← Back
Intellectual

Statistics for everyday life

12 sections · 10 pts/section
Section 1 / 12

The numbers people use to fool you

Numbers can fool smart people every day.

You do not need math to spot the trick.

You just need to ask a few questions.

Here is a real case from 1995.

UK safety officials warned about a birth-control pill.

They said it doubled the risk of a dangerous blood clot.

"Doubled the risk" sounds scary.

So look at the real count of women.

Before / After

Same risk, two ways to say it. Tap to flip.

The clot risk doubled. It went up by 100%.

Both lines describe the very same numbers. One sounds huge. One sounds tiny.

Going from 1 to 2 in 7,000 is one extra woman.

That jump can be told two ways.

relative risk
How much a risk grows next to its old size. Going from 1 to 2 is a 100% rise.
absolute risk
The real chance for one person. Here it went from 1 in 7,000 to 2 in 7,000.
A risk that doubles must be a big risk.Tap to reveal

"Doubled" hides the starting size. Double a tiny risk is still tiny: 1 in 7,000 became 2 in 7,000.

The relative risk increase was 100%, but the absolute risk increase was only 1 in 7,000.

Gerd Gigerenzer et al., "Helping Doctors and Patients Make Sense of Health Statistics"

But "doubled" scared many women off the pill.

Many stopped taking it.

More got pregnant who did not plan to.

Now look at a second case, from World War Two.

Planes came back from battle full of bullet holes.

Most holes were on the wings and the body.

Few holes showed up on the engines.

Your guess

Armor is heavy, so you can only add a little. Where should it go?

The bullet holes told the opposite of the truth.

Same trap as the pill: the numbers fooled smart people.

This section is the numbers half of Cognitive Biases.

Next you will learn the few questions that catch these tricks.

Sources

  • Gigerenzer & Hoffrage, natural-frequencies research
  • Gigerenzer, citing David Eddy
  • Daniel Kahneman, Thinking, Fast and Slow
  • David Spiegelhalter, The Art of Statistics
  • Daniel Kahneman, Thinking, Fast and Slow (2011)
  • Alex Reinhart, Statistics Done Wrong
  • David Spiegelhalter, The Art of Statistics (2019)