Difference between Sensitivity and Specificity

Sensitivity and specificity are two of the most widely used statistical measures which are used to assess the performance of the binary classification test that it also known as classification function. Although, these two terms are used simultaneously, but there is a considerable difference between the two terms. Sensitivity is used to measure the extent of actual positives which are identified e.g. people who are having sickness in a group of people. On the other hand, Specificity is used to measure the extent of actual negatives e.g. the people who are not having any kind of sickness in the group.

Instructions

  • 1

    Sensitivity

    It is a statistical measure which is used to check the proportion of positive outcomes. It can be calculated by dividing the number of true positives with the total number of outcomes (number of true positive + number of false negatives). For instance, a survey has been conducted in a village to identity the people who are suffering from fever. Suppose there are 100 people in the village and 50 people are being identified as sick. The sensitivity of the survey will be 50 [50 / (50+50) = 50]. A test or survey is considered to be highly reliable if the results show high sensitivity. In contrast, if the test shows negative results as compared to high sensitivity test, it means that it has ruled out the disease. It is extremely important for you to keep in mind that sensitivity cannot be called as precision about something or positive values which can be predicted. It is the ratio of actual positive values to the total number of observations.

    Image courtesy: aidsrestherapy.com

  • 2

    Specificity

    It is related to a test which identifies the actual negative results. It can be calculated by dividing the number of true negatives by the total number of observations (number of true negatives + number of false positive). For instance, if you take the example mentioned above, suppose there are 100 people living in a village and a survey has been conducted to find out how many people are suffering with fever. There are 50 people who have fever. Then, the specificity will be 50 [50 / (50+50) = 50].

    Image courtesy: radiology.rsna.org

Leave a Reply

Your email address will not be published. Required fields are marked *


− 4 = four