Menu
Log in


The statistical quandaries of an ecologist

  • 16 Mar 2021
  • 6:00 PM - 7:30 PM (UTC+11:00)
  • Zoom

Registration


Registration is closed

The Victorian Branch, together with the Environmental Statistics Section, welcomes you to our first talk of the year, to be given by Dr David Wilkinson.

The statistical quandaries of an ecologist: Applied and theoretical statistics for conservation

At opposite ends of the spectrum, the field of ecology attracts both die-hard field ecologists who prefer to spend most of their time out in the wild collecting data, and hardcore statisticians looking to apply their skills to complex, real life systems. Quantitative ecology is where the two overlap, but communication isn't always easy. In this talk I'll be using aspects of my own work to illuminate some of the statistical quandaries faced by ecologists when they're working with either applied or theoretical statistics.

The 2019-20 bushfires in Australia were catastrophic. While ecologists have been working hard to coordinate our response to the disaster, it isn't all clear sailing. Conservation objectives, abundance or sparsity of data, data types, and working with government departments are all factors that contribute to the statistical decisions we make. How does data availability impact our choice of modelling method? How does the choice of modelling method impact the interpretation of our results? Once it's all done, how does the government make use of it?

In parallel, the world of model development has its own quandaries to face. Joint species distribution models are a relatively new approach that simultaneously models the distributions of multiple species while accounting for residual co-occurrence patterns. How do we ensure new methods aren't just a black box to those without a statistical background? How does a field ecologist try and identify the similarities and dissimilarities between two proposed methods that use different statistical notation? Why do we make it harder by referring to the same concepts with different names? What can we do to make it better?

About David

David Wilkinson is a post-doctoral research fellow in the School of BioSciences at The University of Melbourne. While currently working as the lead data scientist of the repliCATS project for assessing the credibility of published research in the social and behavioural sciences, his background and primary interest is in quantitative ecology applying his skills towards species conservation.

Post-event social

The talk will finish at around 7pm, after which we will stay on Zoom to chat and socialise. We'll set up breakout rooms and you will be able to wander between them to find people to chat with.

Slides and recordings

Video: https://vimeo.com/ssavic/20210316-wilkinson
Audio: https://tinyurl.com/ssavic-20210316-wilkinson
Slides: https://tinyurl.com/ssavic-20210316-slides

Powered by Wild Apricot Membership Software