This project contains all the scripts required to recreate the data and analyses present in the paper Isaac et al (2014). Statistics for citizen science: extracting signals of change from noisy ecological data. Methods in Ecology and Evolution, 5: 1052–1060, and in an alternate (older) version of the work. In this paper we use simulated recording data to test methods of estimating range change under a range of recording scenarios.
Full details can be found in the vignette, which you can download via this link.
We recommend you use RStudio since this has a good interface with GitHub. You also need Git, this allows RStudio to talk to GitHub. Install git, first, and once you have this installed install RStudio. If you don't have R you will need to download and install that too. Finally, to run occupancy models we need a programme called JAGS installed. This can be downloaded from sourceforge
With R studio and Git installed, open RStudio and go to File> New project> Version Control> Git. Under 'Repository URL' paste the address for this page 'https://github.com/BiologicalRecordsCentre/RangeChangeSims'.
If you have just installed git for the first time you might get an error from RStudio saying it can't find git. To fix this go to Tools> Options> Git/SVN> and browse for the git.exe (usually in a folder such as 'program files/GIT/bin/git.exe').
Once you have created this project run the script 'Run_full_analysis.r' to recreate the majority of results presented in the paper.
The various modelling methods used can be found in the R package 'sparta'. A version of this package is included in this repository, however we recommend you install the most recent version of this package from sparta's repository on GitHub. There you will also be able to read tutorials and ask questions as well as report problems.