[FoRK] re: Libertarian Purity Test

John D. Mitchell johnm-fork at non.net
Wed Mar 10 10:14:21 PST 2004


>>>>> "place" == place org <jbone at place.org> writes:
[...]

> Whump links to some comments Gary Farber has on the Lib purity test.

> 	http://amygdalagf.blogspot.com/
> 2004_03_07_amygdalagf_archive.html#107890400453863577

Note that for folks who don't want to bother with the purity test itself,
you can check out some analysis of the test results done by the author:
http://www.gmu.edu/departments/economics/bcaplan/result.htm


> I thought I'd offer a few comments on his comments.  It's rather amusing
> how much people grouse about tests, ALWAYS.  Without fail.  Myself, I
> never seem to have any significant problem understanding the intent of
> the questions; each of these kinds of tests has a particular context, and
> each question in the test should be interpreted relative to that context.

FWIW, I often take each test multiple times with a consistently different
interpretation each time.


> So, Gary complains:

>> Not because it's libertarian. Because it's simple-minded.

> I would guess that Gary probably wouldn't like *any* yes-no test
> measuring opinion.

Indeed.

However, it's critical to have at least a "maybe"/"neutral" selection
available so that people can express their ambivalence.  For example, this
was very purposefully done in "The World's Smallest Political Quiz" (WPSQ)
http://www.self-gov.org/quiz.html

The WSPQ creators also pointedly created the "Centrist" grouping so that
there would be a distinct identity for centrists.


I've argued with some test/survey creation experts on the need for the
ability to say things like "don't care" and "this question is {stupid,
unclear, misleading, etc.}" but that makes the surveys way too complicated
for the vast majority of people. :-)


>> Somewhere between a third and a half of the questions I find
>> answerable. But the rest simply aren't, to my worldview, yes-or-no
>> questions.

>> Examples: Are taxes too high? Which taxes? For who?

> This is a question about principle.  In principle, an
> archetypical-extreme lib (as defined by the classical Lib thinkers, to
> the extent to which they agree) believes the tax rate should be zero for
> everyone.

Yeah, well, the Libertarian Purity Test (LPT) author was clearly trying to
tease out some gradation between the levels of belief through the
repetition of questions on the same point but with increased severity.

On the other hand, the WPSQ creators tested many different questions and a
different wordings for each of the questions specifically to get a properly
representative outcome for as wide a population as possible.  They were
very clear on the fact that the "best" questions to include in the quiz
would necessarily have to change over time so as to retain the
decisiveness.

Have fun,
	John



More information about the FoRK mailing list