ShinyStats statistics training apps
ShinyStats: A new interactive and engaging statistics training application to complement existing resources
Team Members: Riinu Ots, Paula J W Smith, Catherine A Shaw, Uzma Tufail-Hanif, Ewen M Harrison, Stephen J Wigmore, O James Garden
School: Edinburgh Medical School (Deanery of Clinical Sciences)
“ShinyStats” is a series of web-based interactive applications (Apps) created using the R statistical programming language and a relatively new R extension called Shiny. Students are given already visualised sample data to interact with; for example, they can manipulate the number of observations. Each time an observation is altered, the application displays the p-value (an indicator of statistical significance). This level of experimentation gives the students a more thorough understanding of statistical concepts, benefitting not only their own research and analysis but also the ability to understand the results quoted in published medical literature.
As a pilot, we have created and tested 3 ShinyStats apps: “t-test”, “chi-squared”, “statistical power”. These apps have been trialed within the virtual learning environment of the Edinburgh Surgical Sciences Qualification (ESSQ) suite of programmes. Uptake and feedback from our students has been universally positive and demonstrates the real demand for accessible ‘user-friendly’ statistics training within the post-graduate student community.
With funding, we plan to (i) increase scalability to accommodate more student interactions at the same time, (ii) add at least 6 more statistical tests to the App, (iii) train our current academic staff and postgraduate students to facilitate the online discussion boards associated with this new module.
Ultimately, we anticipate that the ShinyStats statistics resource developed through PTAS funding will be applicable for use by any programme which includes a research element, both online and on-campus.
Download the final report (PDF)
Other project outcomes
Book Chapter (to be published in 2020) : HealthyR: R for Health Data Analysis (to be published with CRC Press/Chapman&Hall in 2020)