Den här sidan finns på svenska. Byt till svenska
Definition — what does that mean?
Of the people without the condition, the share the test correctly clears.
In Swedish: Specificitet
Sp
The share of the people without the condition that the test correctly clears: true negatives divided by everyone without the condition, true negatives plus false positives.
Sp=0.95
Sp equals 76 divided by (76 plus 4), equals 76 divided by 80, equals 0.95
You are given
TN, the true negatives, are the people without the condition whom the test clears; FP, the false positives, are the people without it whom the test wrongly flags. The denominator is everyone who does not have the condition, and the specificity is the share of them the test gets right.
Of 80 people without a condition, a test clears 76 and flags 4: Sp = 76 / (76 + 4) = 0.95, or 95%. The other 5% are false alarms. Turned round, a test with a specificity of 0.90 that raised 4 false alarms cleared 0.90 · 4 / 0.10 = 36 people.
| Condition present | Condition absent | |
|---|---|---|
| Test positive | TP | FP |
| Test negative | FN | TN |
Specificity reads down the second column; sensitivity reads down the first. Together they describe the test itself, independent of how many people tested have the condition, which is what separates them from the predictive values.
A test with a tunable threshold can be made more sensitive by flagging more, and every extra flag is a true case or a false alarm. Raising sensitivity lowers specificity and the other way round; the right balance depends on what a miss costs against what a false alarm costs. 1 − Sp is the false positive rate, and the positive likelihood ratio, Se / (1 − Sp), puts the two in one number.
What is a good specificity?
High enough that false alarms are rare for the purpose: a confirmatory test is set for specificity,
after a sensitive screen has done the finding.
Specificity or true negative rate?
The same number. 1 − Sp is the false positive rate.
Where do I get the interval?
The sensitivity and specificity calculator gives a 95% Wilson interval from the counts.
| Solve for | Rearranged | With the numbers | Comes to |
|---|---|---|---|
| Sp | TNTN | 7676 | 0.95 |
| TN | SpFP1 | 0.95·41 | 76 |
| FP | TN(1 | 76(1 | 4 |
Last reviewed October 20, 2026 ·How pages are reviewed
Statistics and probabilityAll formulas
Copy the code into your page where the formula card should appear. It loads from lovetocalc.com, so it stays current on its own — all we ask is that the credit line comes with it.