McGill et al. (2006) argued that community ecology had lost its way. Shipley (2010) accused community ecologists of acting like a bunch of demented accountants. Strong words – so what’s the issue exactly? And what can we do about it?
Their beef was that when studying groups of species and their environmental association, ecologists often were not thinking enough about the reasons for variation across species. (In this post we’ll focus on variation in abundance or in environmental response of abundance across species. We’re interpreting “abundance” loosely – counts, biomass, 1-0, whatever.) While alternative methods are more readily available nowadays, “accountancy” is still common.
A Beginner’s Guide to Data Exploration and Visualisation with R by Elena N. Ieno and Alain F. Zuur
In 2010 Alain Zuur, Elena Ieno and Chris Elphick published a paper in Methods in Ecology and Evolution entitled ‘A protocol for data exploration to avoid common statistical problems‘ (Volume 1, Issue 1). Little did they know at the time that this paper would become one of the journal’s all-time top downloaded and top cited papers, with a total of 22,472 downloads between 2010 and 2014.
Based on this success they decided to extend the material in the paper into a book.
Zuur and his colleagues at Highland Statistics ltd. give about 25 five-day statistics courses per year. Their typical audience consists of biological scientists at the post-graduate and post-doctoral levels. Early on in each course they have the following conversation with the participants:
Speaker: “Do you review submitted manuscripts for journals?” Audience: “Yes.” Speaker: “Do you like the statistical part of these manuscripts?” Audience: “No!” Speaker: “Do you understand the statistical part?” Audience: “Not always.”
What if there were ways you could make reviewing your paper easier and more enjoyable for reviewers? What if making your manuscript easier to understand and nicer to read would increase the likelihood of your work being published?
In this video Francis Hui introduces boral, a new R package he developed for Bayesian analysis of multivariate data in ecology. It uses Bayesian MCMC estimation to fit latent variable models for unconstrained ordination (read the MEE paper, Model-based approaches to unconstrained ordination, for details), and for multi-species inference while accounting for inter-species correlation: Download boral from CRAN Read the MEE paper Model-based approaches to … Continue reading boral: R package for multivariate data analysis in Ecology
In this video David Warton interviews the organisers of the “Modern Statistical Methods for Ecology” Graybill/ENVR Conference (Sept 7-10 2014, Fort Collins) – Alix Gitelman, Geof Givens, and Janine Illian. They discuss highlights of the conference, current trends in statistical ecology, and where the Graybill conference series (organised at Colorado State every year) is going next. Continue reading Graybill/ENVR 2014 – highlights, current trends and what’s next
David Warton interviews David Borchers, Reader in Statistics at the University of St Andrews, about his latest paper in MEE, “Continuous-time spatially explicit capture-recapture models, with an application to a jaguar camera-trap survey“: Continue reading An interview with David Borchers: Continuous-time SECR
Methods in Ecology and Evolution has been publishing papers on statistical ecology since its inception in 2010. Since the last ISEC meeting, we have published many more papers, of an increasing quality and influence. We have put together a Virtual Issue to showcase some of those papers (but it also misses out many more that will be just as interesting)!. The papers chosen show the … Continue reading Ecological statistics are methods too!
This week David Warton (Methods’ Associate Editor) received the 2014 Christopher Heyde Medal from the Australian Academy of Sciences for contributions in probability theory, statistical methodology and their applications. He gave a talkto the academy, which he’s summarised in this article, originally published onThe Conversation.
Sorry Rick – you should’ve been left behind about three decades ago (along with some algorithms). Claudio Poblete/Flickr, CC BY-NC-ND
By David Warton
It’s an exciting time to be doing statistics. You heard me – statistics: exciting.
It often gets a bad rap, but stats is after all at the business end of the research process. When I’ve collaborated on studies of megafauna, leopard seals, police confessions, a new casino game or climate change effects on biodiversity, the point where researchers find out their results and have those “Eureka!” moments is more often than not in front of a computer rather than out in the field.
Now is an especially good time to be a statistician because the technological revolution over the past couple of decades has blown the field wide open – but despite this, some researchers continue to use outdated and inadequate statistical methods.
The sooner we can change this, the better.
Wikimedia Commons, CC BY
When I started high school (around 25 years ago) computers looked like the one pictured right. They had 64kB memory. And this compressed digital image is more than 100kB, meaning that this poor computer doesn’t have enough memory to look at this picture of itself!
A mathematical tool used by the Metropolitan Police and FBI has been adapted by researchers at Queen Mary University of London to help control outbreaks of malaria, and has the potential to target other infectious diseases.
In cases of serial crime such as murder or rape, police typically have too many suspects to consider, for example, the Yorkshire Ripper investigation in the UK generated a total of 268,000 names. To help prioritise these investigations, police forces around the world use a technique called geographic profiling, which uses the spatial locations of the crimes to make inferences about the criminal’s likely anchor point – usually a home or workplace.
Writing in the journal Methods in Ecology and Evolution, the team has shown how the maths that underpins geographic profiling can be adapted to target the control of infectious diseases, including malaria. Using data from an outbreak in Cairo, the scientists show how the new model could use the addresses of patients with malaria to locate the breeding sites of the mosquitoes that transmit the disease.
“The experts working in the field had to search almost 300 square km to find seven breeding sites, but our model found the same sites after searching just two thirds of this area,” said Dr Steve Le Comber, a senior lecturer at QMUL’s School of Biological and Chemical Sciences.
“In fact our model found five of the seven sites after searching just 10.7 square km. This is potentially important since there is a lot of evidence suggesting that the best way to control outbreaks of malaria is to attack the mosquito breeding sites – but it is incredibly difficult to do in practice.”
The mathematical approach takes just minutes on a computer, meaning that the method could be used in the early stages of epidemics, when control efforts are most likely to be effective – potentially stopping outbreaks before they spread.
This contour map shows the number of cases in Cairo, Egypt. The observed data points are shown as red circles, while the empirically identified sources are shown as blue dots.
Dr Le Comber added: “The model has potential to identify the source of other infectious diseases as well, and we’re now working with public health bodies to develop it further for use with TB, cholera and Legionnaires’ disease.” Continue reading “Criminal profiling technique targets killer diseases”
In the 4th and final installment of Barb Anderson’s INTECOL 2013 podcasts, she asks a number of delegates: What method has transformed your field the most, during your career?
The answers in this podcast are given by the following people:
Steve Hubbell, University of California, Los Angeles, USA (00.21)
Georgina Mace, University College London, UK (00.44)
Carsten Dormann, University of Freiburg, Germany (01.07)
At INTECOL 2013, Methods’ Associate Editor, Barb Anderson, interviewed a number of delegates and asked them: What is the newest method that you currently use?
The answers in this podcast are given by the following people:
Bill Sutherland, University of Cambridge, UK (00.18)
Georgina Mace, University College London, UK (01.04)
