shinymde: A shiny interface to mde, the missing data explorer

2021-12-18

R-CMD-check license lifecycle Maintenance Project Status GitHub last commit GitHub issues GitHub issues-closed PRs Welcome

Goals of shinymde

shinymde provides an application (user interface) to ease the process of missing data exploration. The core functionality accessed is programmed in the code only package mde. It is hoped that this will provide an even easier way to explore missing data.

Installation

To install, please run the following in R.

devtools::install_github("Nelson-Gon/shinymde")
# provide branches for dev version  
devtools::install_github("Nelson-Gon/shinymde@dev")

Running the App

Test the app

shinymde is currently deployed at https://nelson-gon.shinyapps.io/shinymde/ where users can use the app via a browser. Please test the app there and let us know what you think. Thank you!

Currently Implemented features

  • Summary of missingness

This uses mde’s na_summary function to generate a simple missingness report. The user can also download this report for use in other purposes.

  • Recoding as NA and recoding NA as

This allows a user to recode a given value as NA, R’s handler of missing values or convert NAs to some other value. It uses mde’s recode_* functions.

  • Conditional recoding

This allows one to recode values conditionally. This supports recode_as_na_for, recode_as_na_if, and other conditional recoding functions.

  • Drop based on missingness

This uses mde’s drop_na_* functions.

  • Missingness plots

  • Gear icon that allows users to customize plot options e.g. plot themes, axis labels, legends, etc.

shinymde in action

Below is a screenshot showing current functionality.

shinymde in action

View the full app at https://nelson-gon.shinyapps.io/shinymde/.

Features that need further development (help)

If you would like to work on this project, here is a features list that could be a good place to start.

  • Add user guides in the form of tooltips wherever possible.

  • Add unit tests.

  • Add hovers to user input

  • Switchable dark mode

  • Control pattern_type/subset_cols choice in na_summary.

  • Use cached filename in input filename if a user quits the program.

  • Add user help pages that link to mde documentation on click or as an independent tab.

  • Allow user to choose plot type.

  • Support dates in mde

  • Allow control over visible text on visual summaries.

Thank you and keep building,

Nelson

2021-12-18

Please note that the ‘shinymde’ project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.