Hasty generalization

Hasty Harry

“Two examples? Conclusion: obvious.”

What's the problem?

A hasty generalization draws a conclusion that the cases can't support: too few for the claim being made, or not representative of the group the claim is about. There's no magic minimum sample size; what counts as enough depends on the claim and the context. A large sample can still be weak if it's biased, for example if only the most enthusiastic people answered. Small samples aren't worthless: they can support observations, a reason to investigate, and some provisional inferences. But tentative wording alone doesn't make an inference warranted, and overconfidence isn't the only problem: a cautious-sounding claim can still reach past its cases or blame the wrong thing. Overgeneralizing Ollie is a close cousin: Harry asks whether the cases are enough and representative; Ollie asks how far the claim reaches beyond them.

Context: A new manager's first two hires from one staffing agency both left within a month.

✗ Too hasty

That agency only sends unreliable people.

✓ Fits the evidence

Both hires from that agency left within a month. That's worth looking into, including whether the roles were a good fit and what the job conditions were like, before drawing conclusions about the agency.

Why it works

Two early departures are a real fact and raise a fair question. They can't establish what the agency always does, and they don't show that the people were unreliable: the reasons could lie in job fit or working conditions. The revision states the supplied fact and opens those questions instead of assigning blame.