Free explainer2 min read
Specificity calculation
Understand the model, recognise it in a stem, separate the look-alikes, then apply it.
Start with the mental model
SpecificityThe proportion of people who truly do not have a condition that a test correctly identifies as negative.Read more → is sensitivity'sThe proportion of people who truly have a condition that a test correctly identifies as positive.Read more → mirror: of everyone who genuinely does NOT have the condition, what proportion does the test correctly clear? To find it, you look only down the 'non-diseased' column of a 2×2 table. For example, imagine testing 900 people confirmed disease-free: if the test correctly clears 810 of them (true negatives) and wrongly flags 90 as positive (false positives), specificity is 810/900 = 90%. In practice, this means specificity, like sensitivity, is a property of the TEST, fixed by how good it is at leaving healthy people alone, and it does not change with how common the disease happens to be in the population tested.
01Core model
• SpecificityThe proportion of people who truly do not have a condition that a test correctly identifies as negative.Read more → = TN / (TN + FP): of those WITHOUT disease, how many test negative.
• Only the NON-DISEASED column of the 2×2 table is used, true negatives and false positives; the diseased column plays no part.
• Worked example, using the same 1000-person table sensitivityThe proportion of people who truly have a condition that a test correctly identifies as positive.Read more → uses: 810 non-diseased-and-test-negative and 90 non-diseased-and-test-positive gives specificity = 810 / (810 + 90) = 90%.
• A specificity of 100% (zero false positives) means every disease-free person tested negative, not that the test caught every true case (that is sensitivity's job).
• In practice, this means before calculating, first isolate the non-diseased column, then divide the true negatives in it by the column's own total.
02Memory anchor
SpecificityThe proportion of people who truly do not have a condition that a test correctly identifies as negative.Read more → looks only at people who are disease-FREE: TN over (TN + FP).