Editor’s Letter—Vol. 27, No. 2
Dear CHANCE Readers,
The American Statistical Association (ASA) was born in Boston, Massachusetts, in 1839. When I was president of the ASA’s Boston Chapter (2003–2006), one of the chapter’s local leaders and ASA historian John McKenzie suggested writing a letter to the ASA’s Board of Directors about making Boston the location for JSM 2014, noting that the ASA will celebrate its 175th birthday that year. The letter needed to be written quickly, as the ASA selects the location for JSM several years in advance (see the ASA website for locations through 2020). I wrote the letter and sent it to the ASA. I’m unsure of the role the letter played in the selection process, but JSM will be in Boston this year. On behalf of my home city, welcome to Boston. We look forward to seeing you.
Staying with the history theme, CHANCE began its life in 1988. Distinguished executive editors of CHANCE have included William F. Eddy/Stephen E. Fienberg (1988–1991), John E. Rolph (1992–1995), George P.H. Styan (1996–1998), Hal S. Stern (1999–2001), Dalene Stangl (2002–2004), Michael Lavine (2005–2007), Michael Larsen (2008–2010), and Sam Behseta (2011–2013). Readers may wish to know that most of the CHANCE issues are available online at Taylor & Francis.
In this issue, we welcome a new columnist, Di Cook, and her column, Visiphilia. Di is a professor of statistics at Iowa State University who specializes in data visualization using interactive graphics. In her first column, she takes a graphical expedition into a statistics grade book and uses videos to describe the graphical displays and their interpretations.
In a timely article with March Madness winding down, Philip Yates and John Trono answer the question, “How Predictable Is the Overall Voting Pattern in the NCAA Men’s Basketball Post Tournament Poll?” Michael Lopez, Adrian Esparza, Michael Lavine, and Jenna Marquard describe a student project with atypical data in “Your Textbook Can’t Help You Here: Applying Traditional Methods to Atypical Data.”
Derek Young explores the James Bond films and evaluates the association between the actors who played James Bond with measures of movie success. He ranks Sean Connery as the best 007 and predicts the success of the next film starring Daniel Craig.
Meanwhile, Jason Mitchell and Li Lian Ong prepare us for the upcoming 2014 World Cup (possibly the largest sporting event in the world), to be held in Brazil, with “Optimism and the Occult Octopus: Favorites Lose, Underdogs Triumph, and Spain Finally Wins the World Cup.”
Sam Behseta and Mine Çetinkaya-Rundel interview Jim Berger of Duke University. Topics include Bayesian statistics, reproducibility, the misuse of statistics, SAMSI, the Journal of Uncertainty Quantification, and the importance of statistics in the future.
In Taking a Chance in the Classroom, Kari Lock Morgan, Mine Çetinkaya-Rundel, and Dalene Stangl discuss how data from a speed dating study that explored initial romantic attraction (these data should hold the attention of most students) can be used in statistics courses of several levels and topics.
In The Big Picture, Nicole Lazar reminds us that we’re living in a world that is collecting more and more data, but more data does not always imply more reliable results. Big Data is making our role as statisticians more important than ever.
Andrew Gelman and his colleague, Phillip Price, discuss ethics in the context of modeling and prediction. They focus on the distinction between the goals of getting the right answer and using an accepted approach that is easy to understand and accepted by the audience. They note that making poor predictions is not unethical, but failing to recognize problems when available data discredit the predictions is.
In his column, Visual Revelations, Howard Wainer illustrates how ignoring missing data can produce distorted results while using an interesting example of SAT scores and college rankings. He then shows us that missing data methodologies can by understood by the general public, despite their complexities.
Finally, Christian Robert reviews the books Statistical Modeling and Computation by Dirk Kroese and Joshua Chen, Machine Learning: A Probabilistic Perspective by Kevin P. Murphy, and Statistics for Spatio-Temporal Data by Noel Cressie and Christopher Wikle.
Once again, I would like to highlight the “Give Them a CHANCE” campaign and encourage people to sponsor promising students. In this issue, I sponsor Jordan Dworkin, a junior at Haverford College majoring in psychology with minors in statistics and computer science. Jordan is completing applications for the Summer Institute in Biostatistics programs after having several statistics courses in college and going through the AP Statistics program in high school. He is inquisitive, energetic, and the son of Bob Dworkin—a renowned pain researcher and someone who greatly appreciates statistics!
Scott Evans