Please note: Older issues of the newsletter are likely to contain
broken links -- the newsletter is presented here "as published."
Table of Contents
Researcher's Notes by Mike McQuestion
- CIC Learning Technologies Database
- Downloadable American Nat'l Election Studies
- Mexican Statistics Online
- NAFTA Tracking Database
CNI White Paper on Gov Pubs
"Access to and Services for Federal Information in the Networked Environment" is the title of a new white paper produced by CNI, the Center for Networked Information. The paper discusses changes both in technology and governmental policy that require a new look at how government information is distributed to the public.
The Federal Depository Library Program (FDLP) is the traditional way for the public, through libraries, to obtain access to publications created with their tax dollars, at no added charge. Title 44 of the U.S. Code specifically mandates the FDLP and the Government Printing Office Sales Program--which designates the GPO as the primary distribution channel for publications produced by government agencies. For example, the State Historical Society Library on campus is a designated depository library for U.S. government documents in this region of the state.
The networked environment is exacerbating some problems with the FDLP/GPO traditional access points.Many agencies are sidestepping their requirement to print/distribute through the GPO, finding it cheaper to publish documents on their own. Similarly, they may set up a website without going through GPO Access. This decentralization of government information may reduce costs for individual agencies, but without the organization of information established by the FDLP/GPO program, public access to government information is as haphazard as the WWW itself.
Data users are well aware of the benefits of new computer technology for accessing government information. Anyone who has grabbed tables from a government agency website for a presentation, or made use of an extraction program on a government-issued CD-ROM knows this is an improvement from previous distribution methods. Of course, this assumes the availability of adequate hardware and network connections.
Neither format, however, lends itself to standardization. The dynamic nature of the World Wide Web means that both the content and location of pages tend to change frequently, unlike print or microfiche. Data on optical media are produced with various interfaces, changing with hardware and operating system upgrades. The need to learn new software with every publication adds to "information overload."
The white paper refers to other problems revolving around the organization--or lack of it--in networked information. While there is no authoritative index or catalog to the Web, there is a "proliferation of websites acting as pointers only, resulting in a duplication of efforts." Search and retrieval on the web is still "hit and miss," while browsing mechanisms leave much to be desired. Also, the main advantage of networked information-- accessibility from home or office--is undermined by password restrictions for government subscriptions such as the Commerce Departments STAT-USA. Here, one must go physically to the Historical Library to use the databases offered. Read the white paper at http://www.cni.org/projects/fedinfo.
News from NORC
"At the recommendation of the American Statistical Association, the US Bureau of Labor Statistics has undertaken a Redesign of the Current Employment Statistics (CES) Series. This series is a source of some key economic indicators such as total employment by industry, hours, and earnings. In collaboration with the State of Illinois, NORC will develop and evaluate alternative estimators for key economic indicators at various reporting levels. NORC will empirically test potential estimators that may reduce data variance and minimize bias on CES sample data. NORC will also test alternative sample rotation plans and benchmark schedules and will test and develop methods of estimating the variances of the series estimators." (National Opinion Research Center website: http://www.norc.uchicago.edu)
Two other NORC projects currently in the field also may be of interest to data users. The Early Childhood Longitudinal Study (ECLS) in combination with the Head Start Longitudinal Study (HSLS) will follow young childrens transitions from pre-school to the fifth grade, tracking variables associated with success. The Prospective Longitudinal Study of Adolescent Health will interview 96,000 high school students through the use of audio-CASI, a successful survey method for sensitive topics in which the student listens to questions through headphones and enters a response directly into the computer.
by Mike McQuestion
I am a dissertator in Sociology/Demography. Recently I've begun to research poverty in the U.S. at the county level. At the DPLS I have found several interesting datasets for this project. The DPLS staff has helped me find and extract data from both CD-ROM and ASCII files, including USA Counties 1996, Bureau of Health Professions Area Resource File, 1940-1994, and the Census of Population and Housing PUMS 5% Sample (Public Use Microsample), 1990.
From these I have extracted variables on county poverty rates, personal income, housing, household structure, educational and crime levels, local government expenditures and revenues, out-of-pocket health expenditures, occupations, capital investments, payrolls and earnings by industrial sector, levels of employment, wages, social security and other transfer payments. With these data I am modelling micro-level poverty in new ways.The DPLS staff continue helping me identify often obscure datasets which I have not been able to locate elsewhere.
Census Sampling in Jeopardy
Readers should follow a news item regarding a Senate Appropriations Committee decision to attach language to a supplemental spending bill that would prohibit the use of sampling in the year 2000 Census. The full Senate was expected to debate S. 672 after DPLS News went to print. The Census Bureau had planned to save about $1 billion by using sampling techniques to infer the makeup of households after 90 percent of a region has been contacted. Sampling is expected to improve counts for minorities and both the urban and rural poor. While Democrats are expected to support sampling, Wisconsin Senator Herb Kohl, a committee member, is outspoken against the use of sampling. Wisconsin retained one of its seats in the House of Representatives last year after sampling was disallowed for the 1990 Census.
Census 2000 Sampling updates:
May 16, 1997
June 10, 1997
Go to October 1997 DPLS News, "To Sample or Not to Sample"
Online Archival Repository:
Slave Movement in the 18th-19th Centuries
This collection encompasses eleven individual studies compiled by Herbert S. Klein of Columbia University and others during the 1970s. The studies are as follows: Records of Slave Ship Movement Between Africa and the Americas, 1817-1843; Slave Ships of Eighteenth Century France, 1748-1756, 1763-1792; Slave Trade to Rio de Janeiro, 1795-1811; Virginia Slave Trade in the Eighteenth Century, 1727-1769; English Slave Trade, 1791-1799 (House of Lords Survey); Angola Slave Trade in the Eighteenth Century, 1723-1771; Slave Trade to Rio de Janeiro, 1825-1830; Internal Slave Trade to Rio de Janeiro, 1852; Slave Trade to Havana, Cuba 1790-1820; Nantes Slave Trade in the Eighteenth Century, 1711-1791; and Slave Trade to Jamaica, 1782-1788, 1805-1808. http://dpls.dacc.wisc.edu/slavedata
These studies vary in content and provide access to a range of variables such as: ships port of arrival, date of arrival, type of vessel, tonnage, masters name, number of guns, number of crew, national flag, number of slaves, port of departure, number of days of voyage, mortality, date ship sailed, site of slave purchase, site of slave sale in America, number of slaves sold, date ship returned to port, number of deceased crew members, number of children, slave mortality during voyage, slave mortality immediately upon landing, type of slave ship, ships name, home port of vessel, port of construction, year constructed, year of registration, type of owner, origin of slaves, date of departure from Africa, date of arrival in New World, date ship left America, physical capacity of ship, number of sailing days, person or company to whom slaves consigned, ship identification, place of birth, occupation (by sex) of slave, color, captain of ship, and goods carried.
Data Producers Wanted
Are you a researcher whose projects result in machine-readable data? Based on the premise that high quality data, once generated, should be fully and effectively exploited by as many researchers as possible, granting agencies such as the National Science Foundation, the National Institute of Justice, and the Department of Health and Human Services now require researchers to archive data collected as a result of funded projects. The DPLS has a long tradition of assisting investigators in the production and preservation of public use files that will facilitate the dissemination and utilization of their data by others.
More recently, we have been adding archival datasets to our WWW site for users anywhere to browse and download. See http://dpls.dacc.wisc.edu/archive.html for current offerings.
Additionally, if you have purchased data for your own purposes, and would like to make them available to other users on this campus, DPLS is happy to make copies of both data and documentation and place them in the library. In this manner we have added many important studies to our collection. If you are interested in receiving information on how to produce public use files, or would like to submit data for inclusion in the DPLS Archive Online Repository on our website, please contact us.
ASC Call for Papers
A Call for Papers has been issued by the Association for Survey Computing for a one-day conference at the Imperial College in London, September 30, 1997 on the topic "The Internet, the Interview, the Impact." Topics to be addressed include how to do Computer Assisted Self Interviewing on the Internet, whether representative samples can be drawn, and long-term issues for how the Internet will affect survey research. For more information, email firstname.lastname@example.org. The closing date for receipt of abstracts is June 16.
CIC Learning Technologies Database
The Consortium of Institutional Cooperation (CIC) Learning Technologies Database is a Web-based inventory of instructional resources for the use of faculty and staff on the CIC campuses, including the University of Wisconsin-Madison. According to the electronic publication "CIC/LTI Report" (which reports on the CIC Learning Technology Initiative), the database allows teachers "to enter detailed descriptions of instructional resources under development or in use at their own institutions into a common database management system that is accessible throughout the consortium. The system has been designed to include a variety of materials, including course syllabi, electronic texts and databases, archives of electronic media, development tools, training programs, instructional software, and interactive communications systems. Information in the system can be retrieved by users in any of the CIC domains and is searchable by category and by keyword. The database is available at http://lti.cic.net (as of 7/9/98, the link to this database is no longer functional).
Downloadable American Nat'l Election Studies
For the first time, NES is releasing the 1996 ANES directly to the public through the Internet. Users may download the data, ASCII codebook, SPSS and SAS data definition files from the WWW after filling out a brief registration form. Along with the usual focus on the elections, these data contain special foci on the congressional campaign, issue importance and uncertainty, media exposure, social capital, and a module coordinated through the Comparative Study of Electoral Systems, a project developed through a consortium of scholars from 52 polities. More information about the data can be found at http://www.umich.edu/~nes/studyres/nes1996/nes1996.htm.
Mexican Statistics Online
The National Institute of Statistics, Geography and Informatics (INEGI) of Mexico was established in 1983 to gather, process and publish data about Mexico's population, economy, and society through collection of the National Census, household surveys, and administrative records. Geographic data is also collected and maps created within the Institute. Much information is distributed via its website (in Spanish and English), including information about the North American Industrial Classification System, the National Development Plan, a catalogue of products, and highlights about Mexico. Detailed geographic data on the website include colored maps on land area, geology and soils, climate and watersheds, vegetation and fauna and communication routes. Socioeconomic data include short-term economic indicators, as well as tables on population, education, health, employment, productivity and infrastructure. http://www.inegi.gob.mx
NAFTA Tracking Database
The North American Integration and Development (NAID) Center at UCLA was created to conduct ongoing research concerning NAFTA and North American integration, and to assist communities and governments with policies and investment projects for sustainable and equitable development across borders. "Researchers at the NAID Center are developing a unique capacity to monitor and model the impacts of North American integration and the adjustment process. Among the trends that are monitored are trade, capital flows, and migration effects on employment and income throughout the U.S. and Mexico." Sources such as County Business Patterns and INEGI (above) are used to chart NAFTA related changes, such as company layoffs in the United States and rate of increase in the Maquiladora industries. State of the art web-based forms and graphs make this site as easy to use as it is fascinating. http://naid.sppsr.ucla.edu/database.html
End of May 1997 Newsletter.