The Testable Question
Science begins not with an answer but with a question narrow enough to be answered — one that predicts different observations depending on what is true.
Every inquiry starts with an observation that doesn't fit: something surprising, a pattern, an anomaly. But an observation only becomes science when it is turned into a question the world can answer — specific, bounded, and tied to things you can actually measure.
A good scientific question forbids something. “Is this drug good?” cannot be answered; “Does this drug lower blood pressure more than a sugar pill over eight weeks?” can. The sharper the question, the clearer what would count as a yes and what would count as a no.
Vague questions produce vague findings that can never be wrong — and a claim that can never be wrong teaches you nothing. The rigor of the method is front-loaded: most of it is spent making the question answerable before a single measurement is taken.
Sharpen the question first
Start from the anomaly
Notice what actually surprised you — the result that clashes with expectation — and name it plainly before reaching for an explanation.
Make it answerable
Rewrite the question until it names what you would measure and under what conditions. If you can't say what observation would settle it, keep narrowing.
- What exactly am I measuring?
- Under what conditions, and compared to what?
- What answer would count as no?
Decide what would count as an answer
Before collecting anything, state what result would count as yes and what would count as no. Deciding after you look is how you fool yourself.
A question you can't be wrong about is not a scientific question — it's a wish.