Overview
The false positive paradox is the surprising fact that most positive results can be wrong, even for a good test, when the condition being tested for is rare.
Enter how common the condition is, the test's sensitivity and its false positive rate.
Worked examples
| Inputs | Result |
|---|---|
| Prior Probability of Condition (%)2Sensitivity (%)92False Positive Rate (%)8 | Posterior probability P(ConditionPositive): 19.01%False-alarm share among positives: 80.99% |
Show 2 more examples
| Prior Probability of Condition (%)0.1Sensitivity (%)99False Positive Rate (%)1 | Posterior probability P(ConditionPositive): 9.02%False-alarm share among positives: 90.98% |
| Prior Probability of Condition (%)10Sensitivity (%)95False Positive Rate (%)5 | Posterior probability P(ConditionPositive): 67.86%False-alarm share among positives: 32.14% |
How it works
Bayes' theorem combines these three numbers. True positives come from the small group with the condition, while false positives come from the much larger group without it. When the condition is rare, the false positives can easily outnumber the true ones.
The formula
| p | Prevalence, the prior probability |
| FPR | False positive rate |
Practical tips
- With 2% prevalence, 92% sensitivity and an 8% false positive rate, only 19.01% of positives are real.
- Picture 10,000 people: 184 true positives against 784 false ones.
- Testing only people with symptoms raises the prevalence, which makes positives far more trustworthy.
Frequently asked questions
Does this mean the test is bad?
Not necessarily. It means a single positive for a rare condition needs confirming.
Where does this matter?
Medical screening, security alerts, drug testing and fraud detection.
How do I reduce false alarms?
Use a more specific test, a confirmatory second test, or test higher-risk groups.