Skip to content
AP Statistics · Probability, Random Variables, and Probability Distributions

Conditional probability

Conditional probability appears in 25 reviewed AP Statistics questions and sits in Probability, Random Variables, and Probability Distributions. Below are worked examples showing exactly how the exam tests it, with the answer marked and every distractor explained.

How Conditional probability is tested

Example 1Harder
A bag contains 4 red and 6 blue marbles. The Law of Total Probability states P(red on 2nd draw) = P(red 2nd | red 1st)×P(red 1st) + P(red 2nd | blue 1st)×P(blue 1st). If draws are WITHOUT replacement, what is P(red on the 2nd draw)?
  1. A3/10
  2. B4/10Correct
  3. C16/90
  4. D6/45

Why B is right

Using the Law of Total Probability: P(red 2nd) = P(red 2nd|red 1st)×P(red 1st) + P(red 2nd|blue 1st)×P(blue 1st) = (3/9)(4/10) + (4/9)(6/10) = 12/90 + 24/90 = 36/90 = 4/10 = 0.40. By symmetry, the probability of drawing red on any draw equals 4/10, the original proportion.
Why the other 3 choices miss
A
3/10 would be P(red on 2nd | red on 1st), the conditional probability given the first was red. This is not the unconditional probability for the 2nd draw.
C
16/90 = P(red 1st)×P(red 2nd|red 1st) = (4/10)×(3/9) = 12/90, which is only one branch of the Law of Total Probability tree, not the complete calculation.
D
6/45 is not a correct calculation from these values.
Example 2Harder
A rare disease affects 1 in 5,000 people. A medical test for the disease has a 99% true positive rate (sensitivity) and a 98% true negative rate (specificity). If a randomly selected person tests positive, what is the approximate probability that the person actually has the disease?
  1. AApproximately 99%, because the test has 99% sensitivity
  2. BApproximately 50%, because either the person has it or does not
  3. CApproximately 98%, because the test has 98% specificity
  4. DApproximately 0.5%, because the disease is rare and most positives are false positivesCorrect

Why D is right

Using Bayes' theorem: P(disease | positive) = P(positive | disease) * P(disease) / P(positive). P(disease) = 1/5000 = 0.0002. P(positive | disease) = 0.99. P(positive | no disease) = 0.02. P(positive) = 0.99(0.0002) + 0.02(0.9998) = 0.000198 + 0.019996 = 0.020194. P(disease | positive) = 0.000198/0.020194 ≈ 0.0098, or about 1%. Even with a very accurate test, the extreme rarity of the disease means most positive results are false positives.
Why the other 3 choices miss
A
Incorrect. The 99% sensitivity describes the test's ability to detect the disease when present, not the probability of having the disease given a positive test. When the disease is rare, these are very different quantities.
B
Incorrect. The probability is not 50/50. The base rate (prevalence) of the disease strongly influences the posterior probability through Bayes' theorem.
C
Incorrect. The 98% specificity means 98% of healthy people test negative, but the 2% who test positive vastly outnumber the true positives because healthy people far outnumber sick people.

Studied alongside

Other concepts from Probability, Random Variables, and Probability Distributions.