Statistics: The Skeleton Key of Scientific Research
Unemployment is at its lowest level in 30 years. Asthma-related hospital visits soar on smog days.
Headlines like these are familiar to most of us, and they’re often qualified with the phrase “statistics show.” Most of us think of statistics as mysterious rules governing the universe, rules that intrude on our lives only as interesting—and often perplexing—bits of trivia.
Dr. Nancy Reid, the Canada Research Chair in Statistical Theory and Applications, has made a career of statistics and is a leader in the area of statistical inference—drawing conclusions about people or processes from large and complex sets of data.
As modern technology has made collecting data easier, tools for getting the best possible information out of these data are becoming ever more important. At the same time, scientific knowledge is becoming more and more specialized, as scientists focus on finding specific solutions to specific problems. As a statistician, Reid is a sort of jack of all trades, helping solve complex, data-rich problems in areas as diverse as particle physics and air quality.
Reid’s projects involve looking for the things in common between different areas of research and developing principles and techniques for dealing with these commonalities. She will focus on three areas in the coming years.
Building on an interest she developed while serving on review committees for the Health Effects Institute, Reid will look at the statistical methods used to study the health effects of air pollution.
Her collaboration with high-energy physicists, meanwhile, has led to new methods for dealing with the data that will come from large particle physics experiments such as the Large Hadron Collider particle accelerator currently under construction in Switzerland. Finally, Reid also works with the Statistical Methodology Group at Statistics Canada to study complex social-survey data.
Her wide-ranging interests and varied collaborations mean that Reid’s work acts as a skeleton key of sorts, ensuring that scientists and social scientists make the best use of their data—long before they reach the headlines.