Lies, damn lies and statistics.

One of my pet annoyances is when people use percentage figures as though they have some independent relevance - you know the sort of thing: "Figures released today show an increase in (something or other) of 75% over the past year". Other than the fact that there has been an increase, this statement is meaningless without knowing what the base-line figure was. It could simply mean that 4 instances have increased to 7, for example, or more significantly that 10,000 have increased to 17,500. It all depends on what we're talking about of course, but an increase of 3 would probably not be worth getting your knickers in a twist about, whereas an increase of 7,500 might well be highly important.