I use visualizations both for teaching and research purposes (e.g. interactive joint distributions). I mainly work with R, Shiny and Plotly and I am very interested in the potential of interactive data visualization. Most of my projects are published on my github repository. Please contact me if you have any feedback or questions.
- Visualizing Causal Scenarios [interactively]: A shiny app to visualize causal scenarios (see the corresponding paper). The graph above is related to that project.
- Styling plotly layout & margins: A shiny app to illustrate different layout settings for Plotly (in R).
- Styling plotly markers: A shiny app to illustrate different marker settings for Plotly (in R).
- Guessing distributions: Let your students guess what distributions of various variables look like and discuss them subsequently. You can find the app here and the code on github.
- Measurement error (Bull’s eye): Illustration of systematic and random measurement error. Example is a single person that repeatedly measures his/her weight on a scale. Students can change the number of measurements (observations), i.e. how often the person measures his/her weight, as well as the random error and the systematic error underlying these repeated measurements. You can find the app here and the code on github.
- Visualizing functions: Plot functions. User can choose a certain function, decide about the range for which the function should be plotted and the range of x- and y-values for which the plot is displayed. You can find the app here and the code on github.
- Transformations of variables/data: A simple app to illustrate what happens to the the distribution of a variable when it is transformed. You can find the app here and the code on github.
- Joint distributions (discrete variables): Again the idea is that students become familiar with the idea of joint distributions, i.e. develop a “distributional perspective” of data. You can find the app here and the code on github.
- Systematic measurement error in subgroups: Illustration of how distributions of variables change as a consequence of measurement error. It also illustrates how distributions change if there are different systematic measurement errors across subgroups that operate simultaneously. You can find the app here and the code on github.