The proposed project is a 1st attempt to utilise and deeply analyse a variety of engagement analytics collected at ARU, to conduct a comprehensive analysis of students’ behavioural patterns (e.g. correlations between different engagement indicators) and the relationship between engagement measures & student success. We ask whether analytical data can help to predict student performance and retention concerns, and how analytics can help us to help students on our courses.
A variety of technologies are available across the sector providing data on human interaction. Universities are increasingly engaging with ‘learning/learner analytics’, defined as measuring, collecting, analysing & reporting of data about learners and their contexts, with the purpose of understanding and optimising learning environments (Long & Siemens, 2011), but this data is often underused (UUK, 2016). This data can also be used to evaluate the impact of learning design (Hernández-Leo et.al, 2018).
Using Talis Elevate, a new area of analysis becomes available, offering a new level of insight into not just the learner behaviours with content, but also allows for deeper analysis into their interaction & engagement within the content. This enables academics to ascertain the impact and effectiveness and use of resources at a more granular level (where does the cohort tail off within a resource), opening up many new areas of analysis for student & learning design analysis.
We seek to answer the following
Which measures of engagement show the greatest correlation to attainment?
What are the patterns of user behaviour and interaction, and how do these correlate with attainment?
Which elements of analysis provide the greatest insight into learning design effectiveness?
Whilst this project aims to identify the most valuable data for ascertaining potential attainment relationships, we will detail limitations from each dataset, the impact of outliers on the final results, and potential misleading insight from elements across the project. This project is limited by the data available during the project, so is not utilising all possible data recorded with all student interaction points.
Whilst it is acknowledged that the data available can give valuable insight into student interaction, it is accepted that relying on data to predict success in studies needs to also take into consideration a number of other factors, such as personal preferences towards studying which will not be measured as part of this project at this time.
This session will detail the research and analysis undertaken from analytics obtained on a number of courses within the Faculty of Science & Engineering Faculty at Anglia Ruskin University during the 2018/19 academic year, using Attendance Monitoring data, Canvas analytics, and Talis Elevate data, showing interaction with resources on the module. The session will present the methodology and findings from the project, and future research activity following on from this initial project.
By attending this session participants will have an opportunity to explore alternative measures of student engagement & how these relate to attainment. In addition, there will be a chance to consider the limitations of using different types of datasets, and how this type of approach could be taken at their local institution.
Hernández-Leo, Martinez-Maldonado, Pardo, Muñoz-Cristóbal, and Rodríguez-Triana (2018). Analytics for Learning Design: A Layered Framework and Tools. [online], avaiable at https://core.ac.uk/download/pdf/161233912.pdf [accessed 8/3/19]
Long, P. D. & Siemens, G. (2011). Penetrating the fog: Analytics in learning and education. Journal of Interactive Online Learning Dietz-Uhler & Hurn 26 EDUCAUSE Review Online.
UUK (2016), Learning Analytics in Higher Education. [online], avaiable at https://www.universitiesuk.ac.uk/policy-and-analysis/reports/Pages/analytics-in-higher-education.aspx [accessed 8/03/19]