The Certainty Effect: Why We Fail at Calculating Probabilities
The Certainty Effect: Why We Fail at Calculating Probabilities

The Certainty Effect: Why We Fail at Calculating Probabilities

“Not to be absolutely certain is, I think, one of the essential things in rationality” – B. Russell

What is the certainty effect? According to Li & Chapman (2009), the certainty effect happens when people overweight outcomes that are considered certain relative to outcomes that are merely possible. The effect was introduced by Kahneman and Tversky in 1979.

The certainty effect makes people prefer 100% as a reference point relative to other percentages, even though 100% may be an illusion of certainty.

This means, people choose 100% because they think that 100% is the highest and most beneficial probability available, but sometimes 100% may not be as beneficial, as we assume. Lower percentages or probabilities can be more beneficial in the long term.

For example, people prefer a 100% discount on a cup of coffee every 10 days to other more frequent, but lower amount, discounts (Li & Chapman, 2009), even though the second option may save them more money in the long run.

The point is that people are less sensitive to probabilities that are away from reference points. There are two natural reference points, namely 0 (certainly will not happen) and 1 (certainly will happen). 1 can be considered as a 100% probability, and 0 as a 0% probability.

Whenever we move away from the two natural reference points, we get confused about probabilities. It is just a lot easier to use 0% and 100% to make estimates about probability.

Arkes argues that we save cognitive effort by focusing on reference points, and he points to the fact that we use less time in processing certainty than we do in processing other probabilities (1991, as cited in Li & Chapman, 2009).

Research by Dickhaut and colleagues supports this idea (2003). In short, we process 100% with less cognitive processing or effort and more intuition. This tendency seems to have negative side effects as it often leads to wrong estimates. Here is an example of the certainty effect, where people use 10% as a reference point:

People were more attracted to a vaccine that was described as eliminating a 10% risk … than if it was described as reducing the risk for one disease from 20% to 10%.

In this example, people do not make the right estimates because the two vaccines are equally beneficial, i.e. the ‘net risk reduction’ is equal across the two versions.

In a study by Li & Chapman (2009), the authors found that the certainty effect can be applied to real-world decisions.

The study shows that insofar as 100% cannot be exceeded (e.g., 100% probability), it serves as a prominent reference point. This overweighting, however, does not occur when 100% can be exceeded (e.g., a 120% saving in points).

The authors sums up by saying that the appeal of 100% has important implications for both health promotion, consumer decisions and public policy.

They emphasize that after all anything can be described as 100% of something, and it is a cognitive bias that may influence decisions in an unintentional way.

Share Your Thoughts

%d