Edward Gillian 23/03/2021
The messydatapackage has a number of goals:
Firstly, this package uses different tidyverse techniques such as
pivot_longer to group employee data in 2 semi-long formats as the data
for salary and commuting have different factor levels.
Secondly, tidyverse techniques are used to create contingency tables
for calculating statistics and graphing plots. Different plotting
functions are used including ggplot2, corrplot, and base plot.
The plots are displayed on a tabbed summary page in R Shiny.
Thirdly, R Shiny reactive elements are included for the corrplot
plots to allow for custom data visualisations. Also, reactive elements
are included to let the user choose the dependent variables and
independent variables for the logistic regression models.
Finally, automated testing is done through chained test functions using
testthat. These functions allow the developer to add different input
files to test the functions stored in the R folder to be tested for
reliable outputs. The functions use expect_known_value to generate the
test outputs.
You can install the released version of messydatapackage with:
devtools::install_github("EdwardJGillian/messydatapackage")