Scientists Say: Uncertainty

Uncertainty (noun, “Un-SIR-ten-tee”)

In daily life, a person may be certain about some things but uncertain about others. For example, they may be certain they’ll eat breakfast one morning but uncertain if it will rain. In science, though, everything is uncertain. And scientists often measure that uncertainty.

Uncertainty is how much a measurement varies around an already-measured value. No measurement can be completely accurate. There will always be some error. Or there can be natural variation in whatever is being measured. So scientists will try to measure how much uncertainty can be found in their data. To represent that uncertainty, they place error bars around a point or line on a graph or chart. The bars’ size represents how much new measurements might be expected to vary around the value the scientists have found.

Sometimes scientists express uncertainty with the standard error of the mean. These bars represent where all potential measurements might fall, based on a random sample. Another way to express uncertainty is with a confidence interval. This is a predicted range of values that are likely to contain the true value a scientist is trying to find. Confidence intervals are usually expressed as percentages. With a 95-percent confidence interval, any new measurement should fall within that interval 95 times out of 100. 

Uncertainty can also be used to indicate how likely something is to occur. For example, climate change scientists may include uncertainty in their discussions. This doesn’t mean that they are uncertain whether the planet’s climate is changing. They have documented that change in many ways. But there is always some small bit of uncertainty around how much change is happening and where. 

In a sentence

When scientists study how much the nutrition value of a food changes over time, their results include the uncertainty around their measurements.

Check out the full list of Scientists Say here.