The biased sample fallacy (also called selection bias or unrepresentative sample) occurs when a subset of a group is used to make a generalization about the whole group, even though the subset has differentiating characteristics from the rest of the group.

Here is an example:

Elsa: Nobody in this town likes to watch that new TV show, Bonkers Conquers.

Alex: Why do you say that?

Elsa: Because I asked twelve people at the whisky bar if they watch it, and they all said No.

The problem here is that the sort of people who gather at the whisky bar may not be of the type who would watch that particular show. The cure for this would be to gather a random sample of people from across the whole town.