Washington Statistical Society
        Washington Statistical Society on Meetup

May 2009

Contents:



Announcement of the 2009 WSS Election

The 2009 WSS election will be held online from Monday, April 27 through Friday, May 22. Instructions on voting will be sent to members the last week of April. Below are the candidates for this year's election. The results will be announced at the WSS Annual Dinner in June. Note: The order in which candidates appear on the ballot has been determined by a random process.

All WSS members in good standing as of March 31, 2009 are eligible to vote, as per the WSS Constitution. All are urged to vote once the balloting begins.

Please use this link to cast your ballot.

Candidates for President (select one)

J. Michael Brick, Westat and JPSM

Mike Brick is a Vice President and Director of the Survey Methods Unit at Westat, and is a research professor in the Joint Program in Survey Methodology at the University of Maryland. Prior to joining Westat in 1984, he was a mathematical statistician at three federal government agencies for about 11 years. He has a Ph.D. in statistics from American University, is a Fellow of the American Statistical Association, an elected member of the International Statistical Institute. He has published in a variety of journals on topics including sampling, weighting, and bias evaluation.

Brick has served on the AAPOR's Executive Council (2007-2009), as Associate Editor of Survey Methodology (2001-present) and Public Opinion Quarterly (2003-present), as editor of The Survey Statistician (1995-2000), and as publication editor of Landmark Papers in Survey Statistics (2001). He has also served on several panels or committees, including panels for the National Academy of Science and the American Statistical Association.

Chet Bowie, NORC

Chet Bowie is a Senior Vice President and Director of the Economics, Labor and Population Studies Department at NORC. Prior to joining NORC in 2007, he spent two years at Market Strategies, International. From 1972-2005 he worked at the Census Bureau where he spent the last 10 years of his career there as the Director of the Demographic Surveys Division.

Candidates for Methodology Program Chair (select one)

Jill A. Dever, RTI International

Jill A. Dever is a senior research statistician in RTI International's Washington DC office. She recently rejoined RTI after receiving her Ph.D. from the Joint Program in Survey Methodology under the direction of Richard Valliant. Jill has co-authored and reviewed articles for several survey-oriented journals, and regularly attends WSS-hosted events. Her current areas of interest include survey weight calibration, sample designs for multi-phase studies, various aspects of longitudinal surveys, yoga, and dark chocolate with red chilies.

David R. Judkins, Westat

David R. Judkins is a senior statistician at Westat. Except for a 16 month ̉vacationÓ at RTI, he had been at Westat since late 1986. Prior to that, he worked at the Census Bureau in the old Statistical Methods Division. Until about 2000, his major focus was on sample survey design. Gradually since then, he has re-focused on program evaluation, especially education programs. His major current interests are in semi-parametric procedures for imputation and inference, trying to compromise between design-based and model-based procedures. He was the SRMS program chair for the JSM in Seattle in 2006. He has served as a referee for the Journal of Official Statistics, Survey Methodology, JASA, and Science. He is also an associate editor of Survey Methodology and a past editor of the WSS Newsletter. He has 25 peer-reviewed papers and reports, many conference presentations, and several pieces of invited commentary. He wrote several of the entries for the new SAGE Encyclopedia of Survey Research Methods.

He was made a fellow of the ASA in 2007.

Candidates for WSS Representative-at-Large (select two)

Christine Cox, NCHS

Christine Cox has worked for the Federal Government for 25 years and currently serves as the Director of Record Linkage Activities at the National Center for Health Statistics and the Branch Chief of the Special Projects Branch within the Office of Analysis and Epidemiology, NCHS. She currently serves on a variety of interagency committees including the Federal Committee on Statistical Methodology Confidentiality and Data Access Committee and Administrative Records Subcommittee. She also serves as a statistical disclosure expert on two federal agency Disclosure Review Boards and is a member of the Board of Directors for the Society of Biodemography and Social Biology. Her specific research interests involve imputation and data perturbation, and the analysis of linked survey and administrative data in health and health policy research. She has published both substantive and methodological work using administrative records and sample survey data in government reports and refereed journals. Christine regularly attends WSS seminars and has made numerous beneficial professional contacts through WSS. She is looking forward to the opportunity to become more active in the WSS and serving as a representative at large seems like a good way to start.

Robert Santos, The Urban Institute

Robert (Rob) Santos is Senior Institute Methodologist at The Urban Institute in Washington DC. He has served the public opinion and survey research communities for the past 30 years as a sampling/survey statistician, methodologist, project director, and senior administrator, holding leadership positions in academic, nonprofit and commercial research sectors. Rob is an elected Fellow of the American Statistical Association (ASA) and the 2006 recipient of the ASA Founder's Award for excellence in survey statistics and contributions to the statistical community. He has been a member of ASA since 1977, and has served on numerous elected and appointed positions, including Scientific and Public Affairs Advisory Committee, 2007-2009; Nominations Committee, 2007-2008 (chair, 2008); Program Chair, Social Statistics Section; 2000; Committee on Career Development 1996-2001 (Chair, 1999-01); Committee on Minorities in Statistics, 1981-1994, 1999-2004 (Chair 2003-2004); Committee on Privacy and Confidentiality, 1993-1997; Program Chair, Survey Research Methods Section, 1993 and Social Statistics Section, 2000. Rob has been a member of WSS for ten years.

Mel Kollander, Survey Research Consultant

Mel Kollander is presently a survey research consultant to the federal government and private companies that perform contract research with the federal government. Mel was founder and director of Temple University's Institute for Survey Research Washington office. Mel was the director from 1995 to 2005. Prior to his Temple U. position, Mel served as the Environmental Protection Agency's principal survey methodologist from 1979 to 1995. An important highlight was that he was on the staff of the Administrator, Carol Browner, for almost three years. Another highlight was that Mel served as advisor to the government of Kuwait and the World Health Organization regarding the assessment of the public's exposure to the smoke from the burning oil well fires during the 1991 Gulf War.

Mel has published numerous papers, chapters in several books and presented workshops and has been a guest lecturer on survey design for human exposure field studies and qualitative research methods.

Mel has been a member of the Board of the Washington Statistical Society since 1995. Currently, he is co-chair of the Agriculture and Natural Resources Program. Mel received the President's award in 2006 and was candidate for WSS president in 2007.

Zhiwei Zhang, NORC

Zhiwei Zhang is a Senior Research Scientist at the National Opinion Research Center where he has worked for 12 years. His research expertise and experiences include survey methodology, measurement errors, program evaluations, and statistical methods and applications. He was a recipient of the Clifford C. Clogg Award from the American Sociological Association Method Section in 1996, a visiting Bureau of Justice Statistics (BJS) fellow in 1997, a visiting National Institute on Aging (NIA) fellow in 2000, a Center for Spatially Integrated Social Science (CSISS) fellow in 2001, and an Institute for Mathematics and Its Applications (IMA) fellow in 2003. He has taken various roles such as principal statistician, litigation statistician, analytic task leader, co-investigator, principal investigator, associated project director, senior research scientist and technical advisor in projects for various government agencies including the Appalachian Regional Commission (ARC), Census Bureau, DHHS, DOE, DOI, DOJ, DOL, and the State of Qatar. In addition to co-authoring many published government policy related reports, he has published on Proceedings of the Section on Survey Research Methods (with Michael P. Cohen and Douglas Wright; with Fritz Scheuren; with Dean Gerstein; etc) and research papers with statistical applications in journals such as Addiction, Journal of General Internal Medicine, Journal of Behavioral Health Services and Research, Social Behavior and Personality, Medical Care, and Journal of Health Psychology. Dr. Zhang has been a member of ASA for more than 10 years and has presented at WSS seminars and JSM. He did WSS' community outreach educational and reporting activities as a volunteer and has chaired sampling and analytical algorithm related sessions at national and international statistical conferences.

Return to top

The 17th Federal Forecasters Conference (FFC/2009)

Thursday, September 24, 2009
Washington, DC

The conference theme is "Forecasting and Risk." The conference seeks to highlight how forecasters account for low-probability, but high-cost events. Participants will review how forecasters account for the following risks: economic risk, energy supply risk, food supply risk, health care and epidemic risk, transportation disruptions, and natural disasters. The conference will examine the role of federal forecasters in the evolution of public policies that account for these rare events. For more information, visit www.federalforecasters.org or contact Jeff Busse, 12201 Sunrise Valley Drive, MS988, Reston, VA 20192; (703)648-4914; jbusse@usgs.gov

For more info click here: www.federalforecasters.org

Return to top

Symposium on
Games and Decisions in Reliability and Risk
May 27-28, 2009
The George Washington University

The symposiumon Games and Decisions in Reliability and Risk (GDRR) is being hosted by the Institute for Integrating Statistics in Decision Sciences (I2SDS), in the Department of Decision Sciences and the Institute for Reliability and Risk Analysis (IRRA) in the Statistics Department, the George Washington University, Washington DC. GDRR is partially supported by the Army Research Office. The objective of the symposium is to introduce a new theme, that is, use of game and decision theory in reliability and risk analysis and to bring together researchers from disciplines such as economics, engineering, finance, mathematics, medical sciences, probability and statistics who have a lot to contribute to this theme.

The site of the GDRR is the Duques Hall (2201 G Street, NW) which houses the School of Business at the George Washington University. The symposium will start at 9:00 am on May 27, 2009 and will conclude at 5:00 pm on May 28, 2009. The symposium dinner will take place on the evening of May 27th. Registration to the Symposium is free, but since the number of Symposium participants is limited to 60, participants are recommended to register as soon as possible by sending an e-mail to: i2sds@gwu.edu providing their name, affiliation and address.

A blockof rooms has been reserved at The George Washington University Inn (202-337 6620) and at Double Tree Guest Suites (202-785 2000). You can make room reservations by calling the hotels and identifying yourself as a participant of the I2SDS Symposium. The symposium rate is $185.00 per night for single occupancy plus taxes at The George Washington University Inn and is $209.00 per night for single occupancy plus taxes at the Doubletree Guest Suites. You need a credit card to confirm your reservation. These rates are applicable until April 26, 2009.

The list of invited speakers include

  • Daron Acemoglu, Massachusetts Institute of Technology, USA
  • David Banks, Duke University, USA
  • Erhan Cinlar, Princeton University, USA
  • Arthur Dempster, Harvard University, USA
  • Peter Mueller, M. D. Anderson Cancer Center, USA
  • Suleyman Ozekici, Koc University, Turkey
  • Nicholas Polson, University of Chicago, USA
  • Fabrizio Ruggeri, CNR IMATI, Italy
  • Jayaram Sethuraman, Florida State University, USA
  • Ehsan Soofi, University of Wisconsin-Milwaukee, USA
  • Srinivasa Varadhan, New York University, USA
  • Simon Wilson, Trinity College, Ireland.
Return to top

Workshop on How to Do Nonresponse Bias Analyses
in Household and Establishment Surveys

June 10, 2009

Bureau of Labor Statistics Conference Center
2 Massachusetts Avenue NE, Washington, DC

The Federal Committee on Statistical Methodology (FCSM) is sponsoring a one-day workshop designed to help Federal agencies meet the OMB guidance for conducting nonresponse bias analyses when surveys achieve response rates less than 80 percent. The workshop will feature illustrative examples of the research done by Federal agencies and their contractors to examine nonresponse bias in Federal surveys. The workshop is intended to provide a forum for agencies to learn and share experiences in conducting nonresponse bias analyses, with emphasis on the methodology, and each session will include time for floor discussion. A variety of techniques will be illustrated on different surveys, including Federal surveys of individuals, households, business establishments, farms, and educational institutions.

Research presented will include the examination of potential nonresponse bias by:

  • Using data obtained from the survey frame or panel,
  • Using data from response history,
  • Benchmarking to external data and administrative records,
  • Examining additional information obtained from a subsample of nonrespondents, and
  • Evaluating different nonresponse weighting procedures.

In addition, some research will also be presented that compares two or more of these techniques and approaches to investigating potential nonresponse bias.

The workshop is targeted to:

  • Individuals in Federal agencies who manage and conduct surveys;
  • Federal statisticians, survey methodologists, and others who design, conduct, and evaluate surveys; and
  • Individuals in universities, business and nonprofit organizations who are involved in the development, implementation or evaluation of Federal government surveys.

The workshop will assume a working knowledge of data collection methods in survey research. Examples will be presented and only rudimentary statistical knowledge of concepts such as bias and variance of estimates is required of participants.

Registration fee $95. Attendance is limited. The registration form is available at www.copafs.org/NRBiasReg.htm.

For further information/questions regarding registration, please contact:
Edward Spar or Lee Ann Sklar
Council of Professional Associations on Federal Statistics
2121 Eisenhower Avenue, Suite 200
Alexandria, Virginia 22314
Phone: 703-836-0404 Fax: 703-836-0406
E-mail: COPAFS@aol.com

Return to top

Students' Corner

On the birthday problem

Along withthe delta method and optimization techniques, the ability to create statistical graphs and tables is one of the most powerful tools available to the statistician. From state level maps of the United States showing the unemployment rate to tables of individual statistics for sports teams, statistical graphs, tables, maps, charts, and plots surround us. I was again reminded of the importance and beauty of statistical graphs this past week when I attended what turned out to be an extremely popular workshop on envisioning data taught by Professor Edward Tufte (http://www.edwardtufte.com/tufte/). His coffee-table-sized books are full of super diagrams, tables, and graphs from all time periods telling complex data stories in clear and concise ways.

As a child, I enjoyed exploring maps and learning about the world geography. As a student of statistics, now I am intrigued by data maps. In addition to looking at and analyzing graphs, I often create tables and graphs for homework and class assignments. One of my favorite parts of writing a long report or assignment is using the R package to make graphs to show my results. About a year ago, two of my friends, Jill Dever and Kim Henry taught a class on R to some fellow students in my department. One of the highlights was visiting a website (http://addictedtor.free.fr/graphiques/) dedicated to making graphics in R. I was surprised to learn that the same software package that I use to do simulations and run models could be used to create graphs of bunnies, Chernoff Faces, and county level maps of the United States. The website contains over 150 beautiful graphs along with the R source code for each graph, so you can create similar graphs too.

It is obvious to me that my friends and faculty in the survey methodology department also appreciate graphs. Most offices have cut-outs or print-outs of plots, tables, or graphs with amusing annotations hanging outside them. My colleagues who have defended their dissertation nearly always include tables, graphs, or diagrams in their defense. Most have been very insightful and summarized a massive amount of research which could easily take more than ten power point slides to describe. It is often said that a picture is worth a thousand words; a statistical graphic may very well be worth a googolplex or two.

Tim Kennel (tkennel@survey.umd.edu)

Return to top

SIGSTAT Topics

May 20, 2009: GeoDA - Part 5
(https://www.geoda.uiuc.edu/)

GeoDa is the latest incarnation in a long line of software tools developed by Dr. Luc Anselin's Spatial Analysis Laboratory (SAL) in the Department of Geography at the University of Illinois, Urbana-Champaign. It is designed to implement techniques for exploratory spatial data analysis (ESDA) on lattice data (points and polygons). The free program provides a user friendly and graphical interface to methods of descriptive spatial data analysis, such as spatial autocorrelation statistics, as well as basic spatial regression functionality. The latest version contains several new features such as a cartogram, a refined map movie, parallel coordinate plot, 3D visualization, conditional plots (and maps) and spatial regression.

Continuing the April discussion, this month will cover: Bivariate Spatial Autocorrelation, Regression Basics, Regression Diagnostics, Spatial Lag Model, and Spatial Error Model

SIGSTAT is the Special Interest Group in Statistics for the CPCUG, the Capital PC User Group, and WINFORMS, the Washington Institute for Operations Research Service and Management Science.

All meetings are in Room S3031, 1800 M St, NW from 12:00 to 1:00. Enter the South Tower & take the elevator to the 3rd floor to check in at the guard's desk.

First-time attendees should contact Charlie Hallahan, 202-694-5051, hallahan@ers.usda.gov, and leave their name. Directions to the building & many links of statistical interest can be found at the SIGSTAT website, http://www.cpcug.org/user/sigstat/.

Return to top

Note From The WSS NEWS Editor

Items for publication in the June issue of the WSS NEWS should be submitted no later than May 15, 2009. E-mail items to Michael Feil at michael.feil@usda.gov.

Return to top

Click here to see the WSS Board Listing (pdf)
Return to top