Home | Working Group | Research | Courses | Software/data | Links | Contact/bio |

Raftery is new JASA-Applications and Case Studies Editor

by Pete Lindeman, ASA

Amstat News, April 1998

Adrian E. Raftery of the University of Washington became JASA Coordinating Editor and Applications and Case Studies Editor on January 1, 1998, taking over from Diane Lambert. Raftery told me he is excited about becoming editor. "JASA as a whole is in great shape," he said. "It is the most circulated and the most cited statistics journal in the world, standards are high, it is superbly produced, the average time to review is down to four months, and great strides have been made in putting JASA-related material on the Web." ASA members are happy with the journal, too; Adrian reminded me of the recent ASA survey in which 85% of the members surveyed said they found JASA to be useful or very useful.

The new editor is a native of Dublin, Ireland, where he obtained B.A. and M.Sc. degrees in Mathematics from Trinity College, followed by a doctorate in mathematical statistics at the University of Paris 6. Then it was back to Trinity, and in 1985, after five years teaching there, he went to work at the University of Washington in Seattle, where he is professor of statistics and sociology. Before taking over at JASA, he finished a term as editor of the journal Sociological Methodology.

"JASA-Applications and Case Studies is important because it is one of the few places where statisticians and quantitative researchers in different subdisiplines and areas of application can communicate and share methods," said Raftery. "It is also easier to read than many other academic statistics outlets, something that I want to maintain. Our readers are the most important part of the JASA community, and I want to be sure that JASA remains useful to them."

Raftery would like to expand the number of Applications and Case Studies articles published in JASA. "Not by lowering standards, though," he said. Instead, he would like to get more submissions. "I see a great deal of excellent applied statistics being done as part of scientific and other projects, but relatively little of it ends up in JASA-A&CS or similar journals. I think there is pressure on statisticians to help publish results in subject matter journals, and they often don't get around to writing up their own statistical innovations for separate publication. I would like to see that change. If you have made a real statistical contribution to an applied project, I encourage you to write it up and submit it to us."

What constitutes a JASA-A&CS article?, I asked. "A JASA-A&CS paper is genuinely motivated by the applied project from which it sprang, and it advances the state of statistics in its subject area," Adrian answered. "It is not a methodological paper with an illustrative example tacked on. Often it will start by describing the scientific, engineering, business or policy question or problem, its background, the debates and controversies that surround it, and why existing statistical methods are inadequate. Then the proposed approach is described, with evidence that it improves over current practice. It is important to say concretely what was learned from the new approach that couldn't have been with existing methods is important."

Raftery's own applied work has been in social research, environmental statistics, and pattern recognition and image analysis. His current social research is on the effects of different family structures on social mobility; a recent study of his with sociologist Tim Biblarz showed the surprising result that children raised by working single mothers do almost as well professionally as those of intact families, and much better than those raised in families headed by their fathers or stepfathers. His environmental research is on bowhead whales, where with colleague Judy Zeh he has developed methods for estimating the number and rate of increase of the population; his work has been the basis of the International Whaling Commission's policy in the area. Before that, he worked on wind energy and on solar energy, and is starting a project on methods for environmental risk assessment.

His pattern recognition work focuses on spatial point processes, with applications in minefield detection, automatic detection of flaws in textile manufacturing, and astronomy. His methodological interests are in applied Bayesian statistics, spatial statistics, time series and model-based cluster analysis.

"Editing JASA is teamwork," said Raftery, "and we have a great team. I'm very lucky to have a superb editorial board, and a wonderful editorial assistant, Janet Wilt. There is an enthusiastic and supportive community of authors and referees. It has been great working with Theory and Methods Editor George Casella, and with the journals production department at ASA, and I'm looking forward to working with Review Editor Marty Wells. Thanks to Diane Lambert, the transition was really smooth. Being editor is a lot of work, but it's also a lot of fun. In my first few weeks I read papers about dinosaur extinction, red deer management in Scotland, the criminal justice system in California, and sexual harassment in the Army. I'm learning a great deal."