Both repeatability and reproducibility are important in measurement and research. If you are the operator of a particular precision measuring instrument, you have to know these two terms. Your instrument has to be repeatable. It means you take several readings and you get the same readings. If there is a variation, then you have to find the mean value.
And if you are a researcher, then your research must be reproducible. Other people can try it and find the same result.
So, how are these two terms actually different? Why are they important to know in measurement?
According to the Vocabulary of Metrology
repeatability is measurement precision under a set of repeatability conditions of measurement.
And according to the same source,
repeatability conditions of measurement is condition of measurement in a set of conditions that includes the same measurement procedure, same operators, same measuring system, same operating conditions and same location, and replicate measurements on the same or similar objects over a short period of time.
There are 6 factors that decide repeatability:
- same subject (instrument operator)
- same procedure
- same instrument
- same condition
- same location
- same object (the same object to measure)
And according to the same reference for the repeatability source,
reproducibility is the measurement precision under reproducibility conditions of measurement.
What is the reproducibility conditions of measurement?
reproducibility conditions of measurement is condition of measurement in a set of conditions that includes different locations, operators, measuring systems, and replicate measurements on the same or similar objects.
There are 5 factors that decide reproducibility:
- different subject (instrument operator)
- replicate measurement
- different instrument
- different location
- same object (the same object to measure)
The only similarity is the object to measure. While the rests are opposite.
So, What Are the Differences? And How They Matter?
What is Repeatability?
Repeatability is the degree of consistency among multiple readings done by an operator within the same procedure, same instrument, same condition, same location, and short time interval that measures the same object.
For example, when you are taking a digital caliper to measure a single 1-inch gauge block. If the readings are consistently displaying 1″, it means the digital caliper provides high repeatability. The same thing happens to the test indicators and other measuring tools that are expected to give a high degree of repeatability.
The chance of decreasing repeatability can happen due to errors occurring. Unfortunately, you don’t know where it comes from. There are chances of errors when it comes to using a caliper. Also, cosine error can occur to dial test indicators. These errors may affect the reading of your tools.
When the readings are taken, the broken electronic system of your old digital caliper gives inconsistent readings when measuring that 1-inch gauge block. As a result, when it measures different objects (with different lengths), you do not realize it gives you unreliable readings.
Worst, you have to note a number of readings manually to find the mean value. You make some mistakes and put the wrong number several times. It increases the error, then lowers the repeatability score. In this case, using the digital caliper that can transfer the readings on a device is preferable.
Repeatability is an important factor that determines whether your instrument is reliable or not.
What is Reproducibility?
It can be concluded that reproducibility refers to how a particular measurement is proved to generate the same result in other places, performed by different people, and using instruments from other brands.
When you are performing a measurement, make sure other people can do it as well. And if the result they gain is the same, then your measurement has high reproducibility.
In some cases, some researchers made a publication. They are responsible for what they are doing, and therefore, the research must be reproducible.
When you are claiming that the distance between the moon and earth is X km. People around the world have to get the same number, then your measurement is reproducible.