Learning & thinking · a life skill
Simple Bias Spotting
read your own evidence the other way round
"Try to be unbiased" is advice you can't act on. This is one you can. Take one thing you believe, write down what convinced you, and imagine each piece had pointed the other way. What you'd have said about it then is the doubt you skipped. Tonight you turn that doubt on the evidence as it came.
- Easy to start
- Medium impact
- Learn it in 1 week
- One belief, read both ways
The evidence on the left was taken at face value. The same evidence on the right got questions. Those questions were fair, so they apply on the left too.
Tonight, with a pen
Try it now.
One belief and a page folded down the middle. About twenty minutes.
-
Write one belief
Something you think is true about your work, your town or your habits, at the top of the page. Pick one you care about a little, not your deepest conviction.
-
List what convinced you
Down the left half, the two or three things that made you believe it: a number, a story, something you saw. One line each.
-
Flip each one
Down the right half, write the same piece as if it had come out the other way. Same source, same size, opposite result.
-
Write what you'd have said
Under each flipped piece, the first thing you'd have said about it. Under the original, what you actually said. Usually it was nothing.
If the flipped one makes you ask "who ran it?" or "how many people?", that's the question you skipped. -
Turn the doubt back
Ask each original piece the question you'd have asked its flip. Write one line on what still stands after that, and what you'd check next.
The belief can survive. The point isn't to change your mind. It's to find out how much of the case held up when you asked both sides the same questions.
Your flip tonight
Say how you took each piece, then how you'd take its flip.
Write the belief and up to three pieces of evidence. For each, say how hard you looked at it, then imagine it had pointed the other way and say how hard you'd have looked then. The check shows where you used two standards.
The evidence, both ways
The check
Your belief goes here.
Nothing you type is saved or sent anywhere. It's gone when you leave the page, so copy the check if you want to keep it.
Why flip rather than try to be fair
Asking people to consider the opposite corrected their bias more than asking them to be fair.
Charles Lord, Mark Lepper and Elizabeth Preston started from an idea: a lot of biased judgment comes from not thinking about the possibilities that would go against what you already believe. If that's the cause, getting people to think about those possibilities should help.
They tested it in two experiments. In one, people judged new evidence on a social question they already had views on. In the other, people tested an impression of what someone's personality was like. Some were told to be as fair and unbiased as they could. Others were given a strategy: consider the opposite.
In both, considering the opposite did more to correct the bias than the instruction to be fair, even though being fair sounds like the more demanding request.
The honest limits: two lab experiments from 1984, each in one sitting. They show the strategy working in the moment, not that it lasts or that it works on beliefs people care about most.
Be as fair and unbiased as you can.
Consider the opposite: how would you judge it if it had come out the other way?
- Experiment 1, judging new evidence. Weighing fresh findings on a social question people already had views on.
- Experiment 2, testing an impression. Checking a hunch about what someone's personality is like.
The idea
Why it works.
Bias rarely feels like bias. Evidence that agrees with you just looks right, so you don't think to question it. Evidence that disagrees looks suspicious, so you find the flaws. Each reaction seems reasonable on its own. The unfairness only shows when you set the two side by side.
"Be fair" asks you to notice something you can't feel. The flip gives you something to do instead: a second version of the same evidence, so the two standards are on one page where you can see them.
When it goes wrong
Habits that help.
- Keep the flip the same size. Flip the result, not the source. A survey stays a survey, and the same number of people answered it.
- Write the reaction before judging it. Put down the first thing you'd say to the flipped piece, however unfair. That gut line is the useful part.
- Doubt isn't the goal. If the flipped piece would have convinced you just as easily, say so. Then you're treating both the same, and that's what you're checking for.
- Start small. A mild belief gives you practice without a fight. Try the big ones once the flip feels routine.
Practise this in HourLife.
In Decisions: Bias check. Write down a decision, weigh the options, mark the one you chose, and come back later to record how it turned out.
