Representing the Data Doubles research team, Andrew Asher gave a presentation entitled, “Do students care? Student perspectives on personal data use in library learning analytics,” at the 10th International Evidence Based Library and Information Practice Conference in Glasgow, Scotland. The presentation began with the premise that oft-repeated assumptions that “students no longer care about privacy” (for example, see this piece in The Chronicle of Higher Education) are researchable questions that can be productively addressed by talking directly to students about their privacy perspectives, expectations, and practices.
Using interview data from Phase One of the Data Doubles project, the presentation argued that students, in fact, care quite a bit about their personal and academic privacy, and often express complex and nuanced opinions about how their universities should collect, use, protect, and share information about them. However, students also expressed confusion and a lack of knowledge about the extent of learning analytics data collection and analysis practices, indicating serious problems with universities’ consent and disclosure processes. The presentation concluded with an argument that libraries and universities should therefore treat informed consent procedures as an important educational tool rather than simply a legal or regulatory responsibility.
In the discussion following the talk, questions focused on the issue of reflexivity in Data Doubles’ research methodology, specifically whether researchers with strong professional stances about student privacy could effectively conduct unbiased interviews. While the Data Doubles team acknowledges that because of students’ generally low levels of awareness of learning analytics practices and privacy literacy the research interviews contained necessary discursive and educational elements, we also designed the protocols and questions to be explicitly student-driven, with additional explanatory information provided only when a student asked, or when was required for a participant to adequately discuss a particular issue. The Data Doubles team also designed the interview questions to be as value-neutral as possible to avoid biasing students’ responses and implemented review procedures during the analysis process to check for potential interviewer effects within the data. In the interest of transparency, Data Doubles will also be publishing our full interview protocols for interested researchers to evaluate independently in our forthcoming Toolkit.