Joint GSS-SSS Newsletter

March 1996


Social Statistics Section: Chair's Report
Government Statistics Section: Chair's Report

What's the Role of the Federal Statistical System in the Era of Block Grants?
Redefining the Official Poverty Measure
OMB Review of Federal Standards for Data on Race and Ethnicity
HHS Develops Survey Integration Plan
Continuous Measurement System
GSS on the World Wide Web
International Adult Literacy Survey
Bureau of Labor Statistics Wins Ford Foundation/Harvard Award
The International Conference on Computer-Assisted Survey Information Collection
Planning for Chicago in August
Comings and Goings


Social Statistics Section Officers
Government Statistics Section Officers


Social Statistics Section: Chair's Report
Daniel Weinberg, Chief, 
Housing and Household Economic Statistics Division, 
Bureau of the Census 

The best way to keep the membership informed on the many 
activities of the Social Statistics Section in 1995 is to 
summarize the discussions of the Executive Committee Meeting held 
at the 1995 ASA meetings in Orlando. Since this was the only 
meeting of the Executive Committee this year, the topics covered 
provide a comprehensive overview of what the Section has been 
doing, as well as plans for upcoming activities and concerns. 
First, our new officers, for those who haven't yet heard, are 
Miron Straf, National Academy of Sciences, Chair for 1997; Ginny 
DeWolf, Bureau of Labor Statistics, Secretary-Treasurer for 
1996-1997; and Ed Spar, Council of Professional Associations on 
Federal Statistics, Program Chair for 1997. You will be asked to 
vote shortly on new officers for 1998. Continuing officers are 
Dan Weinberg, Census Bureau, Chair for 1996; Sue Miskura, Census 
Bureau, Program Chair for 1996; Dan Kasprsyk, National Center for 
Education Statistics, Council of Sections representative through 
1997; and David Cantor, Westat, Publications Chair for 1996. 
Membership is up very slightly from one year ago. A fellows 
nominating committee has been named. The committee's charge is 
to recommend as many people as necessary for nomination as 
fellows of ASA. The committee will not write the nomination of 
any person that it recommends but will instead find a close 
associate of the individual to do so. The committee chair is Bob 
Groves. (Bob's committee has selected three persons it would 
like to nominate.) 
The first Roger Herriot award was given to Joe Waksberg of 
Westat. The selection committee consisted of representatives of 
the Social Statistics Section, the Government Statistics Section, 
and the Washington Statistical Society. 
The Section contributed $750 to the Council of Sections (COS) to 
help initiate a home page on the Internet for ASA, which did not 
have sufficient funds available to do so. Bob Kominski will 
serve as liaison to the COS committee that is responsible for 
setting up the home page. (The ASA has since agreed to take over 
and fund further development.) 
The cash on hand as of June 30, 1995 was $31,235.63, which 
represents a slight increase from December 31, 1994. This 
increase, however, is running well short of what was projected 
for this period. The major reason for the shortfall to date is 
a sharp reduction in revenue from sales of the Section 
Proceedings. The ASA office reported that sales for all 
Proceedings has been down this year, but could not offer a solid 
explanation for this pattern. The Treasurer (John Czajka) was 
asked to investigate with the ASA office why the sales are down. 
There was also discussion of the merits of lowering the price to 
attract more sales if indeed the sales for the year end up 
substantially lower than expected. 
The COS passed a motion that sections must develop plans to get 
their cash on hand down to no more than twice their annual 
expenditures. The Executive Committee members discussed 
practical ways to do this for the Section. We concluded that the 
surplus ought to go to developing additional continuing education 
and conference opportunities for members. Margo Anderson agreed 
to organize a group to look further at options. A contribution 
of $500 was made to the Clifford Clogg Memorial Fund out of the 
1995 budget. Cliff made great contributions to the Social 
Statistics Section during his too-short life; we will miss him. 
The Program Chair for 1995 (Margo Anderson) reported that the 
Section sponsored 4 invited sessions, 9 poster sessions, 2 
special contributed sessions, and 5 regular contributed sessions 
at the 1995 JSM. A number of other topics came up as part of 
this report, including: (1) the possibility of publishing 
proceedings using electronic access, and (2) methods that might 
assist the program chair coordinate topics with chairs from other 
I hope to see you all at the Joint Statistical Meetings in August 
in Chicago. If you have any concerns, questions, or suggestions, 
I can be reached at (301) 763-8550 (if there are no more 
government furloughs) or at 

Government Statistics Section: Chair's Report
by Bob Lehnen, Indiana University 

During this past year, the Government Statistics Section (GSS) 
Executive Committee has devoted its efforts to three broad, 
overlapping themes: (1) to better serve the GSS membership, (2) 
to reach out to state and local statisticians, and (3) to 
strengthen the financial position of the section. In each of 
these three areas, the section leadership has taken actions to 
improve the services provided to its membership. 
Probably the most innovative step this year is the decision by 
the Executive Committee to participate in the Council of Sections 
Electronic Initiative. This Initiative is designed to make the 
ASA more accessible to its membership via the Internet. GSS is 
contributing monetary support to the Initiative, and more 
importantly, is experimenting with placing the 1995 GSS 
Proceedings on-line. You can read elsewhere in this newsletter 
about the efforts of Al Tupek and other section members to make 
this happen sometime this spring. 
A successful effort to put the Proceedings on-line will allow GSS 
to address two related issues simultaneously--the cost of 
printing proceedings and their accessibility to the membership. 
Printing proceedings is an expensive undertaking for GSS, 
requiring nearly half of the section's budget. Placing the 
proceedings on-line as a service to the membership will make them 
more accessible and also reduce costs. I would like to hear 
your reactions to this project, particularly on the issue of 
whether a paper version is still necessary. 
Another effort to reduce section costs and improve service is the 
production of this joint newsletter. Under the leadership of 
Joan Lee Turek, the joint newsletter format was developed in 
cooperation with the Social Statistics Section. The response to 
the first issue has been overwhelmingly positive. 
Other activities of note are the efforts by section members to 
recruit new faces to section leadership, particular those 
"outside the beltway." If you are interested in working as a 
section officer, or know someone who should be recruited, please 
let me know. The strength of GSS depends on the contributions of 
its members. Also, Council of Section representative Mary Heim 
is preparing a brochure describing the section and its 
activities, and Carolyn Shettle, the 1996 Program Chair, is 
building a strong program. Finally, GSS in cooperation with the 
Social Statistics Section and the Washington Statistical Society 
created the annual Roger Herriott Memorial Award to recognize 
outstanding achievement in government statistics. 
We welcome your participation in any of these activities. It has 
been a pleasure to serve you as section chair for 1995. I 
especially wish to thank the section leadership. Without their 
willing help, little would have been accomplished this past 
year. I look forward to working with the GSS chair for 1996, 
Jerry Gates, to continue serving the GSS membership. 


What's the Role of the Federal Statistical System in the Era of Block Grants? Edward J. Spar, COPAFS The current Congress of the United States is undertaking an experiment in restructuring the Federal-State relationship for domestic social policy. Whether we call it the end of the welfare state, or the dismantling of the responsibility of the federal government to its citizens, decision making regarding who will receive funding for those in poverty will now "devolve" (the latest inside the Beltway jargon) to the states. This will be accomplished through the creation of block grants, whereby over 400 social programs will be consolidated into a handful of grants. This process will affect millions of children and families. This is not an entirely new concept. In 1981, 50 federal programs were consolidated into 9 block grants. Several other block grants have been created since then so that a total of 15 block grant programs with funding of $32 billion are in effect today. These earlier block grant experiences provide important lessons: the most successful grants had provisions to measure and assess impact; the least effective were those for which accountability was missing. Indeed, when there was a lack of accountability, Congress eventually intervened and created more restrictive programs to meet the needs of certain populations, which in turn reduced state flexibility. A report by the Agency for Health Care Policy and Research stated that, "Experience suggests that folding federal programs into block grants can be accomplished with minimal disruption in services when the federal government clearly defines objectives, responsibility and accountability." Unfortunately there seems to be little hope that the current legislation will have requirements for accountability. This includes determining how the states will allocate the dollars as well as outcomes: measures of how well the process worked in reducing poverty, hunger, etc. Except for very few programs, the federal statistical system does not measure social and demographic characteristics at the local level. To be effective for allocating block grant funds, we consider "local" to mean the tract level as the smallest geographical unit. The Census Bureau has produced population estimates for counties and cities since the 1970's. However, they have never produced data by age, sex, and race, much less poverty for small areas. The Bureau of Labor statistics collects data from local areas for measures of employment and unemployment, but they have not linked these measures to demographic or socio-economic characteristics. The Statistics of Income Division of the Internal Revenue Service has developed files from income tax returns, but these are not easily converted to broad measures of the population and households. Indeed, those very populations which would receive funds from block grants probably are the least likely to submit income tax statements. At present, the federal statistical agencies probably don't have a good idea as to what data are feasible to develop at the local level. But let's back up a minute. We have always assumed that there is a role for the federal statistical system. This assumption, of course, is backed up by legislation which requires agencies to produce data for both the allocation of federal dollars and measures of outcomes. Without the data from the Decennial Census long form, the ability to allocate transportation, housing, education dollars based upon legislation would be impossible. Without the Survey of Income and Program Participation, there would be no means of measuring the change in poverty status based upon federal programs. Much of this may change in an era of block grants. First, a number of states have made it clear that they do not want any interference from the federal government. Second, some states may want data, but don't have the infrastructure to make use of it. There are only a handful of states, such as Oregon, which have initiated programs to bring data sources together. The Oregon Shared Information System mandates participation from its Department of Education, Bureau of Labor and Industries, Office of Corrections, and Office of Community College Services. The purpose of the Oregon system is to measure local performance, evaluate outcomes, eliminate duplication of service delivery, improve policy making, and eliminate duplication of data collection efforts. Very few states have reached this degree of sophistication. Therefore, even if the federal statistical system can produce local area data, there is no guarantee that without specific legislation mandating its use, that it will be used. Assuming that there will be many roles for the statistical agencies, how will they meet the needs of the states? To compound the question, how can we expect the agencies to produce more while at the same time the federal budgets are being slashed? For the foreseeable future we can expect either reductions or at best flat federal statistical budgets. These agencies will be expected to do much more with much less. The Bureau of Labor Statistics has already drawn up a list of programs which must either be eliminated or reduced based upon the degree of funding loss. We assume every agency is developing or has already developed an equivalent list. We can expect the agencies to set new priorities. To be of any use to local areas, the agencies will have to concentrate on how to make the most of administrative records ranging from tax returns to Medicare records. In cooperation with other statistical agencies, the Census Bureau should continue its experimental work with a local area continuous measurement survey which would be conducted throughout the next decade. There is so much we do not know about the upcoming era of devolution. How long will it last? How far-reaching will it be? Most important, will the outcome for millions of Americans be positive or negative? For the federal statistical system, this will be a challenge; eventually someone will realize that the efficacy these block grants, especially at the community level, must be measured. The only difference is that, in the beginning, there will probably be no one to tell the federal statistical agencies what to do. They will have to take the lead on their own. Redefining the Official Poverty Measure by Daniel H. Weinberg, Bureau of the Census The National Academy of Sciences' Committee on National Statistics (CNStat) released a report last May entitled Measuring Poverty: A New Approach (C.F. Citro and R.T. Michael, eds.). In that report, the committee recommended that the federal government redefine the way it measures poverty. The Office of Management and Budget (OMB), with the help of the Census Bureau, is about to convene an interagency panel to examine technical methods for doing so. The report recommends changes to the income measure, the poverty thresholds, and the survey used. To change the income measure from the current money income definition, it proposes to add noncash benefits, and to subtract taxes, work expenses, child care expenses, child support paid, and medical out-of-pocket expenses (MOOP). The poverty thresholds are to be based on food, clothing, shelter, and "a little bit more" (75-83% of median expenditures on these items multiplied by 1.15-1.25), a new equivalence scale, an allowance for geographic variation, and annual updates based on growth in median expenditures. Finally, the report recommends that the government use the Survey of Income and Program Participation (SIPP) instead of the March Current Population Survey (CPS). Technical Issues There are several technical issues that must be resolved before implementing a new measure of poverty such as this. These include: -- Estimates of the valuation methodologies for school lunches, food stamps, and housing benefits; -- new estimation methodologies for additional programs -- possible development of a new methodology for valuing Medicare and Medicaid (depending on whether the subtraction of MOOP is adopted or not); -- Development of a methodology for estimating MOOP (e.g. a statistical match of the National Medical Expenditures Survey to SIPP) or reestimation of employer contributions to health insurance using more recent data; -- Completion of development of a tax simulation model for SIPP; -- Estimating work and child care expenses; -- Redesigning the SIPP sampling scheme to maximize reliability of cross-section estimates while maintaining its longitudinal estimation capabilities for program participation, taking account of the need for state-level estimates and minimizing the attrition bias; -- Reviewing the Consumer Expenditure Survey to improve its effectiveness for its new dual role (Consumer Price Index market basket and poverty thresholds) and possibly doing the research for consumption-based rather than income-based poverty estimates in the future; -- Creating a time series of poverty estimates from the SIPP and developing methods to impute additional variables to the CPS to develop comparable time-series data for that survey; -- Doing substantial further work on income underreporting and developing imputation models; -- Adding "child support and alimony paid" questions to the CPS; -- Developing and adding "medical care risk" and possibly medical expenditures questions to SIPP to supplement the poverty measure if medical care costs and benefits are excluded from the measure; and -- Developing poverty estimation models for other surveys that cannot ask the detail that SIPP does. Policy Issues Even if these technical issues can be resolved expeditiously, there are still policy issues that must be debated and resolved before a new measure is adopted. These include: -- Including or excluding medical costs and benefits. The CNStat recommends excluding MOOP, employer contributions to health insurance, and transfer program benefits from income. On the other hand, continuing current (experimental) practice would require revising the current method for valuing medical transfer program benefits ("fungible value" is widely felt to be unsatisfactory but no alternative has arisen) and updating the methodology for imputing employer contributions to health insurance. -- Basing thresholds on a pre-specified fraction of median expenditures. How might the public and Congress react to a new poverty threshold that showed millions more poor persons than the current measure? Are we confident about the quality of (i.e. lack of biases in) the Consumer Expenditure Survey data (even pooling three years)? -- Developing geographical Cost-of-Living variations. The cost-of-living differs substantially from place to place, and different choices of methodology have different implications. -- Annual inflation updating. The panel proposes using the rate of growth in expenditures to index the thresholds. This is an attempt to introduce some deliberate "relativity" into the measure and should be considered against just using the Consumer Price Index. -- Choosing the equivalence scale. The panel recommends an equivalence scale of the form: Number of Adults + 0.7* Number of Children^0.7. Choice of the scale will inevitably alter the distribution of the poor. -- Including cohabiting couples. The decision whether to treat cohabiting couples "as if" they were married for statistical purposes might have political implications, though probably not significant statistical ones. -- Calculating child care expenses. The panel argues that work expenses should be deducted at a flat rate per worker per week worked as there is a substantial choice element involved (e.g. residential location affecting transportation costs). Since the decisions about which child care provider to use (e.g. paid or relatives), whether to choose shift work to avoid the cost, and even whether to have children at all are also choice issues, how should those costs be estimated (i.e. flat rate versus actual)? -- Underreporting. Should the income statistics from the survey be adjusted for underreporting based on administrative data and modeling?; and -- Review and Revision. Does OMB want to set up a cycle of regular review and revision based on pre-specified criteria (CNStat recommends once a decade)? As you can see from this brief summary, these issues are not easy ones to solve. We hope for your help and understanding as we move forward on our research! Daniel Weinberg is Chief, Housing and Household Economic Statistics Division. The views expressed here do not necessarily reflect those of the Census Bureau, the Department of Commerce, or the Office of Management and Budget. Comments can be sent to the author at the Census Bureau (Washington DC 20233-8500) or by e-mail to: OMB Review of Federal Standards for Data on Race and Ethnicity by Suzann Evinger, Office of Management and Budget In 1993, the Office of Management and Budget (OMB) initiated a comprehensive review of the categories the Federal Government has used since 1977 to collect and present data on race and Hispanic origin. These categories, set forth in OMB's Statistical Policy Directive No. 15, have been criticized for not reflecting the increasing diversity of our Nation's population that has resulted primarily from the growth in immigration and interracial marriages. OMB is conducting this review in collaboration with the Interagency Committee for the Review of the Racial and Ethnic Standards, whose members from more than 30 agencies represent the many and diverse Federal needs for racial and ethnic data. This review has also sought public input through a variety of means. Four congressional hearings were held in 1993 focusing particularly on the use of the categories in the 2000 census. OMB subsequently asked the National Academy of Sciences' Committee on National Statistics to organize a workshop that was held in February 1994 to articulate issues to be considered in a review of the categories. OMB's June 9, 1994 Federal Register notice (59 FR 29831-35) seeking public comment resulted in nearly 800 letters and testimony from 94 witnesses at hearings held in July 1994 in Boston, Denver, San Francisco, and Honolulu. In an August 28, 1995 Federal Register notice (60 FR 44674-93), OMB provided an interim report on the review process that included a summary of views expressed thus far and provided a final opportunity for public comment on the research to be conducted during 1996. (The list of additional readings at the end of this article provides further information about the development of the current racial and ethnic categories, the review process, public comments, and plans for research and testing.) An important component of the OMB review process is research to examine possible effects of suggested changes on the quality and comparability of the resulting data. As the first project of the research agenda, the Research Working Group of the Interagency Committee developed a Supplement on Race and Ethnicity for the May 1995 Current Population Survey (CPS), the survey best known for producing monthly data on employment. To gather this information, the 60,000 households were divided into four panels having a random sample of approximately 15,000 households each. Each panel received one of the following four survey instruments: Panel 1 received separate race and Hispanic origin questions with no multiracial category; Panel 2 received separate race and Hispanic origin questions with a multiracial category; Panel 3 received a combined race and Hispanic origin question with no multiracial category; and Panel 4 received a combined race and Hispanic origin question with a multiracial category. The design of the four panels allowed for the collection of data to address several issues under review: (1) the effect of having a multiracial category among the list of races; (2) the effect of adding "Hispanic" to the current list of racial categories; and (3) preferences for alternative names for racial and ethnic categories. On October 26, 1995, the Bureau of Labor Statistics (BLS) issued a news release presenting initial findings from the supplement test; more detailed findings will be issued in early 1996 as a report in the BLS Statistical Working Papers Series. Some highlights of the research findings from this supplement are: -- A higher percentage of persons identified themselves as Hispanic when they were asked a separate Hispanic-origin question that was asked before the race question than when "Hispanic" was included as a racial category. -- The proportion identifying themselves as "White" was higher when the Hispanic origin question was asked separately. -- A little more than 1.5 percent identified as multiracial. A sizable percentage of those identifying as multiracial selected the "something else" category or chose only one race. Hispanics made up a large proportion of the respondents who selected multiracial in panel 2, but not in panel 4, which included Hispanic as a racial category. Some of those who selected only one race, or who indicated "something else," actually went on to specify combinations of two or more races; but a sizable number were Whites who gave both their race and a White ethnic group. When two or more races were selected from the list of races, it was most likely to be a combination including American Indian/Eskimo/Aleut. -- When the multiracial category was included, the proportion of American Indian/Eskimo/Aleut dropped, indicating that it is likely that the American Indian/Alaskan Native population count will be reduced if there is a multiracial category response option. -- The proportions for Blacks and Asian/Pacific Islanders were not affected by either the Hispanic or the multiracial options in the list of racial category choices. -- A substantial majority from each panel preferred the Hispanic racial category option. . -- With respect to preferred terminology, the majority of Hispanic respondents chose "Hispanic"; a majority of Whites chose "White"; a large plurality of Blacks preferred the term "Black," but almost as many chose "African-American" or "Afro-American"; more than half of those identifying as American Indian or Alaska Native preferred either "American Indian" or "Alaska Native," but over a third chose the more generic "Native American"; and almost 30 percent of those identifying as multiracial preferred the term "multiracial," but about as many had no preference. Additional research and testing of issues under review will continue during 1996. As part of the research for the 2000 census, the Bureau of the Census will be testing suggested changes to the racial and ethnic categories in the March 1996 National Content Survey (NCS), which will provide information on how about 90,000 households identify their race and ethnicity in a self-reporting context. Issues being tested include a multiracial category, terminology, and the sequencing of the Hispanic origin question and the race question. In June 1996, the Bureau of the Census plans to conduct a separate Race and Ethnicity Targeted Test (RAETT), which will test and evaluate: (1) adding a "multiracial or biracial" category; (2) a "check more than one" approach to reporting as multiracial; (3) alternative sequencing of the race and Hispanic origin items; (4) combined race, Hispanic origin, and ancestry questions; (5) a combined "Indian (Amer.) or Alaska Native" category; and (6) a "Native Hawaiian" category. The RAETT will target about 90,000 urban and rural households representing American Indians, Alaska Natives, Asians, Pacific Islanders, Hispanics, Blacks, White ethnic groups, and multiracial persons. The design will permit assessments of effects of changes on relatively smaller populations not reliably measured in national samples. In determining whether any changes to the current racial and ethnic categories are to be made, OMB and the Interagency Committee will, of course, take into account evaluations of the results from the May 1995 CPS supplement, the 1996 Racial and Ethnicity Targeted Test, the 1996 National Content Survey as well as other related research. Since the current categories have been useful for a variety of purposes for nearly 20 years, there is also the option to make no changes at all in the categories. OMB plans to complete this review by mid-1997 so that changes, if any, can be implemented in the 2000 census. ADDITIONAL READING Anderson, Margo and Stephen E. Fienberg. "Black, White, and Shades of Gray (and Brown and Yellow)," Chance, Vol. 8, No. 1, Winter 1995, pp. 15-18. Edmonston, Barry, Joshua Goldstein, and Juanita Tamayo Lott, eds. Spotlight on Heterogeneity: An Assessment of the Federal Standards for Race and Ethnicity Classification. Washington, D.C.: National Academy Press, 1995. Evinger, Suzann. "How Shall We Measure Our Nation's Diversity?", Chance, Vol. 8, No. 1, Winter 1995, pp. 7-14. Evinger, Suzann. "Review of the Federal Racial and Ethnic Categories." Proceedings of the Section on Government Statistics, American Statistical Association, 1995. Executive Office of the President. Office of Management and Budget. "Standards for the Classification of Federal Data on Race and Ethnicity," Federal Register, June 9, 1994, Vol. 59, No. 110, pp. 29831-29835. Executive Office of the President. Office of Management and Budget. "Standards for the Classification of Federal Data on Race and Ethnicity," Federal Register, Part VI, August 28, 1995, Vol. 60, No. 166, pp. 44674-44693. National Research Council. Commission on Behavioral and Social Sciences and Education. Committee on National Statistics. Modernizing the U.S. Census. Washington, D.C.: National Academy Press, 1995. See in particular Chapter 7, "Data on Race and Ethnicity," and Appendices D and K. U.S. Congress. House of Representatives. Committee on Post Office and Civil Service. Review of Federal Measurements of Race and Ethnicity. Hearings before the Subcommittee on Census, Statistics and Postal Personnel of the Committee on Post Office and Civil Service, April 14, June 30, July 29, and November 3, 1993. Washington, D.C.: Government Printing Office, 1994. U.S. Department of Labor. Bureau of Labor Statistics. "A CPS Supplement for Testing Methods of Collecting Racial and Ethnic Information, May 1995." News release USDL 95-428 issued October 26, 1995. More detailed findings will be issued in early 1996 as a report in the BLS Statistical Working Papers series. A public use data set is expected to be available in early 1996. HHS Develops Survey Integration Plan by James Scanlon, Department of Health and Human Services The Department of Health and Human Services (HHS) supports a number of health surveys and data systems which form the foundation for the nation's ability to monitor the health of the population and the functioning of the health care system. To respond to critical data gaps and emerging health data needs, HHS has carefully evaluated its health surveys and has developed a survey integration plan. The goals of the plan are to improve the analytical capacity of HHS data surveys and systems, to address high priority data gaps, and to implement a survey framework in which HHS data activities are streamlined and rationalized. The Survey Integration Plan has several major goals, including: to implement a more rational, systematic strategy for collection of data on key health issues facing HHS; to fill a critical gap by producing annual estimates of health care expenditures, insurance coverage, and employer-related insurance costs; to continue to meet public health data needs now met through ongoing population and provider-based surveys within a framework that also allows for expansion to address currently unmet public health data needs; to enhance the analytic capabilities of HHS surveys, allowing multiple HHS data collection efforts to be linked analytically through the use of common core questionnaires, common sampling frames, and common definitions and terms; to reduce the overall burden imposed on HHS survey respondents below those imposed by independent surveys to meet the same data needs; and to achieve efficiencies in sampling, data collection, questionnaire design, and survey operations, allowing more of HHS' resources to be focused on meeting high priority data needs. There are several key elements included in the HHS Survey Integration Plan. These include: redesigning and automating the National Health Interview Survey (NHIS) to serve as the sampling "nucleus" for many HHS population surveys; implementing an ongoing, longitudinal panel survey on insurance and expenditures-- the Medical Expenditure Panel (MEP). The 1996 MEP will be based on the NMES design and will use the 1995 NHIS as a sampling frame. The 1996 MEP would be introduced at a lower sample size than was planned for the 1996 NMES, and then would be increased in size to meet precision requirements as the study continues in 1997; consolidating HHS' surveys of employers so that the National Employer Health Insurance Survey and the insurance follow-up component to the expenditure survey (MEP) are jointly fielded; implementing joint field operations and common core questionnaires for HHS' major surveys of health care providers; redesigning the National Health and Nutrition Examination Survey so that it would be conducted using NHIS as its sampling frame for its next cycle, beginning in 1998; developing a conceptual framework for characterizing health care capacity and the public health infrastructure, and rationalizing and coordinating HHS's provider inventories, health workforce analyses, and public health infrastructure surveys; evaluating the issues involved in possibly coordinating the field operations and sampling of the National Household Survey on Drug Abuse with similar approaches in the NHIS; developing a modular design that will facilitate state-level estimates and provide a mechanism for states to "buy into" national survey efforts to meet their own needs; and implementing improved and more uniform policies regarding privacy, data access, public data release, and sharing among agencies, as well as policies to assure that new survey activities will be conducted through the consolidated framework. Continuous Measurement System by Jay Waite, Bureau of the Census Traditionally, the Census Bureau collects sample detailed socio-economic data once a decade as part of the decennial census. The Continuous Measurement System is a reengineering of the method for collecting this data and for providing it throughout the decade. It blends the strength of small area estimation from the census with the quality and timeliness of continuing surveys. The current plans for Continuous Measurement call for a continuous monthly survey beginning in 1999 called the American Community Survey. Sample Selection The Census Bureau is currently developing a system to build and keep up to date a national Master Address File (MAF). The MAF will be constructed by an on-going computer match of the U.S. Postal Service's Delivery Sequence File with the 1990 Census Address Control File and the TIGER files. Throughout the decade, the USPS Delivery Sequence File will be used to add newly constructed housing units to the MAF. Each month, we will select a systematic sample of approximately 400,000 addresses from the most current MAF for the American Community Survey. The sample will represent the entire United States. Each monthly sample will be selected without replacement from the previous sample over a five-year period. Larger proportions of addresses will be sampled for governmental units (incorporated places, American Indian Reservations, counties, and townships) that have a population of less than 2,500 individuals. This will replicate the over sampling in governmental units used by the 1990 decennial census. The monthly sample size is also designed to replicate the sampling ratio of the 1990 decennial census. Data Collection The American Community Survey will use a self-enumeration mail procedure. The procedure will include several mailing pieces: an advance notice letter; the American Community Survey questionnaire; and a reminder card. A replacement questionnaire will be mailed to addresses in the sample if the original is not completed and returned to the processing office within the prescribed amount of time. For households that do not respond by mail, follow-up will be conducted through either Computer Assisted Telephone Interviewing (CATI). After attempting to obtain responses by mail or CATI, we will use Computer Assisted Personal Interviewing to interview a one-third sample of non-respondents. Data Dissemination Continuous Measurement's goal is to provide the data to the users within six months of collection. For States, populous counties and other governmental units or population groups within a population of 250,000 or more the Continuous Measurement System will provide direct yearly estimates. For less populous governmental units or population groups (those with a population of less than 250,000), estimates will be provided each year through refreshed three or five year data accumulations. Plans for the Continuous Measurement System include the annual release of a microdata file patterned after the 5% Public Use Microdata Sample (PUMS) file of the 1990 decennial census records. This file will contain as many records as possible and will show the lowest level of geography possible within confidentiality constraints. In addition, the Continuous Measurement System will provide summarized data for population and housing estimates, cross-tabulated by various characteristics, down to the block group level. Continuous Measurement and the Federal Statistical System As an on-going survey, the American Community Survey will be a flexible vehicle, capable of quickly adapting to changing customer needs. Questions or specialized supplements can be added to the survey not only to collect new information, but also to help identify special populations or conditions. The American Community Survey can screen for households with specific characteristics. These households could be identified through basic ACS data, or through the use of supplemental questions. Targeted households would then be candidates for follow-up interviews, thus providing a more robust sampling frame for other demographic survey samples or for future follow-up surveys. Currently, there is a trend toward the states and local governments becoming more involved in the administration and evaluation of programs traditionally administered by the Federal government. This is referred to as the devolution of government. Federal programs are moving back to state administration in the form of block grants and through programs initiated by the states. The data collected via the Continuous Measurement System will be useful no only to the Federal Agencies, but also to State, Local, and Tribal governments in the planning, administering, and measuring or evaluating programs. The Continuous Measurement System will provide more timely data for use in small area estimation models. Currently few data are available for less populous governmental units which are consistent throughout the Nation. The Continuous Measurement System will provide data enabling statisticians to model or other national data series down to lower levels of geography such as state, county, city, or metropolitan areas. Announcements GSS on the World Wide Web by Al Tupek, National Science Foundation The Government Statistics Section established its presence on the World Wide Web (WWW) at the end of 1995. The GSS WWW pages provide electronic access to activities sponsored by the GSS as well as links to other WWW sites that may be of interest to its members. The GSS WWW pages can be accessed from the ASA Home Page ( Click on "Sections" then "Government Statistics" to reach the GSS Home Page. The special interests of the Section on Government Statistics are in the production and use of statistics by the governmental bodies at all levels of government -- federal, state, and local -- and in the practice of statistics in the public sector. These interests include concern with statistical policy issues, quality and usefulness of governmental data products, special problems of state and local data, and the role of professional statisticians in public programs. There are WWW pages for each of the topics listed on the GSS Home Page: Charter, Officers, News, Meetings and Training, and Related Sites. From the list of current GSS officers you can send, via e-mail, your ideas for improving the GSS. Electronic versions of several recent newsletters, including this one, are available from the "News" page. The GSS programs for last years' and this years' Joint Statistical meetings are included under "Meetings and Training." Links to abstracts and some of the papers presented at the 1995 Joint Statistical Meetings are under development. Links to training centers that specialize in training of the federal/state/local statistical agency work force are also provided here. "Related Sites" provides links to federal/state/local statistical offices and some other interesting sites. Your help in providing WWW addresses for more state/local statistical offices would be appreciated. Come visit the GSS on the WWW and send me your comments by using the "Feedback" button on the bottom of every page. International Adult Literacy Survey Data from the first International Adult Literacy Survey (IALS), which measures literacy skills in seven developed nations showed that the overall literacy skills of Americans are similar to those of people from the six major industrialized societies included in t he survey. While Sweden is at the top of the participating countries, the Netherlands, Germany, Canada, the United States, and Switzerland performed similarly. Each country did better on some items than others. Poland performed at lower levels than the group included in this survey. However, the distribution of literacy skills within the countries varied considerably. The United States was most similar to Canada in this respect, with more adults at both ends of the literacy scale. In contrast, our European counterparts had more concentrated performance-- the Netherlands, Switzerland (both French and German speaking) and Germany had more of their population achieving in the middle range, while Sweden had the highest concentration at the high end of the scale, and Poland had the highest concentration at the lower end. For further information and copies of the report, Literacy, Economy, and Society, please contact Marilyn Binkley of NCES at (202) 219-2195. Bureau of Labor Statistics Wins Ford Foundation/Harvard Award by Richard Clayton, Bureau of Labor Statistics A 10-year research and implementation project involving the Bureau of Labor Statistics and each of the 50 States and the District of Columbia were awarded a grant from the "Innovations in American Government" award program. This award program, sponsored by the Ford Foundation and administered by the John F. Kennedy School of Government at Harvard University, seeks to identify original, truly innovative and proven successful implementation of new and innovative ways of providing government services. The BLS/State project was one of 30 to reach the award level from over 1400 applications. The award-winning project was the development of a Touchtone Data Entry (TDE) system which allows respondents to government surveys to call a toll-free number and interact directly with a computer. This system was developed in the Current Employment Statistics (CES) program, a monthly employment and payroll survey conducted jointly by BLS and the state employment security agencies. TDE is being used by over 140,000 businesses for the monthly, and the conversion process will eventually cover over 300,000 businesses. A Voice Recognition system is available for respondents without touchtone phones. Under the traditional mail collection method, forms are mailed each month to respondents who fill them out and mail them back to the States for processing, a very time-consuming and expensive process. Under TDE, respondents call a toll-free number and enter their data via the touchtone keypad saving critical response time and avoiding ever-increasing labor and postage costs. This approach is now being used in several statistical programs in the US and in other countries, the IRS is now offering touchtone entry for certain tax forms, and in other information collection applications. Since the award was made public, the BLS has received information requests from several jurisdictions hoping to learn more about how to improve their data collection efforts. Interested parties may contact: Richard Clayton Bureau of Labor Statistics Suite 4860 2 Massachusetts Ave. NE Washington, D.C. 20212 202-606-6520 202-606-6644 (fax) Call for Contributed Papers and Demonstrations The International Conference on Computer-Assisted Survey Information Collection (InterCASIC) is the sixth in the series of American Statistical Association (ASA) conferences on survey methodology. The conference is co-sponsored by AAPOR and IASS, and is scheduled for December 11-14, 1996 in San Antonio, Texas. CASIC is defined broadly to include the general application of computer technology to the entire process of collection, capture, and preparation of survey data. Contributed Papers and Demonstrations are invited. Submissions should be made no later than May 1, 1996 to: Lee Decker, Conference Planner, American Statistical Association, 732 N. Washington St., Alexandria, VA 22314-1943, USA. Please include the name, address, telephone number, fax number, e-mail address, and work affiliation of the author(s) and indicate the primary contact person. Contributed Papers providing insight into current and future CASIC methods are solicited. To have a paper considered, the author(s) must submit a 300-500 word abstract, and a submission fee to the address given above. Demonstrations are similar to contributed papers, but will be set in an interactive forum with the appropriate equipment for the live demonstration of automated systems. Demonstration proposals should follow the format of the contributed papers, including an abstract, with the submission sheet. The abstract should describe the proposed demonstration content and any hardware and software requirements. Submission Fee: a $100 submission fee must accompany abstracts. Purchase orders will not be accepted. Payment must be by check in U.S. dollars on a U.S. bank or by international money order in U.S. dollars payable to the American Statistical Association. For credit card payment, contact Lee Decker of the ASA office. If the abstract is accepted, this fee will be applied to the conference registration fee. Contributors will be notified of acceptance by June 15. If the abstract is not accepted, the submission fee will be returned. For More Information: contact Mick Couper (tel.: 301-405-9523; fax: 301-314-7912; e-mail:; Lee Decker (tel.: 703-684-1221; fax: 703-684-8069; e-mail:, or any other member of the organizing committee. Also, visit our World Wide Web page at: casic.html. Planning for Chicago in August by Carolyn Shettle, 1996 GSS Program Chair Mark your calendar now for the August 4-8, 1996 Joint Statistical Meetings in Chicago! We expect this to be an exciting year for our section-- despite the hurdles placed in our way by the federal furloughs and funding cut-backs. We have two invited sessions lined up: a panel on the effect of downsizing on federal statistics and a panel on improving the Consumer Price Index. Additional GSS sessions are under development on a variety of other topics of interest to our members. These include sessions on electronic dissemination; administrative records; customer surveys; performance measurement; privatization of government statistics; and race and ethnicity classification research. In addition, GSS members will undoubtedly want to attend many of the social and survey statistics sessions that we are co-sponsoring, as well as our luncheon roundtables. Look for information on these and other sessions in the next GSS newsletter Panel on Statistical Programs and Practices of the Bureau of Transportation Statistics The 1991 Intermodal Surface Transportation Efficiency Act of 1991 (ISTEA) established a new Bureau of Transportation Statistics (BTS) within the U.S. Department of Transportation (DOT). ISTEA also mandated a study of the statistical programs and practices of BTS by a panel of the National Academy of Sciences-National Research Council. This panel, the Panel on Statistical Programs and Practices of the BTS, will conduct an 18-month study, and will report on how BTS can best work with other DOT agencies, and the private sector, to improve the quality and usefulness of transportation statistics for private and public sector planning and other uses. Based on its initial review of BTS activities and existing transportation data activities, the panel will select a set of critical issues for more detailed review and analysis. This analysis may focus on data collection, particular DOT or state agencies, or a particular practice, such as dissemination or standard setting. The panel's analysis may illuminate procedural, management, institutional, or statistical policy issues faced by BTS, and may also compare BTS practices with those of statistical agencies in other departments. For more information, contact the project director, Connie Citro, at (202) 334-3096. Comings and Goings Nancy M. Gordon joined the Census Bureau as Associate Director for Demographic Programs on December 10, 1995. Prior to joining the government, Dr. Gordon was Visiting Professor of Economics at the American University. From 1980-1995, she was the Assistant Director for Health and Human Resources of the Congressional Budget Office, a nonpartisan organization charged with providing objective, balanced information to the Congress. Her division was responsible for policy analyses in the areas of health, income security, education, income security, education, employment, and housing. Earlier, Dr. Gordon was the Executive Director of President Carter's Task Force on Women, a Senior Research Associate at the Urban Institute, a Brookings Economic Policy Fellow at the Department of Labor, and an Assistant Professor of Economics at the Graduate School of Business Administration, Carnegie-Mellon University. Dr. Gordon received her Bachelor of Arts degree in Economics and Statistics from the University of California at Berkeley and her Ph.D. in Economics from the Stanford University. She has been active in a variety of professional and volunteer activities, including chairing the American Economic Association's Committee on the Status of Women in the Economics Profession from 1985 to 1987, and serving from 1982 to 1990 as President, Officer, or member of the Board of Directors of Worldwide Assurance for Employees of Public Agencies (WAEPA), a non-profit organization that provides life insurance for federal employees and their families.


Harry A. Scarr, 61, a Census Bureau Deputy Director since 1992, died of pancreatic cancer on November 12, 1995 at his home in Leesburg, Virginia. Dr. Scarr's activities at the Census Bureau included responsibility for the operation of the Bureau, working to cut costs, and to improve the accuracy of the next decennial census in the year 2000. In 1993, in testimony before the House Post Office and Civil Service Subcommittee on Census, Statistics, and Postal Personnel, Dr. Scarr said the next census would be "fundamentally changed". Among the reforms the agency was examining, he said, was a more toward more sampling and away from the traditional attempt to collect information from virtually every home in the United States. Before joining the Census Bureau, he served as a deputy assistant secretary for statistical affairs at the Commerce Department's Economics and Statistics Administration from 1982 to 1992. Prior to that, he served in the Department of Justice from 1973 to 1982, as Director of the Bureau of Justice Statistics, Administrator of the Federal Justice Research Program, Assistant to the Associate Attorney General, (Acting) Director of the National Institute of Law Enforcement and Criminal Justice, Special Assistant for Information and Statistics to the Deputy Commissioner of the Immigration and Naturalization Service, and Director of Research and Statistics for the Immigration and Naturalization Service. Dr. Scarr's published writings include articles on values, cross-cultural research, attitude change, burglary, and criminal behavior. Dr. Scarr attended Baldwin-Wallace College, graduated from the University of Michigan, and received his doctoral degree from Harvard University. Dr. Scarr moved to Washington D.C. in 1963, when he became a staff fellow at the National Institute of Mental Health. Following seven years in the private sector, during which he taught at the Wharton School of the University of Pennsylvania and conducted research for Human Sciences Research, Inc., Dr. Scarr joined the federal government. Dr. Scarr is survived by his wife, Cecilia von Schantz Scarr, and their two children Rachel Scarr and Sarah Scarr, all of Leesburg, and by three children of his first marriage, to Sandra Wood Scarr, Karen Scarr of San Francisco, and Phillip Scarr and Rebecca Scarr, both of Charlottesville. Dr. Scarr was a history buff. One of his interests was the restoration of the George C. Marshall home in Leesburg, Virginia. General George C. Marshall conceived the plan for the reconstruction of Europe after the Second World War. Dr. Scarr's family has established a memorial fund to contribute to the preservation of this home, which will eventually become a museum and education center. Donations for the Harry A. Scarr Memorial Fund may be sent to: The George C. Marshall International Center 20 South King Street Leesburg, VA 22075


SSS Daniel Weinberg, Chair, 1995-1996 Bureau of the Census Housing and Household Economic Statistics Division Iverson Mall, Room 307 Washington D.C. 20233-8500 (301)763-8550 (voice); (301) 763-8412 (fax) e-mail: Margo Anderson, 1995 Program Chair Professor, History and Urban Studies University of Wisconsin--Milwaukee Milwaukee, WI 53201 (414) 229-4361 (voice); (414) 229-6827 (fax) e-mail: Susan Miskura, 1996 Program Chair Assistant Director for Decennial Census, Bureau of Census Rm 3045, FB 3 Washington, D.C. 20233 (301) 457-2933 (voice); (301) 457-3767 (fax) e-mail: John Czajka, 1994-95 Secretary/Treasurer Mathematica Policy Research 600 Maryland Avuenu, S.W. Suite 550 Washington D.C. 20024 (202) 484-9220 (voice); (202) 863-1763 (fax) e-mail: David Cantor, Publications Chair Westat, Inc. 1650 Research Boulevard Rockville, MD 20850 (301) 294-2080 (voice); (301) 294-3928 (fax) e-mail: GSS Bob Lehnen, Chair, 1995 Public & Environmental Affairs, Indiana University 342 North Senate Avenue, 3rd Floor Indianapolis, IN 46204 (317) 261-3041 (voice); (317) 261-3050 (fax) Email: Gerald W. Gates, Chair-Elect Progrwn and Policy Development Office Bureau of the Census, Rm. 2430-3 (DIR) Washington D.C. 20233-1000 (301) 457-2515 (voice); (301) 457-2654 (fax) Virginia de Wolf, 1995 Program Chair Bureau of Labor Statistics, Office of Research and Evaluation PSB 4915 2 Massachusetts Avenue, N.E. Washington, D.C. 20212-0001 (202) 606-7374 (voice); (202) 606-7426 (fax) Carolyn F. Shettle, 1996 GSS Program Chair Division of Science Resources Studies, Suite 965 National Science Foundation 4201 Wilson Boulevard Arlington, VA 22230 (703) 306-1780 (voice); (703) 306-05 1 0 (fax) e-mail: Rich Allen NASS/OA/Room 4117 So. Bldg. 12th & Independence Avenue, S.W. Washington, D.C. 20250-2000 (202) 720-4333 (voice); (202) 720-9013 (fax) e-mail: Joan Lee Turek, Publications Chair Director, Division of Modeling, Computer, and Technical Systems, Office of the Assistant Secretary for Planning and Evaluation, Department of Health and Human Services Rm 438F, HHH Bldg. 200 Independence Avenue, S.W. Washington D.C. 20201 (202) 690-6141 (voice); (202) 401-6228 (fax) e-mail:
Edward J. Spar COPAFS 732 N. Washington St.
Alexandria, VA 22314-1943 (703) 836-0404 (voice) (703) 684-2037 (fax) e-mail: