An introduction to this blog

So why am I creating this blog?

I’m a lecturer in theEnvironment and Sustainability Institute at the University of Exeter and could describe myself as a spatial ecologist that likes finding practical solutions to conservation problems. Every year I supervise about eight or so students. Over the years I’ve noticed the students encounter the same types of problems when analysing their data. This is partly by design. I tend to give students similar types of project as past experience has taught me that these types of projects work.  This means that I get lots of requests for help, often concentrated at the same time of year. It also means that I end up explaining the same types of issues over and over again.

This blog has been created for several reasons. On a personal note, I hope it will cut down on the number of meetings I have, while at the same time do a better job of explaining the issues and how to get round them. On a more general note, I think many academics can forget that our main legacy is not our research, but the students we teach, inspire and motivate.  Unfortunately, not all of them are equipped with the skills most wanted in the workplace. This blog is designed to go some way to redressing that balance by provided, simple, easy to understand advice for dealing with commonly encountered problems when carrying out statistical analyses or using geographical information systems (GIS).

Student projects have ranged from work on chough (top) to rare plants like this pigmy rush (bottom)

How it works

Each blog post covers a separate issue and is tagged using keywords. I’ve also created a website to facilitate the organisation of the blog (you can see an index of all blog posts, ordered by topic, here. The website also hosts example datasets and R code, referred to in the blog posts, which I’m happy for students to borrow.

What types of analyses is it for?

The types of project I supervise tend to involve collecting presence absence and/or abundance data on a species, often of some conservation concern, and relating their occurrence or abundance to some environmental variables. Basically, the projects have the aim of working out why a species occurs where it does, what it “likes” in terms of habitat (or climate) and what conservation could do to benefit it. Examples include working out the microhabitat preferences of Marsh Fritillaries on the Lizard Peninsula, guiding habitat management for Chough across Cornwall, working out whether habitat management can be used to offset the effects of climate change on some of Britain’s rarest plants and a bunch of other projects, which I aim to host on the website in due course. However, it deals with (or will in due course) some of the commonly encountered problems when working with ecological data: my data have loads of zeros! When should I use a mixed model and how? What is AIC and multi-model inference?