DPLS News, February 2001

Please note: Older issues of the newsletter are likely to contain
broken links -- the newsletter is presented here "as published."

DPLS News contains articles about local, national, and international data issues.
It is published twice a semester by the library staff.

Editor: Joanne Juhnke, Associate Special Librarian
Contributors:
Lu Chou, Special Librarian & Cindy Severt, Senior Special Librarian

February 2001


Table of Contents

Census Jockeying for Political Gain
2000 National Election Studies
New ECLS CD-ROM
ADDHealth CD-ROM with Search & Retrieval Software
Researcher's Notes
Querying Public Opinion Polls by Question

Internet Corner
     University Information System - Russia (UIS RUSSIA)
     National Survey of America's Families (NSAF)
     ChildStats.gov
     Fast Facts: Almanacs/Factbooks/Statistical Reports & Related Reference Tools


Census Jockeying for Political Gain

The U.S. elections in November left little doubt that counting is not always a straightforward process. The result of the vote tabulation in Florida has the potential to impact the result of a much more complicated count, the decennial Census.

The first census results, the apportionment counts, were released at the end of December. While the U.S. population grew overall, up 13.2% since 1990, the largest gains were in the south and west. Twelve Congressional seats are slated to shift between the states based on the first results. Initial apportionment counts may be found online at http://www.census.gov/main/www/cen2000.html.

Census 2000 logoA 1999 Supreme Court ruling determined that the apportionment counts should be based solely on raw numbers, not statistically adjusted for any potential undercount or overcount. However, the door was left open for statistical sampling to be used for redistricting and other uses. Redistricting data at the block level is scheduled for release at the beginning of April (see http://www.census.gov/population/www/censusdata/c2kproducts.html for an overview of Census 2000 products and release dates).

In the past, undercounts have been largest among minority and lower-income populations. Since those groups tend to vote more heavily Democratic, political opinion regarding sampling has broken primarily along party lines, with Democrats in favor of statistical adjustment and Republicans committed to "actual enumeration." Former Census Bureau director Kenneth Prewitt, a strong proponent of sampling, resigned in January when he was not asked to remain by President George W. Bush. Acting Census Bureau Director William Barron is slated to decide by early March whether to recommend the release of statistically adjusted Census data, a recommendation that the President might choose to overrule.

Some sources have indicated that President Bush may rule out the use of adjusted data for redistricting, but may allow it for federal funding allocations.

It is expected that the census undercount, although still substantial, will be reduced in 2000 as compared to previous decades. A marketing campaign boosted response rates in 2000, and automated handwriting analysis on the returns proved more accurate than human analysis. A large undercount of infants under a year old in 1990 will likely be eased by the wording of the 2000 form, which asked for birthdate instead of age.

One early challenge to Census 2000 comes from the state of Utah, arguing against the practice of counting overseas military personnel but not overseas missionaries. If Mormon missionaries from Utah had been counted, Utah would have gained a Congressional seat that has instead gone to North Carolina.

The issue of racial categories will also figure prominently in the use of Census 2000 data. For the first time in 2000, respondents could identify with more than one racial background, an innovation that produces 63 categories instead of the traditional five..

Table of Contents

 

2000 National Election Studies

National Election Studies (NES) have been conducted around each national election in the United States since 1952. In presidential election years, the study is conducted both before and after the election (a pre/post-election study), while in congressional election years the study is conducted only after the election (a post-election study).

The 2000 NES Special Topic Pilot Study is now available from the NES web site, http://www.umich.edu/~nes/, by following the Download Data link. This was a 14-minute telephone interview conducted in the spring of 2000. Specific topic areas include: social trust, trust in election, civic engagement and a new method for measuring social desirability norms.

A major change in the design of NES 2000 is that some of the respondents were interviewed over the phone using Random Digit Dialing. This new feature allows researchers to compare NES interviews taken in person with the interviews taken over the phone, and perhaps move to phone interviews as a more cost-effective method. In the past 50 years NES has been conducted by face-to-face interviews only.

To assess the turbulent 2000 presidential election, additional topics were introduced, including: the fairness of elections; satisfaction with democracy; value of voting; divided government; environmental politics; abortion; and religious background of the candidates. The data for NES 2000 will be available on the NES web site in March in early release format.

Table of Contents

 

New ECLS CD-ROM

DPLS has obtained on CD-ROM the Early Childhood Longitudinal Study-Kindergarten Class of 1998-99 (ECLS-K) public-use base year data from the National Center for Education Statistics.

The ECLS-K follows a nationally representative sample of about 22,000 children from kindergarten through fifth grade. Information is collected from the children's schools, teachers, and parents. These children come from both public and private schools, and attended both full-day and part-day kindergarten programs. The ECLS-K provides national data on (1) children's transitions to non-parental care, early education programs, and school and (2) children's experiences and growth through the fifth grade. The data can be used to test hypotheses about the effects of a wide range of family, school, community, and individual variables on children's development, early learning, and early performance in school.

Table of Contents

 

ADDHealth CD-ROM with Search & Retrieval Software

The public-use version of the National Longitudinal Survey of Adolescent Health (aka ADDHealth) is now available to DPLS users via CD-ROM with search and retrieval software. Collected in two waves from 1994 to 1996, the ADDHealth study was mandated by Congress to assess the impact of social environment on adolescent health. Variables include diet and nutrition; eating disorders; depression; violent behavior; injury; suicide; exercise; health service use; and health insurance coverage. The data include in-school questionnaires for adolescents and administrators; an in-home interview; the ADDHealth Picture Vocabulary test; a parent questionnaire; and a specially constructed set of community contextual data (not integrated with the search and retrieval software).

Table of Contents

 

Researcher's Notes
by Paul Manna

I am a graduate student in the Department of Political Science here at the UW. My dissertation explores the rise and continued presence of the issue of education on the national political and policy agendas. In particular, my work relates changes in federal law and national public opinion to developments in education policy at the state level.

Because my project looks at education policy during the last four decades, I have searched high and low for data sources that will allow me to capture over-time changes in public opinion. To that end, I was especially glad to learn of DPLS' membership with the Roper Center for Public Opinion Research, which allows professors and graduate students at the UW to acquire datasets from the Center's archive.

In my particular case, our membership enabled me to obtain over twenty of the Annual Surveys of the Public's Attitudes Toward the Public Schools. These surveys, which began in 1969 as a collaboration between Gallup and the educational organization Phi Delta Kappa, will enable me to evaluate changes in the public's attitudes on a range of issues including school integration, national testing, and vouchers. In many of the surveys the questions are explicitly political--e.g., Which political party is best suited to deal with education?-and tie in directly to the kinds of questions that I am exploring in my dissertation. The Roper Center is an important source for my particular project, but also clearly speaks to issues that researchers in other departments, from Sociology to Education Policy Studies, would no doubt find useful.

Editor's Note: DPLS invites our users to visit the Roper Center online at http://www.ropercenter.uconn.edu, and to contact DPLS staff for Roper Center data requests.

Table of Contents

 

Querying Public Opinion Polls by Question

Ever wonder if the recent increase in the price of natural gas has affected everyone else as it has you? Are you curious about whether or not the public has the right to know if an officeholder or candidate ever abused alcohol? Do most people believe hip hop music contains too much violence? The statistical results of these and other public opinion questions can be found via iPOLL, the Roper Center's full-text, question-level retrieval database designed to enable users to locate and examine questions asked on national public opinion surveys.

On the UW-Madison campus, iPOLL can be accessed through the campus-wide subscription to LEXIS-NEXIS Academic Universe. From http://web.lexis-nexis.com/universe/ click on the Reference link. Click next on the Polls & Surveys link. Of the seven search fields presented, only one - Keyword/Question - is required. A keyword search will look for the search string in the full text of the polls; a question search will look for the search string in the wording of the question. Searches can be greatly widened or narrowed by specifying a date ranging from the current day to the past ten years.

All of the surveys in iPOLL are national in scope, and represent major survey organizations including, but not limited to The Gallup Organization, Louis Harris & Associates, CBS News, Los Angeles Times, and the Washington Post. And since DPLS is a member of the Roper Center, your request for public opinion data is just a phone call or e-mail message away!

Table of Contents

 

 Internet Corner

University Information System - Russia (UIS RUSSIA)

This password-protected collection, primarily in Russian, contains a wealth of information about Russia since the early 1990s. Sources include governmental laws and directives, daily records of the State Duma, Russian Federation data (Goscomstat), economic and social data, election statistics, and much more. The collection is intended for social science education and research, and is geared toward university users. Users will need to visit DPLS in person to be logged in.

The UIS Russia website can be reached at http://www.cir.ru/eng/.

National Survey of America's Families (NSAF)

Sponsored by the Urban Institute, the National Survey of America's Families provides quantitative measures of the quality of life in America, paying particular attention to low-income families.

The survey represents the noninstitutionalized, civilian population of persons under age 65 in the United States as a whole, and in 13 states in particular: Alabama, California, Colorado, Florida, Massachusetts, Michigan, Minnesota, Mississippi, New Jersey, New York, Texas, Washington, and Wisconsin.

Data is available for downloading from the site (free registration required), along with a cross-tabulation tool that can either be used online or downloaded.

The NSAF site is located at http://newfederalism.urban.org/nsaf/index.htm.

ChildStats.gov

Forum on Child and Family StatisticsThe ChildStats.gov website is the home page of the Federal Interagency Forum on Child and Family Statistics. The site provides links to federal and state statistics and reports on children and their families, including: population and family characteristics, economic security, health, behavior and social environment, and education. Reports of the Federal Interagency Forum on Child and Family Statistics that can be found on the web site include America's Children: Key National Indicators of Well-Being, the annual federal monitoring report on the status of the nation's children, and Nurturing Fatherhood.

The name of the site reflects its address: http://www.childstats.gov/.


Fast Facts: Almanacs/Factbooks/Statistical Reports & Related Reference Tools

This resource, compiled by librarian Gary D. Price at George Washington University, is a handy, current, well-organized collection of links to ready-reference sources with fast facts and tables on about 150 topics from agriculture to zoology. The links are not intended to be comprehensive, but instead include a selection of "quick look-up" kinds of resources. While not all the entries are social-science related, the country and state listings alone would make this site worth bookmarking. The topical entries include choices such as health, juvenile justice, labor market, and many other useful pointers.

Visit the Fast Facts page at http://gwu.edu/~gprice/handbook.htm.

Table of Contents