Two key aspects of the reliability of measurement outcomes are accuracy and precision. Consider a series of repeated weight measurements performed on a particular object with an equal-arms balance. From a realist, “error-based” perspective, the outcomes of these measurements are accurate if they are close to the true value of the quantity being measured—in our case, the true ratio of the object’s weight to the chosen unit—and precise if they are close to each other. [Stanford, 200807]
Scientists must grapple with the fact that all measurements are eventually sensed and perceived by a human. We use our eyes to read a ruler. Science is based on a comparison with the ruler, subject to our errors in reading it. (Measurement training helps us develop a perspective of how brief and limited human observations are, and how an accuracy limitation underlies each measurement.)
The scientific community thus automatically subjects any novel facts to tests for repeatability. Science measures repeatability statistically in terms of variability or accuracy. The process of reducing an observation to a measurement is not complete until an assessment of its accuracy is available.
In conclusion: Observations are too general and include subjective perceptions. The first step in objectivity is to compare the observations to standards. In the most general sense, this is the process of making measurements, or creating facts. Science looks for repeatable things called patterns in the measurements. The descriptions of the patterns are models, as we will consider in our next entry in this series. [CrossFit, 200401]