Learning About Data and Fairness

Simon Knight, Kirsty Kitto, Kalvero Gulson, Teresa Swist, Claire Benn, Gradient Institute, Tech for Social Good
Last updated on Nov 14, 2021
View all our research More on project site
Decorative image with title can an automated algorithm make human grading fairer? Image from project website

In a project led by Kalervo Gulson (USyd), centre members are collaborating with the Gradient Institute and ANU’s Claire Benn to co-design an ‘algorithm game’ intending to explore issues around fairness and data, using the case study of the 2020 UK Exams Algorithm Controversy.

The project is using participatory-design to investigate how we support people in exploring how algorithms work, their possible impact, and their relationship with decision-making processes. In the first case, in the UK exams algorithm controversy the government decided to moderate teacher-assigned grades to account for the fact that teachers typically give higher grades than those achieved via an exam. However, the implementation of the algorithm had a range of issues, and caused significant backlash.

Aligned with a ‘technical democracy’ (Callon et al, 2009) approach, the project is an example of the Education Futures Studio (EFS), which Gulson leads, aim to support collective learning and experimentation about the role of automation, artificial intelligence, algorithms, and data across a range of education contexts.

Data-literacy algorithms AI schools HE sociomaterial ethics participatory-design technical democracy
Simon Knight
Simon Knight
Associate Professor

Dr Simon Knight is a senior lecturer in the Transdisciplinary School, co-editor-in-chief of the Journal of Learning Analytics, and Director of CREDS. Simon researches how people find, use, and evaluate evidence.

Kirsty Kitto
Kirsty Kitto
Associate Professor

Kirsty Kitto is an Associate Professor in the Connected Intelligence Centre, where she models the many ways in which humans interact with information, and how this can change as a result of the different contexts in which people find themselves.