twstats: Fetch random table data for use in questions / examples

This package provides a wrapper around various sources of national statistics to provide tables of data that match an approximate template. Whilst an independent entity, it is primarily designed for use within the tutor-web LCMS, so we can generate questions based on actual live data rather than made up examples.

This package does not attempt to try and provide a complete interface to every data source, and is not useful if you wish to search for specific data sets on a particular topic. The aim is to provide as many different data tables as possible, constrained just enough that an R script can generate a question from them.

Prerequisites

Any dependencies of twstats will be installed as you install the package, however the data packages it understands will not be installed automatically. You will need to make sure at least one of the following is installed:

Eurostat

Eurostat requires nlme, which isn’t available on CRAN for the version of R in Debian stable. Make sure you use the debian package instead:

apt install \
    libudunits2-dev libcurl4-openssl-dev libssl-dev libgdal-dev \
    r-cran-nlme # Newer NLME requires R 3.4, not in Debian stretch

Finally, install the package:

install.packages('eurostat')

Installation

You can install using devtools:

devtools::install_github('tutor-web/twstats')

Usage

See the examples vignette for an overview of how to use the module. For reference read the R manuals.

Acknowledgements

This project has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement no. 825696.