The Weekly Newsmagazine of Science
Volume 155, Number 20 (May 15, 1999)
Was it the weight?
I found "Why cutting fats
may harm the heart" (SN: 3/20/99, p. 181) very interesting,
if perplexing. I wonder about what effect the shortness of the
experimental trialsas little as 10 daysmay have had.
Also, no mention is made of any change in weight, even for those
on the diet providing only 10 percent of calories from fat. Perhaps
the observed changes in blood lipids were due to the (unobserved)
fact that the subjects were losing weight on these diets and that
their levels would have returned to something more normal (if
not necessarily more healthy) once their weights stabilized?
Researcher Ronald Krauss says his team tried to keep the participants'
weight steady throughout the trials. And while the researchers would have liked to
run the 10 percentfat trial longer, they found they couldn't. "It was simply too
extreme [and unpalatable] a diet for these people to maintain any longer," says
Krauss. J. Raloff
I believe that an important reason that so many people fail to
respond to the census surveys ("Census
sampling confusion," SN: 3/6/99, p. 152) is that the survey
has been expanded beyond a simple "count." In the government's
increasing quest for information on its citizens, there is a major
resistance on our part to disclose much of this information. There
is resistance to invasion of privacy. Care should be taken to
restrict the information gathering to the core of what is required
The article makes one thing perfectly clear: While most agencies
are struggling to do more with less, the Census Bureau has a long
history of doing less with more. In 1790, each of the 650 census
counters counted about 6,000 citizens. In 1890, each of the 47,000
counters only counted about 1,300 citizens. In 1990, with all
the high technology tools available to them, each member of the
army of 510,000 counters counted less than 500 citizens.
This is of course good news, because it shows that the Census
Bureau is on course toward resolving the problem about using statistical
sampling. At this rate, in the year 2340, there will be one census
counter for every citizen.
James I. Mangi
Falls Church, Va.
The difference between direct enumeration yielding 248.7 million
and the "other surveys and analyses" indicating 253 million is
only 1.7 percent. Suggesting that those missing are "mostly" children,
racial and ethnic minorities, and the poor sounds to me like the
well-used strategy for getting media and political attention by
Has it ever occurred to anyone that the federal government documents
the existence of each and every citizen of this country every
April 15? Those who don't file a tax return are listed as someone
else's dependent. Add a few lines to the decade tax return, and
the census is accomplished with one less bureaucracy. .
Thomas P. Becker
|How to communicate with Science News:
||Use our convenient online form: Feedback to Science News
||E-mail us at firstname.lastname@example.org|
||Send snail mail to:
||Editor, Science News
1719 N Street, N.W.
Washington, DC 20036
|All letters subject to editing.
Back to Top
Copyright © 1999 Science Service