Accuracy vs. Precision . . . Know the Difference?

Bullseye Accuracy vs Precision

I never really thought about this until I became the product manager for pipettes. I had been using the terms synonymously in conversation to describe measurement errors. However in statistical calculations, regulations and pipetting, these two terms have different and very distinct meanings.

Accuracy refers to the deviation of a measurement from a standard or true value of the quantity being measured. We can talk about the accuracy of a single measurement. For example, if a pipette is set to dispense 100 microliters but actually delivers 99 microliters, the accuracy of that particular dispense is off (or the pipette is inaccurate) by -1 microliter.  We can also talk about the accuracy of a group of repeated (replicate) measurements. In these cases, we must first determine the mean of the group, and then compare that average value with the standard or true value. Accuracy for a group of measurements refers to the deviation of the group’s mean value from the standard or true value. For example, if we know that three replicate measurements averaged exactly 100 microliters, we still can’t predict how likely it is that the next dispense will be within some limits. One pipette might deliver 99, 100 and 101 microliters (pretty good), while a second delivers 80, 100 and 120 (pretty bad). The averages of both sets of data are exactly 100 and both are perfectly accurate, but which one would you use if your life depended on the next volume measurement being very close to 100? Highly scattered results can produce an accurate average. Accuracy is expressed in percentage error, E%, or inaccuracy.

Precision tells us how close a group of measurements are to one another. The closer the data replicates, the more likely the results will be similar in the future. For this reason, good precision has predictive value; it gives us confidence in future results. Precision is usually calculated and discussed in terms of standard deviations and coefficient of variation (CV). A precise or closely-clustered data set has a smaller CV and is generally more reliable than one that is widely scattered. One pipette might deliver 96, 97 and 98 microliters, while a second delivers 94, 99 and 98. The first pipette has better precision since the volumes are closer to each other; not to a standard value.

Because precision is concerned with the closeness of two or more measurements to each other rather than to a standard value, it’s possible for a group of values to be precise without being accurate, or to be accurate without being precise.

People frequently use archery or darts to help explain these two terms. Accuracy is the ability to shoot an arrow or dart near the bulls-eye. Precision is the ability to hit the target close to the same spot, no matter where that spot is, several times in a row as shown in the photo above. I hope this terminology refresher is useful.

Share this Post!

About the Author : Jackie Williams

Jackie has a dual role at WHEATON. She serves as both a Product Manager for Liquid Handling and a Content Strategist for all WHEATON products. She has worked in the pharmaceutical, food & beverage and chemical industries prior to joining WHEATON. Her educational background includes a B.S. in chemistry from Ursinus College and an MBA from Rutgers University. Jackie brings products to market with insight gained from hands-on experience.

0 Comment

Leave a Comment

Related post

Wheaton - Beyond the Glass