2004-11-08

Lying With Statistics

My Dad used to teach a Statistics course. One of his lectures was called 'Lying With Statistics,' and the lecture covered misleading statistical representations to alert students to such deceptions.

This post is about a small point, but I think a telling one. In announcing Bush's victory last week, Andrew Card stated that:
In this election, President Bush received more votes than any presidential candidate in our nation's history.

This is true but misleading. The statement conceals the fact that John Kerry also received more votes than any other American President. It conceals the fact that Bush's margin of victory (3.5m) falls short of Clinton's (8.2m), Reagan's (16.9m), Nixon's (18.0m), Johnson's (16.0m), or Eisenhower's (9.6m) second-term victories in absolute votes, and represents an even narrower win in relative terms.

As far as I know, Bush II is the only presidential candidate to defeat his closest rival despite garnering fewer votes, as he did infamously in 2000 (Bush II: 50,456,002 to Gore: 50,999,897). I know nobody has since Dubya Dubya Two.

None of this alters the fact that Kerry lost and lost incontrovertibly. It does point out that political rhetoric must always be viewed skeptically. And so should the media. The fact that Kerry scored more votes this election than any previous President occurred to me immediately. Why has no journalist raised this refutation to Card's empty claim? Bush's victory has the illusion of a mandate, an illusion that is being strengthened by GOP disinformation and pleas from his challenger, during his concession speech, to rally behind the President. Astounding.

PS: My favorite electoral maps covering the election:
CBC Newsworld - the pop-ups on mouse-over are brilliant
BBC News - fascinating and detailed historical results
CNN - telling exit poll data

No comments: