Minutes of the COPAFS Quarterly

December 13, 2002

COPAFS chair Maurine Haver opened the meeting, and turned to Jerry Fletcher, who introduced the slate of officers and at large board members for the coming year. The slate was voted in by acclamation, and we went to Ed Spar for his executive director's report.

Ed took just a moment to note that he has now been COPAFS executive director for 10 years. He then commented that, among the few members of Congress with genuine interest in and knowledge of federal statistics, we will lose three-Tom Sawyer, Steve Horn, and Dan Miller-when the new Congress convenes. And for those keeping a scorecard on agency heads, Gary Phillips, acting commissioner of the National Center for Education Statistics is retiring, and Ashish Sen, director of the Bureau of Transportation Statistics, is leaving. Turning to budgets, Ed reported that nothing has happened since the last meeting. As he described it, the situation is still very unclear, but the outcome will not be good. Ed then drew our attention to the document describing the recently passed "Confidential Information Protection and Statistical Efficiency Act of 2002." He commented that the bill will be signed soon, and expressed thanks to Kathy Wallman, Steve Horn, and Tom Sawyer for their efforts in getting the legislation passed.

The good news on the ACS is that the Census Bureau is going ahead with the 750,000 national sample as well as the 31 test sites. Ed also described the Decennial Census Advisory Committee letter to Commerce secretary Evans concerning the ACS, and the reply from Secretary Evans affirming the Department's support for the ACS. However, full implementation remains unresolved, so Ed will keep COPAFS members updated on developments. The economic census forms have been sent out, but funding is insufficient to cover tabulation. However, Ed reported that Congress is indicating that the economic census is a priority, and that funding will be provided. As has been widely reported, the adjusted 2000 census data are being released, and have already been provided to Congress. Ed indicated that these data will be available through an unspecified website (probably soon), and he will notify COPAFS members when that happens.

Ed noted that there has been much talk of agency downsizing and outsourcing-in particular at the Bureau of Justice Statistics. There is concern that OMB Circular A-76 is being reviewed, and could be invoked to justify the outsourcing of many statistical functions to the private sector. Ed then described as "disturbing," two provisions of the Homeland Security Act. First, meetings called under the heading of homeland security need not be open to the public, and second, data provided as a function of homeland security do not have to be released under the Freedom of Information Act. Combined with all the information being removed from federal websites, these provisions are seen as ominous.

Ed then showed a CD with 2000 census modified race data at the county level. Although it provides full age/sex crosstabulations, it was noted that the age data themselves are not modified (as they were in the 1990 "MARS" data), and that Paul Voss and others have identified problems with the 2000 census age data. Ed concluded by announcing next year's COPAFS meeting dates: March 14, June 13, September 19, and December 12.

Federal Highway Data Programs

Barna Juhasz. Federal Highway Administration

Barna explained that the Federal Highway Administration (FHWA) was established in the late 1960s (at about the same time as the Department of Transportation), but that its function dates back to 1893, when the Office of Road Inquiry was established. In 1904, the first national road inventory was conducted at the request of Congress. It identified 2.2 million miles of public roads and annual expenditures of $80 million on these roads. By 2001, the nation was up to 3.9 million miles of roads and $130 billion in annual expenditures. There was comment that, while total miles has not grown dramatically since 1904, the nature of these roads (number of lanes, surface, etc.) is dramatically different.

Apart from a few surveys, FHWA does not collect its own data, but rather acquires data from states. The states collect the data for their own purposes, and while none of the data are mandated, FHWA establishes data reporting guidelines to promote consistency across states.

Barna described three major components of the federal highway data program. First is the Highway Performance Monitoring System, which provides data on the extent, condition, performance, and use of the nation's highways. Second is Highway Funding and Motor Fuel, covering information on fuel consumption, vehicle registrations, driver licensing, and motor carrier taxation. These data are used to distribute federal transportation funds back to states. Third is the Heavy Vehicle Travel Information System, which provides estimates of heavy vehicle highway travel in response to concerns over the increasing volume of truck traffic.

The FHWA also provides a number of special purpose databases. Among these are the Census Transportation Planning Package, which provides special tabulations of census journey-to-work data for local transportation planners, and the National Household Travel Survey, which supports national policy development by providing descriptive information on the modes, purposes and timing of household travel.

Barna wrapped up by describing some of the FHWA's data products and publications, including the annual "Highway Statistics" printed report, which seems to be a flagship product. The agency also provides reports to Congress, and performance measurement reports. Data products are available through the agency's website www.ghwa.gov/ohim/ohimstat.htm.

FedStats: The Next Generation

Marshall DeBerry. FedStats/BJS
Valerie Gregg.
National Science Foundation

Valerie Gregg introduced Marshall DeBerry as the new FedStats project director-a new position as of October 1. Marshall started with background, recalling that FedStats was developed in the mid 1990s to provide one stop shopping for what is still a very decentralized system of federal statistics. The system went live in May 1997, with the mission of providing effective, efficient and timely access to the full range of federal statistics, and the objective of promoting informed decision-making. The system is funded by the 14 member agencies of the Interagency Council on Statistical Policy, and uses "open source software" designed for the fast delivery of information, and optimized for a wide range of delivery devices.

On-going FedStats projects include working papers on Section 508 statistical tables accessibility. Section 508 (of the Americans with Disabilities Act) requires that all tables be accessible to persons with impairments. For example, a sight-impaired person might need an auditory version of a data table. Nick Zill of Westat commented that this is a big issue, as this requirement places a tremendous burden on FedStats, and is slowing the system down. The extent to which such requirements might apply to other federal statistical agencies was raised as an issue for COPAFS to look into. Other on-going projects include MapStats for Kids, and outreach and promotion. New projects include improved automated access to statistical databases, the feasibility of a city MapStats product, and access to historical data.

Marshall noted that FedStats already fits in well with E-Government objectives for the provision of better services to citizens. In fact, FedStats is part of a cross-agency E-Government solutions working group. In its next generation, we can expect FedStats to provide technological enhancements for improved information finding and extraction-including the ability to communicate information to devices (such as PDAs) with limited screen space. We can also expect improved metadata for the interpretation and integration of data.

Marshall concluded with a description of the MapStats product-that provides maps with integrated data-and a "semi-live" demonstration of the system. Valerie pointed out that MapStats is the first time that data have actually been carried on FedStats, as the system had previously been only a gateway to other federal data sources.

The National Survey of America's Families: Update on Methods and Findings

Ken Finegold and Kevin Wang, Urban Institute

Ken started with a review of recent changes in US social policies, including welfare reform, restrictions on benefits for non-citizen immigrants, and tougher policies on child support. He described the National Survey of America's Families (NSAF) as a component of the Urban Institute's Assessing the New Federalism (ANF) project, which is designed to examine the impact of "devolution" (shifting responsibilities to state and local governments) on income security, healthcare and child well-being.

Kevin noted that the NSAF is the largest component of the ANF project, and is intended to estimate the well-being of children, as well as non-elderly adults and their families. The survey is nationwide, but oversamples 13 "focal" states-Washington, California, Colorado, Texas, Minnesota, Wisconsin, Michigan, Mississippi, Alabama, Florida, New Jersey, New York, and Connecticut. The survey also oversamples low income households and households with children. It is primarily a random digit dial telephone survey, but does include households with no telephone. The survey was taken in 1997 and 1999, and data collection has just been completed for the 2002 survey.

Interviewing is done in both English and Spanish. Up to two children are selected in each household, and the household's "most knowledgeable adult" is asked questions about these focal children, and about their spouse or partner. For households without children, up to two adults (age 18-64, and who are not spouse/partners of each other) are randomly selected.

Survey content includes household composition and demographics, health status, insurance, employment, income, poverty status, welfare or other program participation, and child care arrangements.

Survey methodology reports are available on the NSAF website http://newfederalism.urban.org/nsaf/methodology.html, and public use files can be downloaded at www.urban.org/content/Research/NewFederalism/NSAF/PublicUseDate/PubUse.htm. All reports are available free of charge online at http://newfederalism.urban.org.

Ken concluded the presentation with a review of some NSAF findings. The findings document the drop in welfare rolls, but also that the earnings of those leaving the rolls are low. The survey also identifies that there are still more than two million families on assistance, and that many of the poor are not participating in programs for which they are eligible.

Development of the Bureau of Labor Statistics Time Use Study

Tom Nardone (for Diane Herz), Bureau of Labor Statistics

Tom noted that the American Time Use Survey (ATUS) is a "new" BLS survey, but that work actually started in the mid to late 1990s. The study will be conduced by the Census Bureau starting January 2003, using former Current Population Survey respondents. Asked about the willingness of former CPS respondents to participate in yet another survey, Tom explained that the respondents would be given a two-month break, but that they need to follow up quickly before too many respondents move. Tom also noted that because the ATUS is tied to the CPS, it will have a wealth of other information on its respondents.

The survey will measure how people spend their time, as well as where and with whom they do activities. It will gather information beyond what is provided by other time use studies-such as measures of non-market work. The survey's objectives are to assist researchers in analyzing changes in quality of life, and to reveal changes in work patterns-and in particular, the impact of trends in technology, work outside traditional settings, and how people are combining work and child care. The ATUS will also identify how US time use patterns compare with those of other countries, and will help inform policy debates and business decisions.

Data will be collected throughout the year, and will be reported in the form of quarterly and annual estimates. The first annual estimates are expected in mid 2004. The ATUS will be a CATI survey, but will also include non-telephone households. Because it is a person survey drawn from the CPS household sample, the ATUS will select one person from the household and randomly assign them a day of the week. Persons assigned Tuesday, for example, would be interviewed on Wednesdays. And in contrast to the structured interviews typical of federal surveys, the ATUS interviews will be conversational in style. A diary approach was considered, but tests indicated no data improvement, and response rates actually dropped. The objective is to complete 1,600 interviews per month, and with the first interviews taking place January 1, 2003, Tom quipped that the survey will probably start with a lot of sleeping and other unusual activity.

Tom described as controversial the decision to assume that people do only one thing at a time. Data will not be collected on simultaneous activities, but the possibility will be explored. Other challenges include the recording of activities done during trips of two or more days duration (since the interviews are supposed to be next-day), defining when a child is in a person's care (many parents insist it is 24 hours a day), and defining volunteer activity.

Data will be reported using a three-tier activity code scheme, with each activity having a six-digit code, with two digits per tier. Dissemination plans include tables with estimates that will be made available on paper and on the web, press releases, and data extracts. Also planned are "modules" of information on special topics, such as child care, volunteering, and health. Additional information on the survey can be found on the BLS website www.bls.gov.

Concerns of COPAFS Constituencies

At the conclusion of the meeting, brief suggestions were made that COPAFS look further into the issues related to Section 508 compliance, and the possibility of COPAFS picking up the slack now that the Communications Consortium (TerriAnn Lowenthal) census "alerts" have been discontinued until further notice.

  • Don Muff, Muff Consulting Services
  • Ken Hodges, PAA and Claritas
  • Stephen Tordella, Decision Demographics
  • Linda Jacobsen, Independent Consultant
  • Nicholas Zill, AAPOR
  • Rick Kowalewski, BTS
  • Judie Mopsik, Abt Associates
  • Barna Juhasz, Federal Highway Admin.
  • Lisa Schwartz, NORC
  • Kelvin Pollard, Population Reference Bureau
  • Colleen Flannery, Census Bureau
  • John Cromartie, Assn of American Geographers
  • Seth Grimes, Alta Plana
  • Fred Cavanaugh, Sabre Systems, Inc.
  • Ralph Rector, Heritage Foundation
  • Thomas Brown, I-ASSIST
  • Lee Herring, American Soc. Assn.
  • Amy Welch, GPO
  • Lisa Russell, GPO
  • Dorothy Harshbarger, NAPHSIS
  • Carolee Bush, AAPOR
  • Margaret Martin, COPAFS
  • William Kandel, RSS
  • Patricia Becker, APDU and SEMCC
  • Chuck Waite, NABE
  • Dick Kulka, RTI
  • Jerry Fletcher, AAEA
  • Ed Spar, COPAFS
  • Leslie Scott, ESSI
  • Susan Lapham, ESSI
  • Jennifer Williams, CIS
  • Terri Ann Lowenthal, Consultant
  • Jane Smith, Fenestra
  • Kent Hargesheimer, SRC
  • Mike Cohen, ASA
  • Felice Levine, AERA
  • Miron Straf, NAS
  • Tom Nardone, BLS
  • Tim Triplett, Urban Institute
  • Ken Jones, Urban Institute
  • Kevin Wang, Urban Institute
  • Christa Jones, Census


The Council of Professional Associations on Federal Statistics

20 F Street, NW, Suite 700, Washington, DC 20001 | www.copafs.org

©2010 The Council of Professional Associations on Federal Statistics. All rights reserved.