Saturday, July 14, 2007

A meditation on policy-making

From Wikipedia:

This well-known saying is part of a phrase attributed to Benjamin Disraeli and popularized in the U.S. by Mark Twain:

There are three kinds of lies: lies, damned lies, and statistics.

The semi-ironic statement refers to the persuasive power of numbers, and succinctly describes how even accurate statistics can be used to bolster inaccurate arguments.

Its natural corollary, then:

The value of statistics, or numbers of any kind, in determining the best course for any decision or action, is directly related to the integrity and objectives of the entity presenting them.

This is not to say that statistics have no value. It means that every single statistic you consider must be evaluated in terms of the process by which it was obtained, the objectives of the statistician and the organization sponsoring their work, and the prevailing opinions of the day that govern what findings do or don't get published.

The best guidance I can offer to non-statisticians is: don't try this at home. Arguing numbers across a policy decision is the stuff of major, major mathematical skill and frankly, skepticism. The best way to consider a particular statistic is to ask yourself if it makes sense, how the facts used to enumerate the statistic were generated, whether there is substantial room for bias or purposeful ambiguity.

Lies, Damned lies and statistics. Its the truth. There's only one phrase I've embraced as heartily in my lifetime and it is from Tom Petty and the Heartbreakers: "The waiting is the hardest part".

Tom Petty and the Heartbreakers -- The Waiting

And those two things take me back to the song that ran through my head all day yesterday as I checked in and out of Lyman's blog:

John Mayer -- Waiting for the World to Change (Continuum)


Sandy said...

Cool observation that one. And i like the story u posted on indies blog. indeterminacy.


Lyman said...

Damn you a good writer!

Kathy said...

aw shucks, brilliant cousin. You inspire me.