March 11, 2005 COPAFS Meeting

COPAFS Chair Sarah Zapolsky started the meeting, and we went to Ed Spar for his Executive Director’s report.

Ed drew our attention to his summary report, “Federal Statistics in the FY 2006 Budget.” The report goes beyond the budget numbers to provide a brief narrative of the statistical agencies, and the programs driving their FY 2006 budget requests. Ed explained that the NCHS numbers look lower than last year’s, but that this traces to a re-organization at CDC. Ed assured that the numbers are not really lower, and reminded us that COPAFS is part of a “friends of NCHS” coalition, which has worked to educate appropriators, and preserve NCHS funding. Ed was less reassuring about the reorganization at the Bureau of Transportation Statistics, where the director position is no longer a presidential appointee. Turning to NCES, Ed noted that former commissioner Bob Lerner has not been reappointed, nor has his name been sent to the Hill for reappointment. So for now, NCES has no commissioner.

Ed also drew our attention to a document summarizing the standing subcommittees of the House Committee on Government Reform—the source of most statistical agency oversight. For example, census oversight is now under the Subcommittee on Federalism and the Census. Ed will keep COPAFS members posted on hearings and other activities of interest.

The remaining 2005 COPAFS quarterly meetings are scheduled for June 10, September 16, and December 9.

National Health Accounts – What Do We Really Spend on Treatments?

Jack Triplett. The Brookings Institution.

Jack noted that the U.S. spends more per capita and per GDP than any other nation on healthcare, but that there is debate over the value of these expenditures. He commented that current data do not adequately answer the questions—how much do we spend, and is it worth it?

The current resources, trace to pioneering work by Dorothy Rice at NCHS. These National Health Accounts have been published for about 40 years now, and started to address healthcare expenditure questions. To illustrate, Jack presented a pie chart, showing the distribution of healthcare spending to the basic sectors – public, private, etc. A second pie chart indicated “who gets the money,” or distribution to recipients such as doctors, hospitals, and nursing homes.

What these National Accounts do not tell us is what diseases the money is spent on. For some purposes, it is important to allocate healthcare spending by disease categories – to identify, for example, how much is spent on treating broken bones. Again citing Dorothy Rice’s early work, Jack described a set of Cost of Disease Accounts that NCHS had been producing every four or five years. These accounts identified spending by disease categories, but had limitations. For example, an “unallocated” category was reserved for categories such as “ambulance service” that do not allocate clearly to specific diseases. The “unallocated” category was reduced over time, but this improvement impaired year-to-year comparability. Further challenges included comparisons with BEA’s national accounts, and efforts to adjust “cost of disease” figures for inflation.

Jack pointed out that the last set of “cost of disease” data was produced in 1997, and that there are no plans for updates. So there remains a large body of work on price indexing by disease, but no more disease-specific spending data.

Returning to his question “is what we spend on healthcare worth it?” Jack commented that we still do not have the data needed to address the question. We have data on how much spending goes to doctors, hospitals, medications, and so on, but we do not know what we are getting from these expenditures. There are many hypotheses about the growth of healthcare costs, but we do not have the data to confirm, the extent to which this growth traces, for example, to the increased use of expensive technology or medications—used in the treatment of specific diseases. Instead, we make assumptions about the source of increasing costs, and the ways to contain them. Jack argued that for a truly modest expenditure, we could have a more complete framework of healthcare expenditure measures that would promote understanding and informed decisions.

Statistical Activities at the Department of Homeland Security.

Michael Hoefer. Department of Homeland Security.

Mike is director of immigration statistics in the Office of Immigration Statistics (OIS) at the Department of Homeland Security (DHS) – home of what we used to know as the Immigration and Naturalization Service (INS). Mike was quick to point out that his division is independent of the enforcement functions at DHS.

The agency has been collecting immigration statistics since 1820, but Mike commented that we still do not have very good information on people entering or leaving the U.S. Current OIS emphases include improving data quality and implementing standards, and their stated mission is to lead the development of statistical information useful to decisions and analyses of the effects of immigration in the U.S. As Mike put it, “we’re not there yet,” but he described the major strategic goals that are driving their work.

The first major goal is to lead the development of relevant customer-oriented statistical information on immigration. In pursuing this goal, OIS will continue estimating the unauthorized population (in contrast to the traditional focus on the documented population), and develop annual estimates of the temporary resident population. They also are exploring new sources of information – such as surveys – as alternatives to adding questions to immigrant applications, and other administrative records sources.

The second goal is to improve the quality and common understanding of immigration data. Activities include the creation of a DHS Immigration Data Council, and work with stakeholders to develop better services, as well as guidelines for data collection, processing and documentation.

The third goal is the timely dissemination of high quality user-friendly statistical information to customers and stakeholders. Actions take the form of published OIS analyses and reports, a revamping and promotion of the OIS website, and a broadening and deepening of communications with customers and stakeholders.

Mike described the growing interest in immigration statistics (in part tracing to efforts to improve the Census Bureau’s population estimates), and increases in OIS funding and staffing. He also described US-VISIT – a system being developed in response to September 11 that seeks to identify individuals (citizens and non-citizens) entering and leaving the U.S. Mike described the system as “a long way off,” but one whose incremental implementation could take the form of electronic passports, and other measures to provide “comprehensive flow numbers,” and improved data on length of stay.

Asked about perception issues related to the association of such data with DHS, Mike stressed the independence of OIS from enforcement functions, but acknowledged that it is something they are aware of. And in follow up discussion of the technology that might be involved with US-VISIT, Mike mentioned that there has been talk of “forms with sensors,” that would monitor entry and exit.

Mike closed with a listing of OIS points of contact.

The Federal Data Agenda of the Brookings Urban Market Initiative.

Andrew Reamer. The Brookings Institution.

Andrew described Brookings’ Urban Market Initiative (UMI) as a three year, $5.2 million effort supported by Living Cities – a collaboration of nonprofit, private and public sector investors committed to the revitalization of America’s urban centers.

The UMI mission is to improve the quality and use of data available for urban areas, thereby improving the quality of investment decisions, and increasing the competitiveness of urban markets as places for investment.

A basic UMI assumption is that data can spur action. Andrew noted that, historically, money and regulation have driven urban policy, but that these are costly and hard to fine tune. In contrast, information as a policy tool has the advantage of 1) being effective and easy to fine tune, and 2) being low cost, and offering high return on investment. Major UMI strategies include supporting increased availability of data, and taking a leadership role in creating a National Infrastructure for Community Statistics (NICS). The idea is that investment in information tools can facilitate urban investment.

UMI activities include scanning and evaluating current statistical offerings, supporting funding for the American Community Survey as well as Census Bureau microdata access tools, actively monitoring information collection requests, promoting an organized and informed constituency for federal data, and the development of NICS.

NICS is described as a nationwide web-based broker facilitating access to community level statistics from thousands of local, state, federal and commercial data sources – comprising a “loosely coupled confederation” of data providers, data intermediaries, data collectors, and data tool providers. While not providing direct links to these sources (Andrew described it as links to links), NICS would encourage the development of basic metadata standards, and other conditions that would define data sources as “NICS-ready.”

A successful NICS is seen as promoting enhanced data access, and interpretation of nationwide local data, which in turn would enhance investment decision-making. Major steps in NICS development include concept development, understanding the needs of users in different sectors, creating a business plan, and implementation. An immediate priority is the identification of “use cases” that illustrate the potential of NICS. Brookings already has held several meetings – building a “community of practice” among potential NICS participants, and an April meeting will be devoted to the business plan and proposed criteria for “use cases.”

Because of the NICS focus on data usability, an attendee asked about data quality versus usability. Andrew acknowledged that this question comes up all the time, and noted that there would be some screening. However, he explained that the NICS priority would be to make users aware of data quality limitations, rather ensuring that a level of quality is maintained by all NICS participants.

Reaching Out to Users of the American Community Survey

Nancy Torrieri. U.S. Census Bureau.

After relaying Jay Waite’s regrets for being unable to attend, Nancy opened by proclaiming that “the census has begun, and we do call it that.” The message is that the Census Bureau regards the initiation of full ACS implementation as the initiation of a major component of the 2010 census. She gave us a brief update on the move to the full sample, and noted that they are pleased with the response rates achieved so far.

But Nancy’s topic was stakeholder outreach, and efforts to head off misunderstandings about the ACS. She noted that work with stakeholders has been going on for a long time, and involves Congress, professional associations, as well as federal, state, local and tribal governments. The effort has gone through three stages.

The first stage involved the generation of interest in the ACS. Running from 1995 to 1999, this stage included symposia for data users, outreach meetings in the ACS test sites, ACS test data on CD-ROM, and the identification of early ACS applications.

The second stage involved the building of relationships, and ran from 2000 to 2003. Nancy cited work in the Multnomah County, OR test site—and in particular, the demographers at Portland State University—as a partnership example.

The third stage, providing support, runs from 2004 through 2012. Nancy described a booklet the Census Bureau has prepared that describes the ACS, and solicits support from local officials and others. She also described how the Census Bureau had solicited comments on ACS data products, but that few users—and only one federal agency—responded to the Federal Register notice. Users will have another chance, however, as ACS products released during 2005 will provide the opportunity for users to provide comments online. The Census Bureau will send an e-mail alert about this opportunity to its ACS mailing list, and Ed promised to relay the alert to COPAFS representatives.

With the move to the full sample, Nancy noted that this is a period of heightened activity, and that the Census Bureau is hearing from Congress, communities, respondents, and others. In response, they are holding workshops to answer detailed questions on the ACS, Census staff are working with their subject matter counterparts at the local level, and there are plans to increase the training opportunities for data users. The Census Bureau also is preparing an ACS data user guide, and hopes to have a preliminary version available by the end of this summer.

Nancy closed with a plug for the ACS website, and acknowledged a comment that the ACS needs to provide data products equivalent to the decennial census summary files.

Concerns of COPAFS Constituencies

No concerns were raised, and the meeting was adjourned.

  • Patricia Becker, APDU/SEMCC
  • Ken Hodges, PAA/ Claritas
  • Linda Jacobsen, PRB/ APDU
  • Merrile Sing, AHRQ
  • Felice Levine, AERA
  • Dorothy Harshbarger, NAPHSIS
  • Norman Bradburn, NORC
  • Stephen Tordella, Decision Demographics
  • Charles Hulten, AEA
  • Joe Garrett, Market Strategies
  • Seth Grimes, Alta Plana Corp.
  • Ed Christopher, Federal Highway Admin.
  • Dick Kulka, RTI International
  • Robert Hoslach, Claritas
  • Fred Cavanaugh, Sabre Systems, Inc.
  • William Kandel, ERS-USDA/RSS
  • Paul Zelus, AUBER
  • Tom Witt, AUBER
  • Nancy Bates, AAPOR
  • Maxime Bokossa, Synectics
  • Tracy Lesseter, COSSA
  • Dan Estersohn, Arbitron
  • Wendy Alvey, American Statistical Assn.
  • Susan Doolittle, NABE
  • Carolee Bush, AAPOR
  • Henry Brownstein, Abt Associates
  • Judie Mopsik, Abt Associates
  • Miron Straf, National Academies
  • Margaret Martin, COPAFS
  • Dan Levine, Westat
  • Don Muff, Muff Consulting Services
  • Sarah Zapolsky, AARP
  • Maurine Haver, NABE
  • Jerry Fletcher, AAEA
  • Randy Childs, AUBER
  • Ralph Rector, Heritage Foundation
  • Rick Ayers, ESRI
  • John Thompson, NORC
  • Colleen Flannery, Census
  • Christa Jones, Census
  • Andrew Reamer, Brookings
  • Mary Moien, NCHS

The Council of Professional Associations on Federal Statistics

20 F Street, NW, Suite 700, Washington, DC 20001 |

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