# Specificity

## Specificity is about diagnostic tests identifying healthy people

Specificity gives you a measure of the ability of a diagnostic test to

**identify healthy people**or**people who do not have a disease state**. In order to set up an experiment related to the specificity of a diagnostic test, you must have a current "**gold standard**" to use as a reference for detecting disease states.In regards to the 2x2 table, specificity is focused on this specific part of the table:

Specificity is the number of people without a disease (as rated by the "gold standard") that have negative test results for the diagnostic test. In other words,

Another interesting aspect of specificity is that in some cases, it can

**specificity is the proportion of people that were healthy and tested negative for the diagnostic test.**Another interesting aspect of specificity is that in some cases, it can

**"rule in"**a given disease state. If a test has high specificity (__>__95%), then it can detect healthy people. This means, at the same time, that a**positive****test result**is a strong indication that a person does**HAVE**the disease. With high specificity, the ability to**"rule in"**disease states will increase.Here is a 2x2 table that depicts the calculation of specificity:

**Specificity**

There is a quick and easy calculation that can tell you if a diagnostic test is useful in any capacity whatsoever, the Youden index. The formula is ((Sensitivity + Specificity) - 100). Any

**Youden index value below 50%**means that test is essentially worthless and should be discarded.Click on the

**Receiver Operator Characteristic**button to continue.## Hire A Statistician

**DO YOU NEED TO HIRE A STATISTICIAN?**

Eric Heidel, Ph.D., PStat** **will provide the following statistical consulting services for undergraduate and graduate students at $100/hour. Secure checkout is available with Stripe, Venmo, Zelle, or PayPal.

- Statistical Analysis
- Research Design
- Sample Size Calculations
- Diagnostic Testing and Epidemiological Calculations
- Survey Design and Psychometrics