Taking a measured approach

23 May 2018
Dr Beatrix Jones, Senior Lecturer, Department of Statistics
Dr Beatrix Jones

Thirty years ago, scientists with 400 human blood samples might have measured five or 10 things about each one.

Now, with those same 400 samples, they can measure 200, 2000 or even 20,000 things – but that high dimension brings risks, statistically speaking.  While these additional measurements should be informative, there’s a risk that chance differences are misinterpreted unless the statistical methods account for the c.

That’s where Dr Beatrix Jones, who has just started with the Department of Statistics as a senior lecturer, focuses her attention.

“While the number of things we can measure about a sample is ‘exploding’, we need statistical methods to deal with the high dimension,” Beatrix says.

She is particularly interested in metabolomics, the study of the set of metabolites present within a fluid or tissue.

Beatrix grew up in Michigan, US, and initially wanted to study biology – “but it turns out I am really bad at lab work.” So she changed tack. “Statistics seemed like a different way to be involved in answering questions about biology.”

Beatrix completed her MSc and PhD at the University of Washington, where she met Kiwi student, Daniel Walsh. The couple moved to New Zealand in 2004, to jobs at Massey University at Albany. They now have three children.

In her current role Beatrix is teaching Computational Introduction to Statistics (STATS707). Next semester she teaches Applied Multivariate Analysis (STATS302), and is “pretty excited” about that as her research follows on from the techniques covered in that course.

Celebrated statistician John Tukey once said that the best thing about being a statistician was that you get “to play in everyone’s back yard”, and Beatrix has enjoyed that too, although many of her recent papers are in maternal health.

Her latest paper looks at associations between metabolites in a mother’s hair and cognitive ability in her child. Beatrix also crunched the numbers for a paper that explored how New Zealand builders used construction-related smartphone applications to improve productivity.

Ask Beatrix who her statistical role models are, and she’s quick to identify three – Alastair Scott, Peter Guttorp and Jennifer Brown.

“They are all people who do really good work but also have strong collegial relationships and happy families,” she says.

She describes the late Alastair Scott, one of the founders of the Department of Statistics, as “famous for this”. She also cites Peter Guttorp from the University of Washington; and Canterbury’s Jennifer Brown, whom Beatrix says has “proactively shared” strategies to balance work and family.

“I try to emulate people who have successfully navigated some of the same challenges and constraints.”