A Lesson in Statistics

If anybody ever quotes statistics, always ask for the source. If the source
happens to be a survey, always read the survey carefully. They’re usually
worded in a manner that skews the results in somebody’s favor. Here’s an

I just received a Republican Party Census Document. Apparently they think
I’m one of them. This is an understandable mistake, given that I haven’t
really publicly spoken out against Bush in, oh, fifteen minutes or so.
The survey is a series of 15 questions, each with Yes, No,
and Undecided. Here is a sampling of a biased survey:

  • Do you support President Bush’s initiatives to promote the safety
    and security of all Americans?
  • Do you support President Bush’s pro-growth policies to create more
    jobs and improve the economy?
  • Do you support President Bush’s plan to make our schools more
    accountable to parents and to restore local control of education?
  • Do you support the President’s plan to increase military spending
    to meet our defense needs?

Right, so the natural implication (to normal people, not mathematicians)
to any of these is that if you say No, you oppose the goal, and
not just the plans in question. The fourth one is even more fun, because
it nicely presupposes that our current defense needs aren’t already met.

The form ends by guilting you into giving them money. Here are the choices:

  • Yes, I support the RNC and am enclosing my most generous contribution
  • Yes, I support the RNC, but I am unable to participate at this time.
    However, I have enclosed $11 to cover the cost of tabulating my survey.
  • No, I favor electing liberal Democrats over the next ten years.


Creative Commons Attribution 3.0 United States
This work by Shaun McCance is licensed under a Creative Commons Attribution 3.0 United States.