The first day of this most interesting and probably seminal (for higher education) conference ended with George Siemens leading a discussion and brainstorming session on what it takes to make up an effective “data science” team within a higher education context. Initial proposals included a stakeholder, who defines the problem or question(s) to be addressed; the data scientist, who is able to manage the way in which different systems work together at a higher level; the programmer, who writes the query; the statistician, who analyzes the data; and, interestingly, the “visualizer,” who is able to portray the data in a way that communicates the key areas of interest. As the discussion developed, additional ideas from participants were thrown into the mix. For example, the team could include the following roles, functions or persons:
A “champion” must ensure this capacity operates as an ongoing strategic resource, rather than as an ad-hoc system only. There is a bridging function – connecting the different specialties, experts or departments to internal and external stakeholders. A project manager needs to pull all this together. And it continues to expand. For example what is the role of faculty—which academic communities are represented along with which other services? Who ensures the quality of data? IT is an obvious resource in this – but do they have the time and resources to give it priority? Do things change when the process moves to the Cloud? Ethics and privacy become critical concerns and can rapidly become a showstopper.
And of course: how does the information get folded back into a process of improvement – e.g. to faculty and instructional designers, among others? That is probably the key question in the end.