Sensitivity and Specificity Calculator

Your details

Number of people who have the condition AND tested positive (correct positive result).
Number of people who do NOT have the condition but tested positive (incorrect positive result, also called a type I error).
Number of people who HAVE the condition but tested negative (missed diagnosis, also called a type II error).
Number of people who do NOT have the condition AND tested negative (correct negative result).
Prevalence of the condition in the population you are testing. Changing this shows how PPV and NPV shift without altering sensitivity or specificity.
%
SensitivityExcellent test
90%

True positive rate: probability the test is positive given disease is present

Specificity90%
Accuracy90%
Positive Predictive Value (PPV)69.2%
Negative Predictive Value (NPV)97.3%
Positive Likelihood Ratio (LR+)9
Negative Likelihood Ratio (LR-)0.111
F1 Score0.857
Youden's J0.8
Total cases (N)300
Sensitivity90%
Specificity90%
PPV69.2%
NPV97.3%
Accuracy90%
0.8
Poor<0.3Fair0.3-0.5Good0.5-0.7Excellent0.7+

At 20.0% prevalence: 90.0% sensitivity, 90.0% specificity, PPV 69.2%, NPV 97.3%.

  • Sensitivity of 90.0% is good, though 10.0% of true positives are missed.
  • Specificity of 90.0% is good, though 10.0% of healthy people may receive a false alarm.
  • LR+ of 9.00 is moderate - a positive result meaningfully raises suspicion.
  • LR- of 0.111 is small - a negative result substantially lowers the probability of disease.

Next stepYouden's J of 0.800 summarises overall discriminative power (0 = no better than chance, 1 = perfect). This test discriminates well.

 = Powered by OnlyCalculators