I'm passing this message from The Federation on to BGA members for their
interest and action as they see fit personally. The message will
probably not be of interest to non-US members (sorry!).
Cheers
Mike
Dr. Michael C. Neale Dept. of Psychiatry, Virginia Commonwealth University
Phone: 804-828-3369 Virginia Institute for Psychiatric & Behavioral Genetics
Fax: 804-828-8801 800 East Leigh St. Suite 1-115, Richmond VA 23219-1534
------- Forwarded Message
Date: Mon, 03 Aug 1998 17:42:38 -0400
From: William Butz <wbutz@nsf.gov>
Subject: Re: Help Needed on Upcoming Science
To: Long list of people:;
May I offer a minor correction in defense of the Census Bureau, which is
having sufficient problems! The 1990 census was the second most accurate
since modern measurement of the undercount began with the 1950 census.
This means that it was almost certainly the second most accurate U.S.
census ever, the 1980 census being the most accurate. In any case, it was
very much better than being the "least accurate census in U.S. history."
It's correct that it was the most costly, in total dollars. Hard to say
whether it was most costly in terms of dollars per person enumerated.
Bill Butz
At 3:20 PM -0000 8/3/98, Murphy, JoAnne wrote:
> The House of Representatives began debate on the appropriation bill to
>fund the Departments of Commerce and State and the Judiciary on Friday
>and will pass the bill on Tuesday or Wednesday. There are many
>controversial items in the bill, and one in particular has relevance to
>behavioral science. After the 1990 census, which was the most costly and
>least accurate census in U.S. history, the Congress asked the National
>Academy of Sciences to suggest ways to make the census more accurate and
>less costly. Among the recommendations was to use scientific methods in
>taking the census, and a major one of those scientific methods is
>sampling. While sampling has been used for many years to obtain
>statistics that the government and the citizens of the country rely on
>every day--like the consumer price index, the unemployment rate, and the
>epidemiology of diseases like AIDS and tuberculosis--Republicans in the
>House are attempting to prohibit the Census Bureau from using sampling in
>the 2000 census on the ostensible grounds that the Constitution calls for
>an individual count of all the people in the country. They are
>attempting to block the use of sampling in two ways: By bringing a
>lawsuit against the use of sampling on the contention that it is
>unconstitutional, and by giving the Census Bureau only enough money to
>operate until January of 1999 (through the appropriation bill now being
>debated). They are, in effect, threatening a cutoff of funds to the
>Census Bureau if it persists in its intention to use sampling to achieve
>an accurate census count in 2000.
> The politics here are that an analysis of those who were
>undercounted in
>the 1990 census and those who were counted more than once, or
>overcounted, shows that the undercounted tend to be members of minority
>groups, homeless people, people who don't have English as their first
>language, and individuals with low income. At least 5 percent of the
>Hispanic population was not counted, for example. Those who tend to be
>counted more than once are more affluent--the people whose children are
>counted once at home and once in their college dorms, or who are counted
>first in one residence and then move to take a new job and are counted
>again. The undercounted tend to be more Democratic than Republican and
>the overcounted tend to be more Republican than Democratic. In addition
>to providing the background information on which the states will
>reapportion congressional districts, the census figures will determine
>how hundreds of billions of dollars in federal revenue will be
>distributed to states, counties, cities, and individuals. So the use of
>sampling in the census has been transformed from a scientific to a
>political consideration.
> If the bill passes as it stands, the House version of the
>appropriation
>will cut off funds for the census by January 1999. The Senate bill does
>not contain this provision, nor does it have language regarding the
>unacceptability of sampling. In the House floor proceedings on the
>appropriation bill, one amendment is being allowed to permit full funding
>of the census. It is known as the Mollohan Amendment since it will be
>offered by Representative Alan B. Mollohan (D-WV).
> THE PURPOSE OF THIS ACTION ALERT IS TO ASK THAT YOU CALL YOUR CONGRESS
>PERSON'S WASHINGTON OFFICE TODAY OR TOMORROW AND ASK THE REPRESENTATIVE
>TO VOTE FOR THE MOLLOHAN AMENDMENT. YOU CAN REACH YOUR REPRESENTATIVE'S
>OFFICE BY CALLING THE CAPITOL SWITCHBOARD: (202) 224-3121 AND ASKING TO
>BE CONNECTED WITH YOUR REPRESENTATIVE'S OFFICE.
> It is unlikely that you will be able to talk directly with your
>representative. Instead you should ask for the legislative assistant who
>deals with the Commerce, State, Judiciary appropriation. Then you tell
>this person that you would like your representative to know that you
>believe sampling is a necessary and reliable tool for improving the
>accuracy of the 2000 census. In view of that belief, you urge the
>representative to vote in favor of the Mollohan amendment. That's all
>you need to do. (If neither the legislative aid nor the Congress person
>is available, you can ask the person who answers the phone to leave a
>message expressing your view for the member of Congress and his or her
>appropriate legislative aid.)
> The debate over this issue began on Friday when the representatives
>were
>deciding whether or not to approve the rules under which the
>appropriation bill would be considered on the House floor. So far, it
>has been suggested that the National Academy of Sciences panel that made
>the recommendations commissioned by the Congress nearly a decade ago was
>fixed to achieve a preset set of recommendations. It has been suggested
>that sampling will eliminate real people from the census (when it fact,
>it would eliminate only duplicate counts). It has been suggested that
>sampling will lead to people deliberately not being counted when mail,
>telephone, and door-to-door methods will be used before sampling to count
>as many people without sampling as possible. The aspersions will
>continue in full force this week. In addition to making that important
>phone call to your representative, you might consider taking a look at
>C-SPAN on Tuesday when the two hours of debate on the Mollohan amendment
>is most likely to take place. (The timing is a little uncertain because
>the House's bill on campaign finance reform must be debated and voted on
>before the appropriation bill can come up.) It will give you an
>eye-opening picture of what happens to science and its reputation when it
>comes up against a strong political interest.
>
>
>Dave Johnson
>Federation of Behavioral, Psychological and Cognitive Sciences
------- End of Forwarded Message