Sensitivity (True Positive Rate)
Definition
The proportion of true matches correctly identified as matches. Sensitivity = TP / (TP + FN) = 1 minus FNR. A method with high sensitivity rarely misses a true match.
- Formula
- TP / (TP + FN)
- Alternate name
- True positive rate
- Relation to FNR
- Sensitivity = 1 minus false negative rate
- High sensitivity means
- Method rarely misses a true match
- Field
- Error rate statistics in forensic identification
Common questions
Can a method have high sensitivity but still be unreliable overall?+
Yes, sensitivity only describes performance on samples that are truly matches. A method can rarely miss a true match yet still produce many false positives on non-matching samples, so sensitivity must be read alongside specificity and the false positive rate.
Why does sensitivity matter more in some forensic contexts than others?+
In contexts like missing-persons or mass-disaster identification, missing a true match has severe consequences, so labs may tune methods toward higher sensitivity even at some cost to specificity, while in contexts prioritising against wrongful implication, the balance shifts the other way.
How is sensitivity typically established for a forensic comparison method?+
It is estimated through validation studies using known-match sample sets under realistic casework conditions, counting how often the method correctly calls a true match versus how often it produces a false negative, then reporting the resulting rate with its confidence interval.
Related terms
- Decision Threshold
- The cut-off score above which a classification method declares a positive result. Raising the threshold reduces FPR but increases FNR; lowering it...
- False Negative Rate (FNR)
- The proportion of true matches (or true positives) that a classification method fails to detect, declaring them as non-matches. Also called the...
- False Positive Rate (FPR)
- The proportion of true non-matches (or true negatives) that a classification method declares as matches (or positives). Also called the type I...
- ROC Curve
- Receiver operating characteristic curve: a plot of sensitivity (y-axis) against FPR (x-axis) as the decision threshold is swept from its most lenient...
- Specificity (True Negative Rate)
- The proportion of true non-matches correctly identified as non-matches. Specificity = TN / (TN + FP) = 1 minus FPR. A method...