Index Last Week Contact DQ NEWS Latest Issue
The MIT 1997 Information Quality Conference
-
New Directions for DQ/IQ
Several hundred people who attended the 1997 MIT Annual Information Quality Conference in Cambridge (October 24-26) heard and participated in sometimes-heated discussions about the future of Information Quality and Data Quality. Among the topics were:
-- Do current "inspect/repair" DQ/IQ methods for databases represent "up-to-date" Quality practice vis a vis what is being done in Industry (i.e., Manufacturing)? Do consultants, gurus, and seminar-givers who advocate and use database inspect/repair methods provide "state-of-the-art" DQ/IQ services to their clients?
-- Are "data warehouses" really necessary? What cheaper and more efficient processes can replace data warehouses?
-- Given the pace of regulatory reform in the United States (most recently involving the U.S. Internal Revenue Service) what are the implications for database methodology? How many and what kinds of databases will be "necessary" in the future?
-- Given the interest in data and information quality standards, what standards would be "best" for various organizations and activities?
-- How can DQ/IQ practitioners and organizations best dissseminate information about data and information quality? How can we make DQ and IQ interesting and important to diverse groups in fields like law, medicine, engineering, management, and education?
To obtain information about the 1997 MIT Information Quality Conference and news about the 1998 IQ Conference, see the MIT Information Quality Web site: http://web.mit.edu/tdqm
Data Quality Issues Complicate Election Investigation
According to an article in the October 26th issue of The New York Times, data quality issues are complicating the investigation of a 1996 U.S. House of Representatives election. The contested election - between victor Loretta Sanchez and losing incumbant Robert Dornan - and subsequent investigation now threaten to create legislative gridlock on the House floor during the last weeks of the session.
The investigation has already lasted 11 months, longer than the Iran-contra investigation took a decade ago. Representative Sanchez's margin of victory was 984 votes. Representative Dornan immediately contested the election, contending that many voters were not American citizens. The investigation is causing increased animosity between Democrats and Republicans. According to the Times, the Democratic leadership is threatening to use parliamentary procedure to tie up the House with extraneous votes if the investigation is not concluded soon.
The data quality issue in the investigation involves the cross-checking of 4,000 names from the Orange County, California, voter rolls with data from the U.S. Immigration and Naturalization service. The purported purpose is to determine how many non-citizens cast ballots, and the Republicans assume most or all the non-citizens cast their ballots for Ms. Sanchez. The Democrats claim that the INS data is unreliable. The INS concurs with the Democrats. Times staff writer Lizette Alvarez wrote the article which appears on page 26.
An Extra $46 Billion In Treasury's
Coffers
Puzzles Washington
A front page report in the October 28th issue of The Wall Street Journal examines U.S. Treasury Department estimates that over $46 billion - money largely responsible for shrinking the deficit in the fiscal year ending September 30th - ended up in Treasury's coffers.
Deficit projections assembled by the Congressional Budget Office in March estimated that by now the deficit would be about $124 billion. The projections were off by about $101 billion. Economists can explain about $55 billion of the windfall; a boom in the economy and a respite from health care inflation. But the explanation still leaves nearly $46 billion unaccounted for.
According to the Journal, a team of Treasury Department officials and economists is seriously concerned about finding where the missing $46 billion cam from. If such windfalls are likely to recur, they will predict higher revenues in future years - which could encourage Congress to launch new spending programs or to cut taxes. Most of the mystery money flowed in during the past months from unanticipated income tax payments from individuals.
The real mystery is on what kind of income the taxes were paid. Detailed data on taxes paid in April 1997 are only now starting to arrive at Treasury. It appears that those responsible tend to be higher income taxpayers with complicated returns. The hunch is that the surprise is related to the stock market - specifically income tax paid by wealthy individuals on capital gains, stock options, and bonuses. Budget estimators also guess the unexpected revenue may be linked to less predicable and increasingly complex ways in which top executives are being paid. Data from Wall Street indicate that 1996 bonuses paid to traders and investment bankers soared over 30% from 1995. The article was written by Journal staff writer Christopher Georges.
Fidelity Records Incorrect Prices For 18 Funds
According to a report in the October 30th issue of The New York Times, millions of mutual fund investors found no prices for their mutual fund shares in newspapers that were published on October 29th. An even bigger problem appeared at Fidelity, America's largest mutual fund company. Heavy stock trading and computer problems on the NASDAQ complicated Fidelity's attempts to adjust the prices of its funds on October 28th.
According to the Times, Fidelity was not able to calculate accurate prices for 18 of its mutual funds. Moreover, some mutual fund analysts have been been perplexed with Fidelity because of Fidelity's pricing of overseas investments. It should be noted, however, that in the same report the Times also acknowledged that Fidelity's competitors and several pricing services either were unable to supply mutual fund price data or supplied inaccurate mutual fund price data because of chaotic market conditions the week of October 26th. The article was written by Times financial reporter Edward Wyatt and appears on page D1.
People Can Claim One or More Races
On Federal Forms
A front-page report in the October 30th issue of The New York Times discusses the Clinton Administration's decision to adopt new rules for listing racial and ethnic makeup on Federal forms, allowing people for the the first time to identify themselves as members of more than one race. Under the Administration's proposal, people can identify themselves in a single racial category or combination. The change could affect Government policies like affirmative action and the drawing of legislative districts. It is the Government's first revision since 1977 in its definition of racial groupings.
The first official use of the new rules will be in next year's dress rehearsal of the 2000 Census. The rules took effect on October 30th after more than four years of study, debate, and testing. The new rules are expected to have a significant impact in places like California, which has a high percentage of people with mixed racial backgrounds.
According to the Times, the Administration has yet to issue a policy about counting people who denote themselves as multiracial. The counting issue is important because various statutes like the Voting Rights Act mandate various actions based upon race. About 1% of Americans consider themselves as being of more than one race. At least 2 million children live in mixed-race families.
The issues over how to list people of mixed race are intricate and resolving them will be arduous. For the most part, demographers, academics, and groups representing multiracial people would prefer dividing people into as many sub-groups as possible, so as to better study economic, social, and cultural trends. On the other hand, civil rights groups fear the dilution of the number of people identified with a single group. The article was written by Stephen Holmes.
CBS to Pay Print Journalists for 'Story' Tips
A report in Iver Peterson's "Media" column (The New York Times, October 27th) discusses a CBS News offer to pay print journalists $500 to $1000 for news tips that get on the air. CBS has already sent E-mail to some court and police reporters nationwide, offering $1000 for an original tip that leads to a story on a news magazine program like "60 Minutes," and $500 for a tip leading to a story on the "CBS Evening News." The news reports cannot already been broadcast elsewhere on television, and the preference is for news items that have not appeared anywhere - including the reporter's own newspaper. According to CBS, the network is looking for the best stories, wherever it can find them. And people in the television news business are finding it increasingly difficult to find interesting "stories" in a time of peace and prosperity.
According to the Times, newspaper editors are not happy about the CBS offer. They generally feel that any ideas their reporters come up with belong to the newspaper. At least one editor interviewed for the column said he would fire any reporter providing a news tip to CBS. The CBS reward for news tips is significant. Most entry-level newspaper reporters at smaller papers earn about $500 per week. And most reporters across the nation earn less than $1000 per week, according to the Newspaper Guild of America. The "Media" column appeared on page D11.