Accuracy:
There are 2 definitions of Accuracy
- In math's, science accuracy refers to close a measurement value is to the actual value.
- The ISO (International Organization for Standardization) definition means accurate measurement has no systematic error and no random error. i.e. ISO advice that accurate be used when a measurement is both accurate and precise.
Precision:
- Precision is how consistent result when measurements are repeated. Precise values are differ from each other because of random error, which is a form of observational error.
- It is how close the measured values are to each other. It is related to Repeatability and Reproducibility in Measurement System Analysis (MSA).
Examples:
- Your office starts at 8:00am. Arriving to work after 8:15am is considered late. You take a subway train to commute to work. Usually you board the same train every (working) day that departs at 7:30am. Barring any unusual circumstances, you report at work between 8:22am and 8:24am every day. Your arrival time is always within the 2 min window. In other words, your arrival time has very little scatter, which means it’s very precise. Here, the true or target value is 8:00am. Your arrival time is 22-24 mins away from the target. Therefore, you are “not” accurate in terms of arrival time, as your company considers arriving after 8:15am as late.
What if you arrive at work between 7:32am and 7:34am every day. Does that make you any more accurate?
The answer is - No. Now you may argue that arriving at work early every day makes you accurate (and makes your boss happy) but in statistical terms, you are still “not” accurate because you are 26-28 mins away from the target value of 8:00am.
Post a Comment
Post a Comment
Please do not enter any spam link in the comment box.