Exploring Election and Census Highly Informative Data Nationally for Australia
The R package eechidna provides data from the Australian Federal elections in 2001, 2004, 2007, 2010, 2013, 2016 and 2019, along with the Australian Census information for each House of Representatives electorate from the 2001, 2006, 2011 and 2016 Censuses. Additionally, Census information is imputed for electorates in years 2004, 2007, 2010, 2013 and 2019. It also includes tools for visualizing and analysing the data.
This package was developed during the rOpenSci auunconf event in Brisbane, Queensland, during 21-22 April 2016. It has been updated many times since to include election and Census information for 2001, 2011 and 2019. Peter Ellis’ work on the NZ electoral data was an important inspiration for this package.
How to install
You can install the latest release of the package from CRAN like this
install.packages("eechidna")Or you can install the development version from github, which may have some changes that are not yet on CRAN, using devtools, like this:
devtools::install_github("jforbes14/eechidna",
build_vignettes = TRUE)
library(eechidna)If you are using Linux, you may need some additional libraries for the mapping functions, you can get these with this line:
apt-get install libgdal-dev libgeos-dev -yHow to use
The package consists of several datasets, which includes Australian Census data at electorate level, Australian Federal election (House of Representatives) voting data from electorates and polling booths, and shapefiles for Australian electoral districts at various points in time. In addition to the data provided, eechidna includes a highly interactive web app for exploring the election and census data together. This app uses the shiny framework, and can be run locally on your computer with the command eechidna::launchApp(). There is a video demo of the app here: https://vimeo.com/167367369.
We have many vignettes that show how to access these data in the package, and demonstrate how to analyse the data using R. These can be found in the articles tab at the top of this page.
An introduction to eechidna: an overview of the package contents and examples of how to use the data.
Exploring election data: examples of wrangling data from the 2016 Federal election data to gain insights.
Exploring Census data: visualizing 2016 Census data to analyse patterns in electoral population characteristics.
There are also three vignettes that demonstrate how to use the spatial data to make maps. Mapping election data for Australia is not trivial because of the extreme variation in electorate size. In these vignettes we show some methods for effectively visualizing election data in Australia. These too are found in the articles tab at the top of this page.
Mapping federal electorates: how to plot a map of Australian federal electorates, and how to better visualize electoral voting data using a cartogram.
Mapping polling booths: examples of how to plot the polling booth locations and associated voting data.
Getting Australian electoral maps: details the process to produce usable electoral maps from the original shapefiles.
Additionally, there is a vignette on how we have imputed electoral Census data in election years for which a Census does not exactly align.
- Imputing Census data for non-Census years: details the procedure used to impute Census data for 2004, 2007, 2010, 2013 and 2019.
Feedback, contributing, etc.
Please open an issue if you find something that doesn’t work as expected or have questions or suggestions. Note that this project is released with a Guide to Contributing and a Contributor Code of Conduct. By participating in this project you agree to abide by its terms.
Acknowledgements
Thanks to Xiaoyue Cheng for her cartogram package which supplies the Dorling algorithm for this package. Thanks also to Roger Bivand for his rgdal and rgeos packages which has some key functions for working with shapefiles. Thanks to Scott Chamberlain and Yihui Xie for help with troubleshooting.