Last week I had the good fortune of attending my first CAIR–no, not the Council on American-Islamic Relations, but the California Association for Institutional Research. Over 300 administrators, analysts, and data nerds decended upon the Millenium Biltmore Hotel in Los Angeles to network, share tips and tricks, and even watch a live shoot of ABC’s The Catch.
From the beginning, it was clear R was still relegated to the fringe: R was unconscionably omitted from the “favorite statistics package” group mingling session, Tableau was the visualization software of choice, and of the hundred or so odd sessions, only three were about R. They were:
- Creating Dashboards of Direct and Indirect SLOs Measures in R (Boosalis, Myhre & Binod)
- R for Everything: Getting, Cleaning, Analyzing, and Visualizing Data (Ranzolin)
- Leveraging the power of regression discontinuity designs for program evaluation: an institutional simulation using R (Orona)
I can’t speak for my own, but the first and last presentations were superb. If interested, I’ve uploaded my presentation to RPubs here. Session description below:
R is a powerful tool for statistical computing and graphics. It has, unfortunately, been caricatured as weird, unforgiving, and applicable only to hardcore statisticians. The objective of this session is thus twofold: (1) to demonstrate R’s versatility up and down the data pipeline; and (2) to demonstrate R’s overall accessibility to users who do not have a computer science background. Towards those ends, I will narrate a typical R workflow at Scout from start to finish including scripts, plots, HTML and PDF reports, and an interactive web application.
While far from perfect, I felt good about the presentation on the whole.