Working on Research and Innovation at Statistics Canada

September 23, 2013

in At Work With Statisticians

By Jean-François Beaumont

A love of hockey and math
When I was a kid, I had a strong interest in hockey and mathematics in that order. But I soon realized that I had better chances of having a successful career with the latter. I wanted to do something where I could apply mathematics to solve real problems. I registered for the Master’s Program in Statistics and I really enjoyed it. I knew then that I would do my career in that discipline and I have never regretted it.

Research is rewarding
A large part of my work is devoted to consultation and activities related to research. I like the consultation work. It ensures that I stay connected to real problems and forces me to maintain knowledge on different topics. Most of my research projects are born from consultations.

In statistics, there is usually no single solution to any problem. Some solutions are “better” than others depending on criteria such as theoretical justification, convenience from an applied point of view, simplicity, interpretability for users of data, etc. The goal of research work in a statistical organization such as Statistics Canada is to find a solution that may not optimize any of these criteria but that achieves a good compromise among them so that the research is useful. Seeing the results of research being implemented is actually quite rewarding.

Making a larger contribution
The contribution of a single statistician to the improvement of the global society is often quite modest and I am certainly not an exception to this rule. It is the joint contribution of many statisticians that can make a significant difference over the years. This can be observed by looking at the evolution of methods used in practice in the last century. That being said, my role at Statistics Canada, like other statisticians, is to ensure that state-of-the-art methods are used and understood in statistical programs of the Agency. My aim is, in part, to develop innovative methods that lead to better quality of the statistical output, and also to develop innovative methods of quantifying that quality. This in turn enables governments and businesses to make more informed decisions, and academics to obtain a better picture of economic or social phenomena.

Beaumont is chief methodologist in the Statistical Research and Innovation Division at Statistics Canada.