• Biotechnology: molecular diagnostics, reading DNA to answer questions

What do sensitivity and specificity mean?

Written with AI (How we create content)Target level: Grade 12Last updated: Report an error
Quick answer

Sensitivity is the share of people who have a condition whom a test correctly finds positive; specificity is the share of people without it whom the test correctly finds negative. A test with 90% sensitivity misses 1 in 10 true cases; one with 95% specificity wrongly flags 5 in 100 healthy people. What a positive result means also depends on how common the condition is.

The topic explained

Four kinds of result

What do sensitivity and specificity mean? Every diagnostic test, from ELISA to PCR to a rapid strip, is judged by how often it is right. Comparing a test with a reliable reference gives four kinds of result: true positives, people with the condition who test positive; false negatives, people with it who test negative; false positives, people without it who test positive; and true negatives. They are arranged in a two by two table. The reference, sometimes called a gold standard, must itself be reliable: if it misses cases, the new test's numbers are wrong too. Sensitivity and specificity describe the test at a given cut-off; measured in a careful trial, they do not depend on how many people in the trial had the condition.

In an invented trial of 1,000 people, 100 have the condition: 90 test positive and 10 negative. Of the 900 without it, 855 test negative and 45 positive. Sensitivity is the share of those with the condition whom the test finds: 90÷100=90%90 \div 100 = 90\%. Specificity is the share of those without it whom the test clears: 855÷900=95%855 \div 900 = 95\%.

A two by two table for an invented test of 1,000 people: of 100 with the condition, 90 test positive and 10 negative; of 900 without it, 45 test positive and 855 negative.has the conditiondoes nottest positivetrue +: 90false +: 45test negativefalse −: 10true −: 855
Invented counts. Sensitivity = 90 ÷ 100 = 90%; specificity = 855 ÷ 900 = 95%.

What a positive result means

A person holding a positive result asks a different question: given this result, how likely is it that I have the condition? That is the positive predictive value. In the same trial, 135 people tested positive, 90 true and 45 false, so 90÷135≈66.7%90 \div 135 \approx 66.7\% of positives are true. Of the 865 negatives, 855 are true, a negative predictive value of about 98.8%.

The most common mistake is to read sensitivity as this answer, to think that with a 90% sensitive test, a positive result means a 90% chance of the condition. Sensitivity starts from people known to have the condition; the predictive value starts from the result. A useful rule of thumb: a highly sensitive test rarely misses, so a negative result from it is reassuring; a highly specific test rarely raises a false alarm, so a positive result from it is convincing.

Why rarity changes everything

Now use the same test on 10,000 people of whom only 1%, 100, have the condition. It still finds 90 of them. But 5% of the 9,900 without it, 495 people, also test positive. Of 585 positives only 90 are true: 90÷585≈15.4%90 \div 585 \approx 15.4\%. Nothing about the test changed; the condition became rarer, and most positives became false.

This is why a screening test given to many people is followed by a second, more specific test for anyone positive, and why a single result is interpreted by a doctor who knows how likely the condition was to begin with.

A bar chart of the share of positive results that are true for the same test: 66.7% when 10% of people have the condition and 15.4% when 1% have it.% of positives true20406080010% have it1% have it66.7%15.4%
Same test, same sensitivity and specificity; only how common the condition is changes.

Moving the cut-off

Many tests give a number, an absorbance in ELISA or a Ct in qPCR, and a cut-off decides what counts as positive. The values of people with and without the condition overlap. Moving the cut-off to catch more true cases raises sensitivity but lets in more false positives, lowering specificity; moving it the other way does the reverse. Test designers choose the balance by what a missed case or a false alarm would cost. In the invented trial, a lower cut-off might find 95 of the 100 true cases while flagging 90 healthy people instead of 45: sensitivity rises to 95% and specificity falls to 810÷900=90%810 \div 900 = 90\%.

Two overlapping bell-shaped curves of a test value, one for people without the condition centered lower and one for people with it centered higher, and a vertical cut-off line between them.test valuepeople0withoutwith
Schematic. The vertical line is the cut-off: values right of it count as positive.
Free printable worksheet

Practice: sensitivity and specificity

For 12th grade biotechnology students. All numbers are invented. Show every calculation, then check your answers.

Free, no sign-up. The answer key prints on the second page. Regenerating a new worksheet on the same topic needs a free account.

Free printable worksheetlessoons.com
Practice: sensitivity and specificityName:
  1. Fill in the hidden cells of the table, using the totals: 100 people have the condition and 900 do not.

    The two by two table of the invented test with the numbers of false negatives and false positives hidden.has the conditiondoes nottest positivetrue +: 90?test negative?true −: 855
  2. Define sensitivity and specificity in one sentence each.

  3. A new test is tried on 200 people with a condition and 800 without. It is positive for 180 of the first group and negative for 760 of the second. Calculate its sensitivity and specificity.

  4. For that test, what share of positive results are true?

  5. Fill in the second bar, and explain why it is so much lower than the first.

    The bar chart of the share of positives that are true: 66.7% when 10% have the condition, and a question mark for when 1% have it.% of positives true20406080010% have it1% have it66.7%?
  6. A student says: the test is 90% sensitive, so my positive result means a 90% chance I have the condition. Explain the mistake.

  7. If the cut-off line is moved to the left, what happens to sensitivity and to specificity? Explain from the curves.

    The two overlapping curves of a test value for people without and with the condition, with a vertical cut-off line between them.test valuepeople0withoutwith
More topics and worksheets at lessoons.com© 2026 Lessons

Common questions

Which matters more, sensitivity or specificity?

It depends on the use. A test meant not to miss cases, such as a first screening, needs high sensitivity; a test that confirms a result before an important decision needs high specificity.

What is a false positive?

A positive result in a person who does not have the condition. A false negative is a negative result in a person who does.

Can a test be 100% sensitive and 100% specific?

In practice almost never, because the values of people with and without a condition usually overlap and any cut-off misclassifies some of them.

Why are positive screening results checked with a second test?

When a condition is rare, many positives from a screening test are false. A second, more specific test confirms which positives are true.

Do sensitivity and specificity change with prevalence?

Not in the simple model: they describe the test itself. The predictive values, what a result means for a person, change with how common the condition is.

Still have a question? Ask Lessi

Any question about this subject. Lessi answers right here, step by step.

Try asking

Asking needs a free account; your question is kept while you sign up. Answers are written by AI and can contain mistakes. Lessi doesn’t give medical or legal advice.

Learn it with a teacher

Learn this topic live with Lessi

Lessi, the AI teacher, explains it out loud on a board, with short quizzes, at your pace. The first 60 minutes are free.