Year: 2015
Type: conference-paper
Abstract: Learning analytics discourse occupies an increasingly prominent position within online distance learning literature. Learning analytics refers to the large-scale collection and analysis of student data for the purpose of providing feedback to students, instructors, and organizations. Digital, networked, and computational technologies facilitate these measurement and intervention processes. Learning analytics are highly touted by some in the education and data mining fields, and it is said that analytics will revolutionize online learning with their sophisticated predictions, modelling, and adaptive machine-learning techniques. Despite claims that learning analytics offer actionable intelligence to their users, learning analytics are not so intelligent that they can compute, predict, or recommend their own legitimacy. Instead, this paper argues that the promulgation of learning analytics depends on an ideology called the algorithmic sublime. To describe what this ideology is and how it works, this paper uses critical discourse analysis as a qualitative methodology to interrogate learning analytics discourse. This study presents four patterns of learning analytics ideology: Modal Bootstrapping, Ethicality, Epistemic Displacement, and Collect It All. These patterns are explained using examples from the literature. This study is significant because it encourages debate about an emerging issue and introduces a novel research methodology into the field.
Author: Michael M. Widdersheim
Institution: University of Pittsburgh
Language: English
Subfield: Computer Science Applications
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: green
OpenAlex ID: W2239047154
