The central purpose of Statistical Science is to convey the richness, breadth and unity of the field by presenting the full range of contemporary statistical thought at a moderate technical level, accessible to the wide community of practitioners, researchers and students of statistics and probability. It complements the main scientific journals of the IMS (the four series of the Annals) by coordinating, integrating and explicating important current statistical research. In line with this goal, the journal will publish reviews of influential new methodological and theoretical topics, articles describing substantive areas of scientific research with promising statistical applications, interviews with distinguished statisticians and probabilists, and articles on the history of statistics and probability. It may also, exceptionally, publish other material such as evaluations of research papers and books, discussions of classic articles with commentary on their contemporary impact, and articles on educational programmes. A high standard of scholarship is expected, and ideally an intellectually rigorous analysis that helps to set the scene for future work in the area. Statistical Science is not a forum for specialists in a topic, whether applied or theoretical, to communicate their latest findings to other specialists. For further advice, please consult the Editorials by former editors: Morris H. DeGroot (1986), Robert E. Kass (1992), particularly on writing review articles, and George Casella (2002), and the Guidelines on Writing for Statistical Science(1994).

Conversation pieces (often known as ‘interviews’)

The interviews in Statistical Science are one of its most popular features. The personal opinions and experiences given in these interviews add a human touch, while also giving insights into the intellectual history of our field. The ideal interview will

• focus on a subject who has insights into the development of concepts and into the interplay of personalities and circumstances that led to such concepts, an overview of how far we have come, and a perspective of where our field should be heading
• tell the reader things that might not have come out without the interview
• probably not be conducted by someone too close to the subject: a little distance and more independence can help lend perspective.

To summarize, the type of interview most appropriate for Statistical Science is one that focuses on ideas and their development, rather than only upon the details and dates of an individual’s career. For more advice, please see Leon Gleser’s Guidelines for Interviewers. Before approaching the subject, the interviewer is advised to discuss his or her plans with the Editor.