One conclusion. Three conditions.

You know the conclusion.

I’m good under pressure.
She’s difficult.
He’s brilliant with people.
I’m terrible at speaking up.

Maybe you are right.

The problem is that we can build surprisingly large conclusions from surprisingly small samples.

So take one judgement you have made about yourself or someone else. Positive, negative or somewhere in between.

Now widen the sample.

Think of three different conditions in which you have seen that person. Different places. Different people around them. Different levels of pressure, support, familiarity, tiredness or expectation.

One extra check: were you there each time? If so, you were part of the conditions too. If you can, make one of your three a situation where you were not in the room.

Don’t try to explain anything yet. Just look at what happened.

Did the same quality appear each time?

Perhaps the person who seemed confident with friends was much quieter at work. Perhaps someone you thought of as patient was patient at home, patient at work and still patient when it cost them something. Or perhaps your own discipline was strong when life had structure and much harder to access when that structure disappeared.

Sometimes the wider sample changes the conclusion. Sometimes it strengthens it.

And sometimes it reveals something more useful: 

the behaviour was real, but it was more dependent on the conditions than you knew.

This is not an instruction to doubt everything you see. Sometimes the first reading is right. And understanding conditions does not mean excusing harm or lowering a necessary standard.

It is simply a better way to earn a verdict.

One conclusion. Three conditions.

See whether the evidence still carries it.

Companion to this week’s Tuesday article: The Behaviour Was Real. Was The Verdict?