For an event B with positive probability, P(A given B) = P(A and B) / P(B). The numerator is the probability that both requirements hold. The denominator is the probability of the information already given.
In the class example, let A mean membership of the club and B mean studying the optional subject. Then P(A and B) = 12/40 and P(B) = 18/40. Dividing gives (12/40)/(18/40) = 12/18.
The formula and restricted-group method give the same result. The formula is useful for more complicated questions, while the group interpretation helps you check whether it has been applied sensibly.
If the conditioning event has probability zero in the elementary setting being used, this ratio is not defined. Do not insert a zero denominator into a calculator and treat its error message as an answer.
A two-way table keeps the groups visible
Build an original example with fifty students. Twenty study Art, fifteen study Music, and eight study both. The Art-only group contains twelve, the Music-only group contains seven, and twenty-three study neither.
Check the totals: eight plus twelve plus seven plus twenty-three equals fifty. There are twenty Art students and fifteen Music students, as stated.
The probability of Music given Art is 8/20 = 2/5. The probability of Art given Music is 8/15. The same eight students appear in both numerators, but the conditions choose different denominators.
A table is especially helpful when a question gives totals and overlaps in words. Fill the overlapping cell first when appropriate, calculate the remaining cells, and confirm that every row and column total agrees with the information supplied.
Reversing the condition changes the question
A common mistake is to see the same two event names and assume their conditional probabilities are interchangeable. They are not. “Among Art students, how many study Music?” differs from “Among Music students, how many study Art?”
Use a verbal check before writing the fraction. The phrase following “given” names the group you are inside. If the question says “given that the student studies Music,” your denominator must represent Music students.
This distinction also matters in interpretation. A large proportion of one small group may belong to a larger group, while only a small proportion of the larger group belongs to the smaller one.
Do not rely on symmetry in a diagram. Two overlapping circles drawn with similar sizes do not establish that the underlying event probabilities are equal. Use the actual counts or probabilities.
Conditional probability on a tree diagram
A tree describes stages. The probability on a second-stage branch is conditional on reaching the point where that branch begins. It may differ after different first-stage outcomes.
Suppose a bag contains four blue and three yellow counters. Two are drawn without replacement. After a blue first draw, the probability of a blue second draw is 3/6. After a yellow first draw, the probability of a blue second draw is 4/6.
Both branches describe “blue on the second draw,” but they have different histories. Write the remaining counts beside each branch rather than copying the same fraction everywhere.
The probability of blue then blue is 4/7 × 3/6 = 2/7. The probability of yellow then blue is 3/7 × 4/6 = 2/7. These equal products happen in this example; that equality should not be assumed for unrelated questions.
A later condition can require several complete paths
Continue with the same bag. Suppose you know that the second counter is blue and want the probability that the first counter was yellow.
The qualifying joint event is yellow then blue, with probability 2/7. The conditioning event “second is blue” includes both blue-then-blue and yellow-then-blue. Its total probability is 2/7 + 2/7 = 4/7.
The required conditional probability is therefore (2/7)/(4/7) = 1/2. Reading only one second-stage branch would miss the fact that the condition can be reached through different histories.
This illustrates a useful tree method. Mark every complete path that satisfies the condition. Add their probabilities for the denominator. Then identify which of those paths also satisfies the target event for the numerator.
Independence is a statement about unchanged probability
If A and B are independent, then P(A given B) = P(A), provided the condition has positive probability. Learning B does not change the probability of A.
You can test independence using P(A and B) = P(A)P(B). In the fifty-student Art and Music example, P(Art) = 20/50 and P(Music) = 15/50. Their product is 0.12, while the overlap probability is 8/50 = 0.16. Those events are not independent in this model.
Do not confuse “not independent” with “mutually exclusive.” Art and Music overlap, so they can occur together. Dependence means the probability changes under the condition; mutual exclusivity means the joint event cannot occur.
When the question states an assumption of independence, use it. When it asks you to establish independence, show an appropriate comparison rather than writing that the events seem unrelated.
Deal with complements inside the condition
Suppose the probability of A given B is 0.7. Within the same condition B, the probability of not A is 0.3. Together, A and not A exhaust the possibilities inside B.
Be careful not to move the complement to the condition itself. P(not A given B) is generally different from P(A given not B). The first asks about the opposite target within the same group. The second changes the group under consideration.
A two-way table can help you see this. Cross out the group excluded by the condition, then divide the appropriate remaining cells by the total of the group you kept.
Write the words before the notation if symbols feel ambiguous. “Among students who study Art, those who do not study Music” gives a clearer starting point than manipulating several complements without a defined group.
Check answers with a small list of outcomes
For simple equally likely cases, enumerate the restricted sample space. Suppose two fair dice are rolled independently and you are told their sum is five. The possible ordered outcomes are (1,4), (2,3), (3,2) and (4,1).
Given that condition, the probability that the first die shows one is 1/4. The denominator is four eligible ordered outcomes, not the thirty-six possible outcomes before the information was given.
The word “ordered” matters because the dice can be distinguished. Treating (1,4) and (4,1) as a single equally likely outcome without adjusting the model can distort a calculation.
Enumeration is a good revision check, but only when the listed outcomes are equally likely under the stated process. A list of outcome descriptions does not automatically make each description equally probable.
Build a review routine around denominator mistakes
After an incorrect attempt, name the denominator you used and the group it represents. Then compare it with the group selected by the condition. This often reveals the problem more quickly than reworking every calculation.
Classify mistakes into a few useful categories: reversed condition, incomplete conditioning event, wrong remaining counts, assumed independence or arithmetic. Give each entry a small follow-up task, such as writing five conditions in words before calculating.
Practise with tables, trees and outcome lists. A method that works on a table may feel less familiar on a tree, even though the same conditional-probability principle applies.
For a wider study routine, read Abeel’s Cambridge past-paper guide and the AS and A2 planning guide. Choose questions that match your current syllabus rather than relying on an S1 label alone.
Ask a teacher about your event definition
A precise question makes feedback easier. Send your event labels, the condition in words and your proposed numerator and denominator. Ask which group has been restricted incorrectly.
After receiving feedback, attempt a fresh question with the condition reversed. This tests whether you understand the reference group or have only memorised one example.
Abeel Eschool provides a starting point for exploring online subject learning. Check the course catalogue or admissions and FAQs for current options. Confirm the coverage and support relevant to your examination plan.
Frequently asked questions
What does “given” mean in conditional probability?
It means additional information has restricted the possible outcomes. Calculate the target probability inside the group described by that information.
How do I choose the denominator?
Use the total count or probability of the conditioning event. Translate the condition into words before selecting a number.
Is P(A given B) equal to P(B given A)?
Not generally. They often share a joint-event numerator but use different conditioning denominators. Check each question independently.
Can I multiply probabilities for dependent events?
Yes, when the second factor is the appropriate conditional probability. P(A and B) equals P(A) times P(B given A).
Why can one condition use several paths on a tree?
A condition such as “second draw is blue” may occur after different first draws. Include every qualifying complete path when calculating that condition’s probability.
Have a question about this topic?
Send your query to Abeel Eschool and our admissions team will get back to you.