Day 94 — Hasty Generalisation

There’s a number of alternative names for the fallacy “hasty generalisation”, such as unrepresentative statistics, inductive generalisation or insufficient sampling; it involves poor conclusion making and over-generalisation.

It’s not a difficult fallacy to fall for either; it may seem that there’s enough support or significant endorsement of an argument, and yet if you only get a representative group that is pro- or anti- about a cause, it can lead to misunderstandings.

For example, if I only asked everyone in my classroom whether they should do the exam that’s happening next week, they’d probably agree that it was a good idea. However, if the sample also included the teachers, the school and the examination board, I’d get a different answer! It’s not good sample to cancel an exam over, even if I have more students than examination board members making the decision.

Here’s an example in politics, at the 1992 Republican convention in Houston, where Mrs Marilyn Quayle made a speech:

I sometimes think the liberals are so angry because they believe the grandiose promises of the liberation movement. They’re disappointed because most women don’t wish to be liberated by their essential natures as women. Most of us love being mothers or wives, which gives our lives a richness that few men or women get from professional accomplishments alone, nor has it made for a better society to liberate men from their obligations as husbands or fathers.’

While Mrs Quayle seems very confident about her ‘most women think this’, she’s not demonstrating where she got her sample from, if it’s a fair representation or it’s a conclusion that we can really depend upon.

Further Resources:

Logically Fallacious