Student success

“It is a truth universally acknowledged that” early identification of students at risk and timely intervention must lead to greater success. One of the successes of the AD Programme is that it has demonstrated, through LIDP’s results, that activity data can identify students at risk from patterns in activity data. These students could be supported by early intervention. It has also been demonstrated in work in the US that it can help students in the middle to improve their grades.
Projects working in this area
Recommendations
In year 2, JISC should fund research into what is needed to build effective student success dashboards
Work is needed at least in the following areas:
  • Determination of the most useful sources of data that can underpin the analytics
  • Identification of effective and sub-optimal study patterns that can be found from the above data.
  • Design and development of appropriate algorithms to extract this data. We advise that this should include statisticians with experience in relevant areas such as recommender systems.
  • Watching what others are doing including in the areas of learning analytics, including Blackboard and Sakai developments. This can also draw on the work of STAR-Trak:NG.
  • Development of a common vocabulary
At this stage it is not clear what the most appropriate solutions are likely to be; therefore, it is recommended that this is an area where we need to “let a thousand flowers bloom”. However, it also means that it is essential that projects collaborate in order to ensure that projects, and the wider community, learn any lessons.
Pilot Systems
In year 2 or 3, JISC should pilot the following systems developed in the current programme:
  • LIDP - further refinement of the algorithms used for instance to look at the effect of different patterns of activity as well as the overall level of activity.
  • EVAD - trial the EVAD approach and (part) codebase elsewhere at the other Sakai implementations in the UK and/or apply the approach to other VLEs (Blackboard / Moodle).
  • STAR-Trak:NG - Further development with trial(s) elsewhere. This could include developing a generic framework to support identification of students at risk