Quantitative Methods Forum

When:
October 21, 2013 @ 10:15 AM – 11:45 AM
2013-10-21T10:15:00-04:00
2013-10-21T11:45:00-04:00
Where:
Norm Endler Seminar Room (BSB 164)
Cost:
Free

Speaker: Matthew Sigal, York University
                   Department of Psychology

Title: The Grammar of (Interactive) Graphics

Abstract: In The Grammar of Graphics, Leland Wilkinson (1999; 2005) outlined a fundamental system for understanding how to approach the visualization of data. The language defined in this work describes rules that connect all of the underlying theoretical structures of graphic representation to our data. These revolve around mathematics (e.g. algebra, scales, and geometry), visual properties (aesthetics and facets), and geometric objects ("geoms"). The concepts outlined in this book are still relevant today, as they are implemented in one of the leading R packages for statistical visualization, ggplot2 (Wickham, 2009). However, this framework was developed primarily with static visualizations in mind.
      With developments in computing power and programming languages optimized for interactivity, we now have the tools to incorporate dynamic elements into such graphics. These features can be as simple as adding overlays that give viewers additional information about your data, to full-fledged graphical user interfaces that can load in external data sets, generate multiple plots, and yield summary statistics in a concise panel.
       The initial goal of this talk is to introduce you to the language of graphics. Building upon that, I will present some notable features that we can incorporate into an interactive environment. Finally, I will compare and contrast some of the presently available platforms for producing such graphics, show some motivating examples, and outline some future projects that I am working on which utilize these new methods to aid researchers in conducting model diagnostics.

Suggested Readings: 
          Cleveland, W. S. (1994). The elements of graphing data. Summit, New Jersey: Hobart Press.
          Few, S. (2009). Now you see it: Simple visualization techniques for quantitative analysis. Oakland, California: Analytics Press.
          Murray, S. (2013). Interactive data visualization for the web. O'Reilly Media.
          Theus, M., & Urbanek, S. (2009). Interactive graphics for data analysis. Boca Raton, FL: Taylor & Francis.
          Wickham, H. (2009). ggplot2: Elegant graphics for data analysis (2nd ed.). New York, NY: Springer.
          Wilkinson, L. (2005). The grammar of graphics (2nd ed.). New York, NY: Springer.

Additional

1) Link to the presentation itself (which should work on all modern browsers, such as Chrome or Firefox): http://mattsigal.github.io/InteractiveGraphics/. Note: as this presentation is all in HTML, monitor resolution can affect how this looks, so please don't try to load it at 800x600.

2) Link to the Github page where all the files for the presentation can be sourced: https://github.com/mattsigal/InteractiveGraphics (includes Initialization Script for getting the required packages, as well as subdirectory containing the code used to generate many of the plots in this branch).