Guest post from the SIG training reps: Susan Jarvis and Duncan Proctor.
In the second post covering responses to our survey we are in the main covering computational issues, though there is often overlap with statistical methods when applied to ecology. Below we have detailed what helpful information we could find covering R (including package creation), GIS, Python, INLA and MARK.
There are a number of places that run courses on R:
There is also the R tutorial package swirl, which teaches you R inside the R interface.
Several times in response to our survey people were looking specifically for data management in R. To be honest it is difficult to know what to recommend here. A quick search will find a huge amount of options for data management in R. Some potentially useful packages would include reshape2, dplyr and tidyr that come with this rather useful cheatsheet. Hadley Wickham has a paper promoting the organisation of data into a tidy form. Also one of our members Dylan Childs has provided the materials for an exploratory data analysis with R course online, which covers a certain amount of data wrangling.
Creating R packages
The first port of call when looking for advice on creating r packages would have to be the book R packages by Hadley Wickham, but there are also a number of guides that can be found online here, here, here, and a short but very clear video example here.
There are a huge amount of resources available for GIS, here are a few but dig yourself and you will find many more
ArcGIS – ESRI have a pretty hefty catalog of training available.
Open source GIS platforms:
OSGeo have compiled a list of open source geospatial training
QGIS has a pretty extensive training manual available
GRASS has many tutorials
If you are happier in R you can also do a lot of spatial data management tasks there, often faster than in a GIS once you know what you are doing
A range of resources are available online e.g. pythonforbiologists.
Have a look within your institution too; I regularly get emails on python short courses run in house.
Searching ESRI’s training catalogue will also find you training on incorporating Python into GIS, should you need to.
There were a couple of mentions of INLA, which if you don’t know stands for Integrated Nested Laplace Approximation, a method of performing Bayesian inference. I have to admit this is well beyond my knowledge. However there are parts of an applied Bayesian modelling course in October that may be relevant. There is also a community based around INLA and R that may be of use and a 98 minute youtube lecture on Bayesian computing with INLA.
MARK is software for the analysis of mark-recapture data, which provides substantial documentation as well as a support forum and a list of training courses. Other than their provided support we haven’t been able to find much, perhaps activity on their forum is the best move if you are struggling.