2016 Causality in Statistics Education Prize Awarded
Online Materials, Nominations Sought for 2017
The American Statistical Association awarded the fourth Causality in Statistics Education Prize to Onyebuchi Arah, professor of epidemiology in the Fielding School of Public Health at the University of California, Los Angeles (UCLA), and Arvid Sjölander, associate professor at the Karolinska Institutet (KI) in Stockholm, Sweden.
Established to “encourage the teaching of basic causal inference methods in introductory statistics courses” in 2013 by a donation by Judea Pearl, recipient of the 2012 Turing Award and professor of computer science and statistics at UCLA, the annual award recognizes the work of an individual or team that enhances the teaching and learning of causal inference in introductory statistics coursework. Funded this year by Microsoft Research and Google, $5,000 was presented to each recipient at the 2016 Joint Statistical Meetings in Chicago.
“While the study and practice of statistics is growing in popularity and demand in both academia and professional occupations, there remains a glaring gap when it comes to causal inference. Even with the recent development of causal inference tools, which are currently sweeping new insights and application areas, most statistics educators and textbooks do not convey any material on these tools,” said Pearl, who is co-chair of the prize-selection committee. “In giving this award, we not only recognize the dynamic efforts of renowned scholars, but also show other researchers and scientists that teaching causal inference can be fun and formative.”
Arah is honored for his graduate-level course titled “Logic, Causation, and Probability,” which embraces the current developments in causal inference using nonexperimental data and equips students with both theory and practical tools. The 10-week course features an introduction to principles of deductive logic; allows for substantial practice in identifying and estimating target quantities using directed acyclic graphs, probability logic, and potential outcomes language; and employs as a teaching tool “hands-on” data analysis exercises.
Sjölander is being recognized for teaching a one-week introductory course for doctoral students in epidemiology on causal inference that covers fundamentals of causal inference, counterfactuals, causal diagrams, confounding, mediation and colliding, and advanced estimation techniques such as inverse probability weighting and marginal structural models. He also conducts a one-day course in causal diagrams for medical doctors at KI’s Research School, where students learn about problems with traditional covariate selection strategies and how causal diagrams can be used to formulate better strategies.
These winning materials and those of previous Causality in Statistics Education Award winners are available online.
The 2015 prize was awarded Tyler VanderWeele, Harvard University professor of epidemiology with joint appointments in the departments of epidemiology and biostatistics. VanderWeele was honored for his innovative book titled Explanation in Causal Inference: Methods for Mediation and Interaction, which provides accessible and comprehensive coverage of causal explanations and science around the fundamental aspects of mediation and interactions. It offers statistics educators a scholarly course material that is teachable, comprehensive, and rigorous. He was presented the award at the 2015 Joint Statistical Meetings in Seattle.
The 2014 prize was awarded to Maya Petersen and Laura B. Balzer for developing a path-blazing course, “Introduction to Causal Inference,” at the University of California, Berkeley. With clear lectures, detailed discussion assignments, and innovative labs and homework using R, Petersen and Balzer have prepared a new generation of scientists, who have acquired the tools of modern causal analysis and are equipped to tackle each step of the causal roadmap. Peterson and Balzer’s course was chosen primarily on the basis of its “teachability” and its appeal to a broad range of statistics-minded disciplines.
The inaugural Causality in Statistics Education award in 2013 was given to Felix Elwert of the department of sociology at the University of Wisconsin-Madison for his innovative two-day course, “Causal Inference with Directed Acyclic Graphs.” He was awarded the prize at the 2013 Joint Statistical Meetings in Montréal, Québec, Canada. Slides covering about eight lecture hours of this short course and accompanying publications are available online.
2016 Prize Committee Members
Maya Petersen, University of California, Berkeley
Dennis Pearl, Penn State University, CAUSE co-chair
Judea Pearl, University of California, Los Angeles, co-chair
Felix Elwert, University of Wisconsin-Madison
Daniel Kaplan, Macalester College
Michael Posner, Villanova University
Tyler VanderWeele, Harvard School of Public Health
Larry Wasserman, Carnegie Mellon University