The history of metrology is of great significance to repeat the results of experiments and improve the credibility of scientific research. Everyone is talking about repeatability, at least in the biomedical and social sciences. Over the past decade, there has been a growing awareness that a result must be repeated independently in order to be accepted as true.
The history of metrology is of great significance to repeat the results of experiments and improve the credibility of scientific research.
More than two centuries after Henry Cavendish measured gravitational constants with twisting scales, econometrics are still unable to agree on the exact value of the constant. Source: The Royal Society
Physical science also needs to focus on repeatability. A section of Nature-Physics focuses on this topic, and the econometricists Martin Milton and Antonio Possolo write that people should look at repeatability in a different light. When the results of measurement science cannot be repeated, they reflect the effects of different research methods and are a good opportunity to help the public understand the scientific process, the authors wrote. (M.) J. T. Milton and A. Possolo Nature Phys. 26, 117–119; 2020)。
The authors, from the International Bureau of Statistics and the National Institutes of Standards and Technology and the National Institute of Standards and Technology, cite three cases in which humans have tried to measure a fundamental constant in nature.
These include the speed of light (c); the Planck constant (h), which correlates photon energy with its frequency, and the gravitational constant (G), which measures the gravitational intensity between two objects.
In the case of Planck constants and speedofe light, different laboratories have obtained the same values in different ways — that is, repeatability. Among them, the value of the Planck constant, which was widely accepted, was set out last May to define the kilograms in the International System of Units.
But while countless experiments have been carried out over the past three centuries, the exact value of G has not been determined. Scientists are at a loss to see the root causes of this uncertainty: it may be because there is an unknown error in the way the value is measured, or it may signal the urgent need for new physics. One possibility that scientists are exploring is that G-values change over time, in which case scientists may first change their thinking that G is a constant.
Physicists may not find it possible, but if so, it will serve as an example of the fact that non-repeatable data is part of the scientific process: the results of the experiment sit until it challenges long-established theories, or points directly to another possible theory.
Problems in the biomedical and social sciences cannot be reduced to a basic constant in the measurement of nature. Repeatic experiments with results in areas such as cancer biology may include more variability sources and are extremely difficult to control than metrology.
However, metrology reminds us that when researchers try to repeat the results, they follow a common and very precise set of experimental criteria, known in the field of measurement, as the measurement of trace. It is this, the authors argue, that helps to establish the reliability and credibility of the scientific research process.
What we’ve learned from Milton and Possolo’s comments is that researchers in different fields should keep a conversation and share their experiences of repeatability. At the same time, when researchers follow the most accepted criteria and repeat a result without being allowed, let’s not say little about what’s going on.
Non-repeatability should not be automatically labeled as failure. It may be a reminder that it’s time to rethink our assumptions.