Hasty Generalisation

What it means

A hasty generalisation is a conclusion drawn from a sample that is too small, too narrow, or too unrepresentative to support it. It’s the leap from a handful of examples to a sweeping claim about an entire category - a group of people, a type of product, an institution, a country.

Humans generalise constantly, and for good reason. Generalisation is how we navigate a complicated world without having to evaluate every single encounter from scratch. If three restaurants in a row give you food poisoning, it’s reasonable to be cautious about the fourth. The problem arises when the sample is too thin to justify the conclusion - when we treat a few data points as if they represent the whole picture.

What makes this fallacy so persistent is that it feels like evidence-based reasoning. “I’m not prejudiced - I’m just going by what I’ve seen.” But what you’ve seen is a tiny, non-random sample filtered through all your existing biases about what to notice and what to remember. The generalisation feels grounded in experience, but the experience itself is a biased sample.

In the real world

Stereotypes are hasty generalisations that have calcified into cultural assumptions. “French people are rude.” “Young people are lazy.” “Politicians are all corrupt.” Each of these started with someone’s limited experience and hardened into a general claim through repetition and selective memory. The few rude French waiters are remembered; the hundreds of perfectly pleasant ones are not.

In the news, hasty generalisations often masquerade as trend pieces. A journalist finds three examples of something - young professionals leaving London, parents choosing home schooling, people quitting social media - and writes it up as a movement. Three examples isn’t a trend. It’s three examples. But the narrative is compelling, the examples are vivid, and the piece gets shared as if it describes something real and widespread.

In everyday conversation, we do this constantly. “I tried therapy once and it didn’t help - therapy doesn’t work.” “My friend went vegan and got ill - veganism is unhealthy.” “I hired a graduate once and they were useless - graduates these days have no skills.” Each conclusion is built on a sample of one, extrapolated to millions.