This Exam P sample reference tests Conditional Probability. This is a Bayes-theorem update after a positive test outcome. True positives contribute 0.18 of the tested population and false positives contribute 0.16, so the posterior is 0.18/0.34=0.5294 and choice D.
These notes identify the calculation error associated with each wrong letter when that error is reproducible.
AThe value 0.10 is the false-negative probability. It concerns a negative outcome within the pregnant group, not pregnancy within the positive-result group.
BThe value 0.20 is the prior pregnancy share and also the false-positive rate here. Reporting either number without population weighting skips the Bayes update.
CThe value 0.50 comes from normalizing the prior 0.20 against the false-positive rate 0.20 as though they were comparable positive masses. The likelihood and the non-pregnant base rate are both missing.
EThe value 0.90 is the sensitivity P(+|P). It reverses the conditioning direction and is not P(P|+).
Original practice · fully worked
Original variant: two diagnostic flags from prior odds
A service incident comes from configuration R or configuration S, with prior odds R:S of 1:4. Two diagnostic flags are conditionally independent given the configuration. The first flag has likelihood ratio 3 in favor of R, and the second has likelihood ratio 2 in favor of R. Both flags appear. Calculate the posterior probability that configuration R caused the incident.
A 0.2000
B 0.3333
C 0.4286
D 0.6000
E 1.5000
Variant answer in brief
Conditional independence allows the likelihood ratios to multiply. The prior odds 1:4 become posterior odds 6:4=3:2, which convert to probability 3/5=0.6000 and choice D.
Setup
Setup
Represent the prior information as odds for R against S.
Oprior(R:S)=41
Model
Model
For conditionally independent evidence, the combined likelihood ratio is the product of the two individual ratios.
LRboth=3(2)=6
Opost(R:S)=OpriorLRboth
Compute
Compute
Update the odds and then convert odds to a probability.
Opost(R:S)=41(6)=23
Pr(R∣both flags)=3+23=53=0.6000
Answer
Answer
The posterior probability of configuration R is 0.6000.
The 3108-page Probability Proof Manual reorganizes 718 verified Exam P solutions by syllabus skill and adds formula proofs, error patterns, and original worked practice.