The Data Charrette is a biannual service event in which professionals, students, and community members from a wide range of fields – computer science, data science, math and statistics, business, biology, public health, graphic design, and the social and behavioral sciences, among others – collaborate with local nonprofit service organizations on data analysis. Over the course of two days, they take the project through the complete data science cycle: obtaining data, scrubbing/manipulating data, exploring/visualizing data, modeling data, and interpreting the results of their analyses. At the end of the event, the nonprofit receives a complete analysis, all visualizations and code, and a template for continuing the work so they can continue to improve the services that they provide.
The Data Charrette follows the examples set by the globally-oriented organization, DataKind.org and the wonderful, student-run A2 Data Dive of the University of Michigan, as well as our own experience with previous iterations of this event.
What makes the Data Charrette unique is that it is hosted by an undergraduate institution – Utah Valley University – with an emphasis on including non-technical students, professionals, and community members. We’ve learned from our experience that the ability to frame questions, coordinate activities, facilitate communication, and make the results interpretable and useful – which can be done by people from any background – are as important as the technical aspects of data manipulation and statistical modeling. What this means is that people with little or no experience in working with data are fully encouraged to come and participate, along with the coders, analysts, and designers.
(We’ll mention that while it is possible to set up instances of AWS servers and do artificial neural networks at this event, that is something that would be left to the discretion of individual participants. As with the most data science, the vast majority of work at the Data Charrette is spent on cleaning, organizing, and exploring data in the not-so-exotic environment of everyday spreadsheets. On the other hand, SQL, Bash, SPSS, R, and Python are always welcome!)