Washington Statistical Society
        Washington Statistical Society on Meetup

December 2007

Contents:



Nominations Sought for 2008 Julius Shiskin Award

Nominations are invited for the annual Julius Shiskin Memorial Award for Economic Statistics. The Award is given in recognition of unusually original and important contributions in the development of economic statistics or in the use of statistics in interpreting the economy. Contributions are recognized for statistical research, development of statistical tools, application of information technology techniques, use of economic statistical programs, management of statistical programs, or developing public understanding of measurement issues. The Award was established in 1980 by the Washington Statistical Society (WSS) and is now cosponsored by the WSS, the National Association for Business Economics, and the Business and Economics Statistics Section of the American Statistical Association (ASA). The 2007 award recipient was Arthur Kennickell, Senior Economist and Head of the Microeconomic Surveys Unit at the Federal Reserve Board, for his leadership of the Federal Reserve's Survey of Consumer Finances and his achievements as an international expert on the design and implementation of household economic surveys.

Because the program was initiated many years ago, it is little wonder that statisticians and economists often ask, "Who was Julius Shiskin?" At the time of his death in 1978, "Julie" was the Commissioner of the Bureau of Labor Statistics (BLS) and earlier served as the Chief Statistician at the Office of Management and Budget (OMB), and the Chief Economic Statistician and Assistant Director of the Census Bureau. Throughout his career, he was known as an innovator. At Census he was instrumental in developing an electronic computer method for seasonal adjustment. In 1961, he published Signals of Recession and Recovery, which laid the groundwork for the calculation of monthly economic indicators, and he developed the monthly Census report Business Conditions Digest to disseminate them to the public. In 1969, he was appointed Chief Statistician at OMB where he developed the policies and procedures that govern the release of key economic indicators (Statistical Policy Directive Number 3), and originated a Social Indicators report. In 1973, he was selected to head BLS where he was instrumental in preserving the integrity and independence of the BLS labor force data and directed the most comprehensive revision in the history of the Consumer Price Index (CPI), which included a new CPI for all urban consumers.

Nominations for the 2008 award are now being accepted. Individuals or groups in the public or private sector from any country can be nominated. The award will be presented with an honorarium of $750 plus additional recognition from the sponsors. A nomination form and a list of all previous recipients are available on the ASA Website at www.amstat.org/sections/bus_econ/shiskin.html or by writing to the Julius Shiskin Award Committee, Attn: Monica Clark, American Statistical Association, 732 North Washington Street, Alexandria, VA 22314-1943. Completed nominations must be received by April 1, 2008. For further information contact Steven Paben, Julius Shiskin Award Committee Secretary, at paben.steven@bls.gov.

Return to top

Capital Science 2008

The Washington Academy of Sciences and its Affiliates Present Capital Science 2008 to be held March 29-30, 2008.

On Saturday and Sunday, March 29-30, 2008, The Washington Academy of Sciences and its Affiliated Societies (including WSS) will hold the third in the series of biennial pan-Affiliate Conferences, Capital Science 2008. It will be held in the Conference Facility of the National Science Foundation in Arlington, VA at the Ballston Metro stop. With about 20 of the Affiliates participating, the Conference will serve as an umbrella for scientific presentations, seminars, tutorials, and talks. These pan-Affiliate Conferences underline the fact that the Washington, DC area is not only the political capital of the country but, in many respects, the nation's intellectual capital -- with several major universities and government laboratories that are the homes of an astonishing number of Nobel laureates.

Keep checking the Web site http://www.washacadsci.org/capsci08/Index.htm for more information as it becomes available.

Return to top

Washington Academy of Sciences "STARS" Program

Global competition is creating an urgent need to motivate and attract some of the best and brightest students into science and technology careers. In the U.S. Congress, K-12 education is receiving renewed bipartisan support in both the House and Senate.

The Washington Academy of Sciences (W.A.S.) (of which the WSS is an affiliate member) is expanding its "STARS" (Science and Technology Aptitude Recognition for Schools) youth-in-science outreach program.

In recent years, the program has included not only active participation in several Senior High School Science Fairs, but also science events at Elementary and Middle Schools. The AAAS and the IEEE have supported our efforts by grants to the "STARS" program.

The W.A.S. is planning 9 major events in the 2008 school year. They will offer Challenge Cups, cash prizes, plaques, student memberships in Scientific Societies, certificates and other awards. Award winning student projects will also be listed in the W.A.S. Journal and Web site. Outstanding school teachers will also be recognized.

You will not need any special preparation, and the Washington Academy of Sciences will provide everything needed on site at the school.

Typically this will involve about a 4-hour commitment at a participating school (in the DC, MD or Northern VA area) sometime in the January-March 2008 time frame.

The number of schools and students benefiting from the "STARS" Program depends directly on your commitment to participate.

Note: This is a separate activity from the WSS Science Fair Judging Program.

If you are interested, please e-mail a one-liner by December 15th 2007, saying: "I would like to be included in the 2008 W.A.S. Judges' Roster" to Paul Hazan, WAS Vice President for Junior Academy Affairs (pmhazan@comcast.net / Tel: (301) 603-0536) and include your name, e-mail, affiliation, address, and telephone number.

You will be notified as soon as the schools give us their event dates.

Return to top

Seminar on Survey Respondent Incentives: Research and Practice

March 10, 2008
L'Enfant Plaza Hotel
Washington, DC

Hosted by the Council of Professional Associations on Federal Statistics

Incentive payments to survey respondents have been used extensively for many years to improve response rates. Considerable research evidence supports the value of monetary incentives to increase cooperation and improve the speed and quality of response in a broad range of data collection efforts. In 1992, a Symposium on Providing Incentives to Survey Respondents, hosted by the Council of Professional Associations on Federal Statistics (COPAFS), brought together a broad spectrum of survey research professionals from government, business, academia, and research organizations to focus on these issues in detail to review the "state-of-the art." (To read the report go to: www.members.aol.com/copafs/incentives.htm). Since that time, the use of respondent incentives in survey practice has increased dramatically across all sectors. Yet there has been no professional forum since the 1992 symposium that has sought to bring survey professionals together with a specific focus on these issues. The purpose of the proposed seminar is to fill that void.

The first session of the seminar will describe current practices in the use of respondent incentives across the three major domains of surveys: 1) surveys sponsored by federal agencies; 2) surveys conducted by academic investigators, including those funded by federal or other grants; and 3) surveys sponsored and conducted by private sector organizations and commercial establishments.

Session two will be a panel discussion bringing together survey research professionals who have conducted and are knowledgeable of major research, practices and trends on the use of respondent incentives in government, academic, commercial surveys. The theme of this session is: who, what, where, when, why and how do we pay? The panelists will focus in detail on what we are doing and what we know about the use of respondent incentives.

The concluding session will bring together a panel of survey researchers who have had considerable experience in the design and implementation of sample surveys. They will attempt to consolidate and synthesize the seminar discussion, identify common elements/themes, and suggest future directions for implementation and research.

Seminar Registration: $125.00. For a copy of the program and registration information contact COPAFS at copafs@aol.com or call COPAFS at 703-836-0404 and ask for Edward Spar or Lee Ann Sklar. The program and registration form are also available at the COPAFS site at: www.copafs.org

Return to top

George Washington University
Course Offerings Spring 2008

The Statistics Department at the George Washington University will offer the following graduate, and special topics undergraduate courses during Spring 2008 (January 14 - May 14, 2008) at the main campus.

Enhance your statistical analysis skills by taking one or more of these courses. Registering as a non-degree student is easy - please visit www.gwu.edu/~regweb/ for pertinent information.

For questions or further information please contact Dr. Reza Modarres, e-mail: Reza@gwu.edu, ph: 202-994-6888.

Description of courses at www.gwu.edu/~bulletin/ugrad/stat.html

List of available courses at my.gwu.edu/mod/pws/courses.cfm?campId=1&termId=200801&subjId=STAT

Return to top

Georgetown University
Department of Biostatistics, Bioinformatics and Biomathematics

Are you interested in genetics, bioterrorism, international health, bioinformatics, epidemiology or health policy? Get the analytic tools to meet the demands of the 21st century!

You are invited to apply for the Masters of Science Program in Biostatistics with tracks in Bioinformatics and Epidemiology.

Graduates with an MS degree in Biostatistics go on to successful and lucrative positions in academic centers, pharmaceutical companies, biotech companies and private consulting firms.

For more information, visit: http://dbbb.georgetown.edu or e-mail Caroline at ctw26@georgetown.edu.

Apply now for Fall 2008 at: http://grad.georgetown.edu/pages/apply_online.cfm

Return to top

Students' Corner

A friend of mine once observed that if all scientists published using a p-value of 0.05 as their significance threshold, then we could take a walk through the stacks in the science library, and every 20th paper we passed, on average, would be based on a false positive. (In retrospect, my friend's conjecture seems to be based on an erroneous interpretation of the p-value, but you get the idea.) We got a grim chuckle out of the thought and moved on.

Perhaps weshould not have dismissed the thought so easily. This past September, an article appeared in the Wall Street Journal (WSJ) online, entitled "Most Science Studies Appear to Be Tainted By Sloppy Analysis."

http://online.wsj.com/public/article_print/SB118972683557627104.html

It discusses the work of Dr. John Ioannidis, an epidemiologist with joint appointments at the University of Ioannina School of Medicine, in Greece and Tufts University in Medford, Massachusetts. His unsettling thesis is that, for various reasons, most published scientific findings are wrong.

The article drew a swift response from Dr. David Lide, the editor in chief of the CRC Handbook of Chemistry and Physics (who happens to be writing from Gaithersburg, Maryland!). Dr. Lide declares that while he could not speak to the medical research literature as a physical scientist, as far as he could tell there wasn't a problem of sloppy analyses in the physics and chemistry literature.

One statement in the WSJ online article with which Dr. Lide took particular issue was: "No one actually knows how many incorrect research reports remain unchallenged." Dr. Lide writes, "Such statements convey no factual information and plant unfair suspicion on the vast majority who abide by professional standards of integrity." I think he is right; this statement is what writer Stephen Campbell calls an "Unknowable Statistic," in his book "Flaws and Fallacies in Statistical Thinking."

The WSJ online article mentions that Dr. Ioannidis published a paper in 2005 that sported the provocative title "Why Most Research Findings are False." The title says it all. Here is a link to the paper: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=118232 7. I encourage my fellow students to read it, and see what you think. (The WSJ online article also mentions that Dr. Ioannidis has a paper published in the August 2007 issue of JAMA; I will need to look that paper up.)

In the 2005 paper, Dr. Ioannidis gives six "corollaries." Disconcertingly, a majority of the six could very well apply to my own field of interest, functional mapping of the human brain. As an interesting exercise, go through the six corollaries and see how many are applicable to the scientific areas that most interest you.

Dr. Ioannidis points out that negative findings are important, too, but that they tend to be neglected. As an aside, I am reminded of the "File Drawer" problem, whereby the preference to publish positive findings results in a positive bias in the literature, while many of the negative findings are squirreled away into desks and cabinets, languishing unpublished beyond the light of day. Interestingly, an online journal has been established to help remedy this problem, providing a place where scientists can publish null findings (at least, in the social sciences). It is called the Journal of Spurious Correlations (no, not the Journal of Irreproducible Results!); here is the link: http://www.jspurc.org/ Perhaps if you find yourself with a null finding, you can publish it in this journal!

Dr. Ioannidis' 2005 paper has of course received its share of criticism. Perhaps the most serious objection comes from Drs. Steven Goodman and Sander Greenland, from Johns Hopkins University. I won't go into detail here regarding their arguments, but you can find their letter here: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1855693 (The letter references a more extensive paper: http://www.bepress.com/jhubiostat/paper135)

And Dr. Ioannidis' rebuttal is here: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1896210

Tolkien once wrote a warning not to meddle in the affairs of wizards, and in keeping with that warning I am hesitant to come out on one side or the other in this debate. Some very prominent names are involved, and I am but a mere student, a "sorceror's apprentice." Still, I think it's safe to say that Dr. Ioannidis raises important issues for our consideration, and that the debate is of interest to my fellow students. Whether or not we agree with Dr. Ioannidis' argument, we can all agree that as students of statistics, we owe it to ourselves -- and to the people who will depend upon us for proper data analysis -- to always strive for statistical rigor.

That's all for this month. If you have any feedback on this column or ideas for future topics, please email me at jmm97@georgetown.edu. As always, your thoughts will be greatly appreciated.

Joe Maisog
Georgetown University / Medical Numerics

References:

Campbell SK, "Flaws and Fallacies in Statistical Thinking," Mineola, NY: Dover Publications, 2004. See chapter 4. Available online here: http://books.google.com/books?id=3q-GAsILFWoC

Goodman S and Greenland, Why most published research findings are false: Problems in the Analysis, letter to the editor, PLoS Med 4(4):773.

Hotz, RL, "Most Science Studies Appear to Be Tainted By Sloppy Analysis," Wall Street Journal (WSJ) online, September 14, 2007.

Ioannidis JP, Why most published research findings are false, PLoS Med 2(8):696-701.

LideDR, "Unfair Suspicion Cast On Scientific Research," letter to the editor, Wall Street Journal(WSJ) online, September 20, 2007.

The Tolkien quote is from "The Fellowship of the Ring," book 1, chapter 3. The full warning is "Do not meddle in the affairs of wizards, for they are subtle and quick to anger."


A minorcorrection for November's column: I have mispelled my friends name! :-( Instead of "Steven J. Fromm," it should be "Stephen J. Fromm." greatly appreciated.

Return to top

SIGSTAT Topics for Fall 2007

December 19, 2007: Survival Models in SAS: PROC LIFEREG - Part 1

Continuing the series of talks based on the book "Survival Analysis Using the SAS System: A Practical Guide" by Paul Allison, in November we'll start Chapter 4: Estimating Parametric Regression Models with PROC LIFEREG. Topics discussed are:

  1. the Accelerated Failure Time model
  2. alternative distributions
  3. categorical variables and the CLASS statement
  4. maximum likelihood estimation
  5. hypothesis tests

See http://www.sas.com/apps/pubscat/bookdetails.jsp?pc=55233 for more information.

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 January issue of the WSS NEWS should be submitted no later than December 10, 2007. 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