All measurements are wrong! Your weight shown on a bathroom scale, your blood pressure taken at the doctor’s office, the diameter of a shaft we measure at work are all wrong. Your true weight may be 151.6325… pounds, but your bathroom scale shows you weigh 151.6 pounds. Your true blood pressure may be 137.433… over 72.185…, but the nurse gets a reading of 135 over 70. The true diameter of the shaft is 2.503111… inches, but your micrometer reading says it’s 2.503 inches. So what’s going on?

When we accept a measurement we have no idea how many digits are trustworthy and how many are meaningless. In order to understand the results we get we need to study the measuring instrument itself. This is known as Measurement System Analysis or MSA and is a collection of the instruments, gages, standards, methods, fixtures, software, personnel, environment, and assumptions used to quantify a measurement. It is the complete process used to obtain the measurement.

From the above paragraph you can see that the measurement of anything involves many variables. The method a person uses to take a measurement, the instrument’s program and software, the person’s training, experience, attitude and assumptions, the fixtures they use and the environment in which the measurement is taken can all affect the value recorded.

In order for us to analyze a measurement system, it’s important to understand several key concepts and terminology. These concepts include bias or accuracy, stability of the measurement system, and precision.

- Bias is the systematic difference between the average of several measurement results and a reference or true value. Bias is also known as accuracy. Let me give you an example. Let’s assume you have a standard whose true value is certified to be 2.5000 inches The average of ten measurements of the standard using our measuring device is 2.4995 inches. The bias is then 2.5000 – 2.4995 = 0.0005 inches. The calibration of our measuring device on some periodic frequency is used to control bias.
- Stability of the measurement system represents the change in bias over time and usage when the system is used to measure a master or standard part. A stable measurement system is one in which the variation seen in the measurements is in statistical control and helps us establish our calibration frequency.
- Precision is the closeness of agreement between randomly selected measurement results. It is this aspect of measurement that addresses the repeatability or consistency of our measurement system when an identical item is measured several times.

Since a picture is worth a thousand words, let me illustrate what I mean.

This diagram shows the true value as the dashed line to the left. The distribution of the measurements is shown as the bell-shaped, normal curve with the mean or average shown as the dashed line at the center of the distribution. The bias or accuracy is the difference between the true value and the average of the measured values. The precision of the readings is shown as the width of the measurement distribution.

A more traditional way of showing this relationship is shown below:

In the upper left, we see the five readings all around the 7:00 area. This grouping has low variation, is precise, but not accurate (an indication of bias). The lower left shows the grouping as neither precise or accurate. The upper right shows the grouping as accurate and precise and illustrates the type of measurements that we hope to see when we measure something.

It goes without saying that if you can’t trust your measurement system, then you can’t trust the data that it produces. This is why it is important to conduct various measurement studies and understand your measurement system and its variation.

Accuracy and precision can be easily evaluated through many measurement systems analysis tools including Gage Linearity and Bias Studies and Gage R&R Studies, which can help you reveal if a scale needs to be recalibrated or if your newly hired operators are measuring products consistently.

What should you do if you detect accuracy and/or precision errors? Focus on improving your measurement system before relying on your data. Allow the results of your MSA to help you decide if recalibrating a or conducting more training for new hires might be just what you need to get your measurement systems back on track.

One Response to “Measurement Accuracy & Precision Determines The Quality of Our Data”

It is important to do MSA to validate your measurements system. And to obtain accurate measurements result. Agree All measurements need to validate.