Accuracy and precision

The difference between accuracy and precision.

Written by Andy Connelly. Published 15th May 2017.

Introduction

In science, words matter and their meanings matter even more. It might be too much to say that some scientific texts read like poetry; however, the ideas in those texts can certainly be beautiful. It is a beauty expressed in words that have specific means; words that must be used correctly. In everyday speech many of these scientific words often get mangled, their means blurred.

In English, these mangled scientific words oddly seem to come in pairs: stress & strain, weight & mass, accuracy & precision. Knowing the difference between the words in these pairs is important to understanding the field in which they are used. The final pair are what I will discuss here; the precise accurate usage of the terms accuracy and precision is vital in helping our understanding of data, particularly in our understanding of uncertainty in our data and in science (Figure 1):

  • Accuracy: how close to the “true” value you are (e.g. how close to your true weight your bathroom scales weigh you to). This can be applied to a single measurement, but is more commonly applied to the mean value of several repeated measurements, or replicates.
  • Precision: Precision is defined as the extent to which results agree with one another. In other words, it is a measure of consistency, and is usually evaluated in terms of the range or spread of results. Practically, this means that precision is inherently related to the standard deviation of the repeated measurements.

Using my pay packet as an example: the statement “I earn between £1 and £100,000 per year” is accurate but not precise. The statement, “I earn £154,302,731.82 per year” is very precise but-as a technician-not accurate, sadly.

precise_or_accuracy
Figure 1: The difference between accuracy and precision.

DISCLAIMER: I am not an expert in analytical chemistry. The content of this blog is what I have discovered through my efforts to understand the subject. I have done my best to make the information here in as accurate as possible. If you spot any errors or admissions, or have any comments, please let me know.

Types of error

Your accuracy is generally more affected by systematic errors in your method: for example, a poorly calibrated piece of equipment. How well you can define your precision depends on how precise you equipment is AND on random error. So, unless there is a real problem, the more measurements you take the more precise you can be. There is a third error, often called Gross error, which is simply the result of a blunder or a mistake. However, be careful not to assume a strange data point is one of these; it might be an important result!

  • Systematic errors – These are sources of inaccuracies are repeatable inaccuracies that are consistently in the same direction. Systematic errors are often due to a problem which persists throughout the entire experiment: for example, zero error or inaccurately calibrated instruments. You should aim to design your experiment to remove systematic errors OR calibrate out the error using certified reference material (CRM) or a standard addition approach – if you follow this method you must state what you have done. Clearly, the results for a CRM will still be subject to random error but your results should still be within the specified uncertainty. If they are not repeated measurements may be required to trace the source of this error.
  • Random errors – these are statistical fluctuations (in either direction) in the measured data due to the precision limitations of the measurement device. Random errors usually result from the experimenter’s inability to take the same measurement in exactly the same way every time to get exact the same number. Sources of this poor precision include noise from the instrument and careless reading or recording of data. A reference material will help follow these errors over time. Otherwise, good experimental technique is key to reducing random error. Another method is repeated measurements as random errors tend to cancel each other out.

Your aim, as a scientist, is to get the best possible accuracy and precision by minimising errors. In reality, the decision often becomes – what level of accuracy and precision is acceptable.

Summary

Improving accuracy is about getting closer to the ‘true’ value of the result. It is often very difficult to check for accuracy in research as you are working in new areas of science. However, using CRMs and comparing your work to others is a start.

Getting an idea of your precision is much easier. If your work is precise you can repeat the measurement and get a similar answer. Sounds so simple doesn’t it, somehow it never ends up being that simple.

 

Advertisements

1 Comment

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s