Learning analytics tools, systems, initiatives, frameworks, and models: An annotated bibliography Publications
This annotated bibliography reviews 261 papers on learning analytics tools, systems, frameworks, models, research studies, and tertiary education provider (provider) initiatives in New Zealand, Australia, Canada, the United Kingdom, and the United States.
Author(s): Peter Guiney, Tertiary Sector Performance Analysis, Ministry of Education.
Date Published: December 2016
The key findings of this report are:
- To successfully introduce learning analytics initiatives, providers need skilled and expert staff and an appropriate organisational culture. Leadership support, staff buy-in, training of existing staff, and policies are required. These initiatives should align with organisational strategic goals and objectives and be supported by the necessary infrastructure.
- Some providers, such as Purdue University in Indiana, Austin Peay State College, Rio Salado Community College, the United Kingdom's (UK's) Open University, and the University of Wollongong, have implemented significant learning analytics initiatives. In addition, the Australian and UK governments have funded a number of learning analytics projects. However, overall adoption is low and many of the existing initiatives and evaluations are immature.
- The main type of learning analytics systems are early alert or warning ones that provide personalised and timely interventions for struggling or under performing students. The most commonly used learning analytics tools are 'dashboards', which visually represent indicators of student progress and academic performance.
- The major barriers to learning analytics are an unsupportive organisational culture, a lack of leadership support, a lack of staff and student capability, insufficient time to undertake the necessary work, and staff resistance. Many of the available tools are immature and not sufficiently user friendly. Implementing learning analytics can also be expensive because of staff training costs, support costs, and additional system costs (eg, algorithm development).
- A major issue confronting learning analytics is accessing and using student data. It is important that providers consult with staff about which data is most useful for learning analytics purposes and the terms and conditions of its use. It is also recommended that providers make clear to students who can access their data and under what conditions and for what purpose it will be collected and used.
- Some learning analytics and projects can help students acquire self-directed learning behaviours and attributes. However, in other cases these initiatives and projects can further reduce motivation for struggling students and may not sufficiently challenge or motivate high performing students.