| Index | Last Week | Contact DQ News | Latest Issue |
| Featured Articles | Data & Information | Science & Technology | Education |
| Economics | Business | Law | Medicine |
Data Quality Issues Complicate Slavery Archives
An article in the November issue of The Library of Congress Information Bulletin, (Vol. 57, No.11) examines the first-hand narratives of former slaves that were written and recorded between 1920 and 1945 by American folklife scholars. The men and women who survived slavery had much to tell. And as the first generation of black people born in freedom came of age, fears that the slave experience would be lost forever troubled some scholars, particularly at Black colleges. The accounts published by black historians at these colleges during the 1920's had little impact on the larger historical profession.
With the advent of the Great Depression, and subesquent Roosevelt Administration programs like the Federal Writer's Project (an expansive federal effort that obtained interviews from thousands of former slaves in 17 states) which collected tens of thousands of pages of typescripts. Another group of scholars began to record the words and songs of the former slaves. Using primitive recording equipment, forlklore specialists - including John and Ruby Lomax and others - were able to capture the voices of former slaves. In 1939, control of the Federal Writer's Project passed from the federal government to the states. The interviews and the sound recordings were eventually deposited at the Library of Congress.
The written interviews and recordings have caused several information quality controversies since the 1920's. The first interviews were conducted almost 60 years after slavery ended. The 1920's interviews by Black historians were criticized because the respondents were thought to be too old and represented about 2% of the ex-slave population. The Federal Writer's Project interviews were conducted largely by white interviewers in the then-segregated South. Moreover, typescripts of "conversations" were often heavily edited reconstructions with "objectional" language, characterizations, and events eliminated or changed. And local authorities sometimes prevented ex-slaves from being interviewed. A core group of interviews and recordings conducted by experienced and sensitive folklorists are thought to have accurately captured the ex-slaves' experiences. Linguists found the recorded interviews a rich opportunity to study the development of Black English Vernacular. Moreover, recent digital remastering of the old acetate and aluminum discs made a fuller rendition of the recordings possible. The article was writtten by slavery historian Ira Berlin and is available on the Library of Congress Web site: http://www.loc.gov/loc/lcib/ .
Telephone Pollsters Switch to Recruiting Respondents
According to an article in the November 2nd issue of USA Today, the increasing reluctance of voters to submit to telephone interviews has caused a significant decrease in polling by telephone. From the 1930's to the early 1980's, people who owned telephones were respectful when answering the telephone because telephone calls were deemed "important." According to USA Today, during the past decade Americans have become increasingly unfriendly to callers. This appears to be the result of the number of annoying calls Americans receive each year. Consequently, the response rates for telephone polls have dropped from a peak of 75-80% a decade to about 45-50% today. The response rate for face-to-face interviews has remained about 70%. Refusals are important because the unwillingness to be interviewed might correlate with something being measured, like income or race.
Pollsters are now seeking alternatives to telephone interviews. Unlike face-to-face interviews, telephone interviews are cheap and the results can be quickly tabulated. Some researchers are trying polling by mail. Others are recruiting respondents and providing electronic polling equipment. Respondents are being selected from many lists and databases, not just by telephhone number. The article was written by Philip Meyer, a journalism professor at the University of North Carolina at Chapel Hill, and appears on page 19A.
Food Portion Data May Be Flawed
A report in the November 3rd issue of The Washington Post health section claims that Americans' concept of food serving size far exceeds the "standard" portion sizes the U.S. Department of Agriculture uses to calculate dietary recommendations.
For example, the USDA defines a "serving" of pasta as 1/2 cup cooked, or about 4 forkfuls. A "serving" of meat is 2-3 ounces. A "serving" of wine is 4-5 ounces. But the portion size of some common foods - breads like muffins and bagels and take-out meals from restaurants and delicatessens - may be 2-5 times the USDA standard portion size. The concept of "serving size" is important because the USDA conducts food consumption surveys in which respondents are asked how many servings of various foods they eat.
To add to the confusion, everyone appears to have a different idea of what the words "small," "medium," and "large" mean. A "medium" potato may weigh 4-8 ounces. A "medium" serving of frozen yogurt may be 12 ounces. A "medium" deli sandwich may have 12 ounces of meat (3-4 USDA "servings").
Although the USDA nutrition studies claim that women eat about 1,600 calories a day and men eat 2,400, some nutritionists say the numbers can't be right, given the level of obesity in the United States. The Post article was written by Lawrence Lindner. The article was based on research conducted by the American Dietetic Association.
SEC Focuses on Muni-Bond Lawyers' Opinions
The U.S. Securities and Exchange Commission, which has focused on underwriters and public officials in its five-year crackdown in abuses in the municipal-bond market is now shifting its attention to municipal-bond lawyers, according to an article in the November 6th issue of The Wall Street Journal. Although the SEC has brought about 60 actions against underwriters and other public officials, bond lawyers have been relatively unscathed. Now the agency has decided to add municipal-bond attorneys (often called "bond counsels") to those it scrutinizes.
Acting as advisors to cities and states issuing bonds, attorneys typically provide opinions that certify bonds as tax-exempt, and often help municipalities select bond markets and underwriters. The SEC is particularly concerned about the type of advice lawyers give unsophisticated, smaller municipal-bond issuers. The SEC also wants to crack down on attorneys who fail to provide investors with adequate information about muni bonds being issued.
But the Commission can't merely bring a case against a bond lawyer for providing "bad advice." Rather, the SEC must prove that the lawyer "consciously knew or recklessly disregarded facts" in bond documents circulated to the public. The Journal article appears on page C19 and was written by staff reporter Charles Gasparino.
Devious Information Brokers Tap Private Financial Data
According to a front-page report in the November 5th issue of The Wall Street Journal, personal financial information is routinely revealed to financial "information brokers," a group of fast-growing firms that specialize in finding financial data. Although many information brokers use only legitimate public records to conduct their searches, others employ ingenious ruses to pry into customers' bank accounts and brokerage records. The most widely used practice involves impersonating an account holder to obtain balances, according to federal and state regulators.
The rise of unscrupulous asset searchers has been fueled by the Internet, which has made it easier for them to advertise and gather information on individuals. These data, in turn, make it easier to impersonate an account holder when calling a financial institution. Dishonest and illegal practices include impersonating bank customers, mailing phony rebate checks to bank customers to obtain the customers' account numbers, or use bogus offers of credit to trick people into divulging bank information and Social Security numbers. Dishonest information brokers also call telephone companies and obtain both unlisted numbers and account information by impersonating customers. And unscrupulous information brokers may obtain credit reports they aren't eligible to receive by submitting phony documents to credit reporting firms.
Where all the financial data about individuals finally ends up is anyone's guess. It is obvious from other articles abstracted on this Web site that financial and other data are being brokered that are both fragmented and of poor quality. (Is it any wonder that people don't want to answer the telephone, much less respond to telephone polls?) The article was written by Journal staff reporters Jon Auerbach, Mark Maremont, and Gary Putka.
Go to: Data Quality Home Page
Comments: dqemail@aol.com (11/1/1998)