Ecology is a science that demands from researchers a decent amount of mathematical thinking and good analytical skills. To be fair, these are must have traits for all of us working in this data-rich era. Despite the obvious mathematical reasoning that comes with studying how organisms and populations thrive, interact and evolve, most ecology graduate programs don’t provide a formal mathematical training for students, thought advanced stats and programming courses are offered in most departments out there. I see this trend as a “lets go straight to what matters” type-of-strategy for learning and teaching analytical methods in ecology graduate programs – which works, but is this the best strategy? I believe the lack of a more traditional training on the basic stuff, such as algebra and probability theory, makes it really hard for early-career ecologists to get their statistics skills developing in a steep learning curve. Fortunately, there are ways to overcome that – and the sooner the better to start going around these limitations through working on improving math and programming skills.
As an ecologist ‘under development’, I believe the first way to get around the limitations in our analytical training is by losing the fear of math: in other words, get the puppy face off and go rough my friend, throw yourself in the mud, and have fun trying to walk on very slippery terrain until you become a pro at doing so. My inspiration for writing this post comes from my recent experience as an ecologist in an environmetrics conference: Graybill/ENVR Conference – Modern Statistical Methods for Ecology. The Graybill Conference is hosted every year by the Department of Statistics of the Colorado State University, and it’s a great opportunity to get to know people that are the actual developers of the statistical approaches we apply in ecology and evolution. Some topics discussed in the conference were hierarchical modeling, occupancy modeling, modeling spatial data, latent variable modeling, and estimating species diversity taking phylogenetics into account. As any other ordinary grad student in Ecology, I also didn’t receive a formal mathematical training, besides undergrad level calculus zillions of years ago. Hence, I definitely wasn’t able to understand most talks as thoroughly and completely as I (probably) would in an ecology-related conference. However, I was indeed able to scoop enough information that will help me to improve my work in progress–and that’s exactly what I was looking for. If you’re a grad student in ecology, and frequently find yourself trying to answer questions that would take advantage of a more advanced statistical approach, keep an eye on environmetrics meetings and workshops, as these might be a handy resource for you.
If this post inspired you, check out these links:
- Phidot is a great source to find out about workshops, conferences and navigate on stats forums: http://www.phidot.org/
- Ben Bolker’s website–his book is great for learning R, plus he’s a very active and helpful person in R forums: http://ms.mcmaster.ca/~bolker/
- Marc Kéry’s book, very useful for learning Bayesian analytical framework: http://www.mbr-pwrc.usgs.gov/software/kerybook/
- Wanna get into coding and find out how R packages are built? Get familiar with GitHub, one of my faves is Hadley Wickham’s: https://github.com/hadley/r-on-github
- Find out what people from the Australian Centre of Excellence for Environmental Decisions do – it’s good stuff! http://ceed.edu.au/
- Read the Dynamic Ecology blog: http://dynamicecology.wordpress.com/
- Finally, visit my former advisor (Dr Gonçalo Ferraz) webpage–Very interesting research, great lab group, and awesome PI (he’s my source for most of these sources above): http://www.ferrazlab.com/LabSite/Welcome.html
I’ll leave you with a remarkable quote from S. J. Gould in the book “The Mismeasure of Man”, which always inspires me to go beyond in my learning process, in an attempt to understand this beautiful thing called nature.
“We naturally favor, and tend to overextend, exciting novelties in vain hope that they may supply general solutions or panaceas–when such contributions really constitute more modest (albeit vital) pieces of a much more complex puzzle.”