Blog post – How should UQ deal with the educational data revolution?

If you are involved with teaching and learning at UQ, you are probably contributing to the collection or generation of data on students, including data on their learning activities, learning processes, learning experiences or assessment results. While collecting or analysing student data is not new, due to advancements in technology, the volume, variety, velocity, and veracity of educational data has increased dramatically over the last decade during the ‘big data’ revolution era. This has motivated many universities around the world, including UQ, to invest resources in implementing analytical technologies and educational tools as well as analysing educational datasets to better understand and support student learning.

As UQ embraces the challenges and opportunities big data creates in higher education, we have the chance to develop our shared vision and understanding on how educational datasets should be collected, accessed and utilised to improve teaching and learning. How can the educational data revolution influence student learning? How do you use student data to inform your teaching practices and improve student learning? Are you concerned that student data may be misused?

Join the conversation and share your views, ideas and stories by taking our poll. I’ve shared my views below to start the conversation.

How can the educational data revolution influence student learning?

The educational data revolution has empowered scholars from a broad range of fields including learning science, educational psychology, higher education research, educational data mining and learning analytics to research various aspects of education such as student learning, student modelling, teaching methods and teaching dynamics. While these research fields share an aspiration about improving teaching and learning and have a general consensus that the conducted research should be rigorous, their approaches, assumptions, research methods and consequently findings may vary.

With the acknowledgement that there are different worldviews on the issues of teaching and learning that do not necessarily always agree, below are a few examples of how the educational data revolution can influence higher education.

Empowering educators and universities with rich analytics

It is increasingly being recognised that educational tools and technologies are not aimed at replacing educators but rather are used as support tools to enable educators to improve their teaching practices. One of the greatest benefits of having large educational datasets is that it enables educators to meaningfully zoom in and zoom out to gain new insight and answer questions about students’ learning processes and activities, which consequently empowers better decision-making in education. For example, it provides the ability to gain new insight from studying the behaviour of students on a particular learning activity within a particular course and then to zoom out to determine whether the findings are generalisable in other contexts.

Similarly, it is possible to use university-wide datasets to determine patterns between performance and behaviour of students and then to zoom in to determine whether the findings are true within a particular cohort, discipline or demographic. Additionally, having access to big educational datasets provides the ability for developing models that may be used to visualise or predict the performance and behaviour of the students. These models are used to:

  1. provide feedback for students to reflect on their learning
  2. provide insight for educators on students, which allow them to provide personalised feedback as well as improve item and course design, and
  3. help universities identify at-risk students to potentially utilise retention strategies.

Aligned with this vision, UQ has developed a new tool called Course Insights which assists teaching staff in gaining actionable insights and delivering personalised support based on student data collected from various sources.

Developing evidence-based teaching practices

There are many theories and methods that claim to improve learning. The educational data revolution has provided educational researchers the ability to use a variety of research methods to validate or debunk many of these claims. Let’s consider two popular educational theories:

  1. active learning activities increase student performance, and
  2. students will learn more effectively if they are taught in a way that matches their preferred learning style.

A meta-analysis of 225 studies that had collected and analysed student data on active learning activities has demonstrated that active learning increases student performance in STEM courses. In contrast, based on numerous educational studies that have been conducted, there seems to be no adequate evidence to support the existence of learning styles, suggesting that “students are not hard-wired to learn in different ways”. 

The educational data revolution also enables individual educators to reflect on their own teaching practices to investigate their effectiveness. To promote scholarly inquiry about student learning and the development of evidence-based teaching practices, UQ is introducing policies, frameworks and educational tools (such as UQx, Course Insights, Survey tools, UQ Active Learn, RiPPLE, Semant and MOOCchat) that enable the UQ community to collect, access and utilise educational datasets.

Providing adaptive and personalised learning opportunities

Educators continue to face significant challenges in providing high quality, post-secondary instruction in large online or on-campus classes. A significant portion of these challenges emerges from high levels of diversity in learners' academic ability. Adaptive learning systems provide a potential solution for this problem. They make use of data about students, learning processes and products to provide an efficient, effective and customised learning experience for students by dynamically adapting learning content to suit their individual abilities or preferences.

A consistent and growing body of knowledge has provided evidence about the effectiveness of adaptive learning systems, which include intelligent tutoring systems. UQ has developed a new tool called RiPPLE that recommends personalised learning resources to students based on their current learning needs from a pool of learning resources that are generated by educators in partnership with the students themselves.

Why should educational data be handled with care in higher education?

Use of educational data has provided many opportunities in higher education; however, even with the best intentions, data can be misinterpreted or misused. As such, there is an obligation that universities handle educational data with care and to ensure that it is being used ethically and responsibly. Common reasons for misinterpretation or misuse of data include statistical misinterpretations, algorithmic bias, insufficient domain expertise, data breaches, and mistaking correlation for causation.

Regardless of the reason, universities must be careful to ensure that the use of student data does not lead to profiling students in ways that brute force intervention which may harm their performance or learning experience. Instead, student data should be used benevolently to create systems of support, encouragement and to provide scaffolding for students to personalise and enhance their learning experience.

The ethical considerations behind using student and educational data have been well studied in the field of learning analytics. A recent discussion paper from this field raises awareness on the importance of handling student data with care, providing insightful guidelines, protocols and processes for ethical use of educational data.

What do you think?

Poll – educational data revolution

Please contribute to the conversation and share your views below.

Last updated:
6 August 2019