Re: poly: An idealistic reason to lie to the census

From: Carl Feynman <>
Date: Thu Mar 26 1998 - 09:26:07 PST

At 07:35 AM 3/24/98 -0800, you wrote:
> (Carl Feynman) writes:
>>The Republicans in congress are currently working to make sure that the
>>Census Bureau is not allowed to use certain well-understood statistical
>>techniques in order to correct for a persistent undercount of poor people.
>>Since poor people tend to live in pro-Democrat districts, the present
>>undercount is advantageous to the Republicans.
> Those statistical techniques may be understood well by experts, but the
>average person doesn't understand them well enough to distinguish
>honest uses of them from attempts to inflate the count in districts
>that vote for higher government spending.

Well, I'm an expert. The techniques involved are easily understood, at
least to a level of detail needed to see that it's not some sort of scam.
But I haven't seen an explanation of them at that level. Perhaps this is
because this issue is too minor for the newspapers to devote an article on
the science page. I really only care about it because I'm a semi-pro

What really pisses me off is that it is precisely the uncertainty you
describe that makes the Republican position tenable, and this they have
been doing their best to increase uncertainty by propagating exaggerations,
half-truths, and the occasional outright lie. I don't like seeing a
powerful institution devoting its efforts to making people more ignorant.
Listening to Democrats talk about taxes and social security pisses me off
the same way, and for the same reason.

By the way, I remember a message that Perry sent around the list back in
January or so, bemoaning the lack of discussion on topics other than the
Final Conquest of the Cosmos. My sending out a message about the census
was an attempt to introduce a near-term topic of conversation. But it
seems to have lost out to the current favored topic, how to finance the
Final Conquest.

Received on Thu Mar 26 17:30:58 1998

This archive was generated by hypermail 2.1.8 : Tue Mar 07 2006 - 14:45:30 PST