Based on (too) few examples, a rule is established that purports to be universally valid.
My ex-boyfriend cheated on me.
All men are cheaters!
As devastating as bad experiences in relationships can be – surely not all men are cheaters. The same is true for women, for that matter.
- Faulty induction
This is an error that occurs when a generalization is based on only a (too) small number of observations of a phenomenon.
This is particularly objectionable if the generalization refers to a diverse group and one could therefore find example cases for virtually any attributes (Nutpicking).
In the worst – and probably also most frequent – case, this is applied to groups of people, who can thus be associated with prejudices. Certainly, in any group of people one can find people who seem to confirm common stereotypes: undoubtedly there are radical Muslims, greedy Jews, hard-drinking Irish or Germans who are Nazis, as well as sexist men and irrational women. In none of these cases, however, are the examples sufficient to ascribe these attributes to the respective group as a whole (Anecdotal evidence).
It is a normal process in the language acquisition of children that they like to use crude generalizations in a certain phase (usually at the age of about 2-3 years). For example, every four-legged animal is called a “dog”, or every means of transportation is called a “car”.
Politician A is corrupt.
Therefore all politicians are corrupt.
There is no doubt that there are politicians who have proven to be corrupt – and quite a few who behave at least dubiously. But to conclude from this that all politicians are corrupt is clearly unfair.
There are certainly also people who enter politics out of idealistic or even altruistic motives. How large their share is, however, can be discussed controversially.