Visit to the Open University on 29th October

Assessment for Learning: automatic feedback for draft essays – Professor Denise Whitelock, IET
Writing assignments or essays still form the backbone of higher education assessment. Students are given very little immediate support when they are actually drafting their assignments, and this is the time when they would most like some assistance, especially if they are returning to higher education after a break. A system known as OpenEssayist has been built by the SAFeSEA project to assist students in these particular circumstances. The software provides immediate feedback on draft essays. This session will discuss the empirical findings from the testing of OpenEssayist with students from The Open University. It will also highlight the challenges of learning analytics to provide meaningful visualisations to students.

Open approaches to OER impact research – Dr Rob Farrow, IET
In this session the work of OER Research Hub is outlined and the merits of open approaches to understanding impact are discussed. OER Research Hub(funded by the William and Flora Hewlett Foundation) works internationally in collaboration with a range of research partners to assess impact across different education profiles. This often requires rethinking methodological assumptions or working in adaptive and responsive ways in order to facilitate sharing of impact data. OER Research Hub thus uses open methods to investigate the impact of OER on education and learning. Open forms of dissemination employed include open access publication, blogging, open release of research data and research tools, open online courses and the OER Impact Map, which draws together data relating to our research hypotheses and shows how a plurality of content can be brought together through data visualisation and graphical mapping. In addition to discussing how some of these techniques might be used to assess impact in other areas I will present some of the headline findings from the project.

Predictive analysis of at risk students – Prof Zdenek Zhrahal
Universities nowadays collect large volumes of data about everyday learning activities of students. By analyzing these data it is possible to evaluate student’s progress, identify those who are at risk of failing and recommend how to improve his/her performance. We will present machine-learning based models that build predictions from data from the Virtual Learning Environment (VLE) and students demographic data. Since January 2014, we predict weekly for two courses at risk students and provided this information with justifications to the course teams. Predictions are presented as in a dashboard with the views for the whole course and for individual students. The dashboard includes a recommender that navigates each student to the course material most beneficial for his/her progress.

Gathering student’s feedback from social media – Dr Harith Alani
Universities strive to gather feedback from students to improve courses and tutorship. Such feedback is often collected via survey forms when courses or modules end. However, such methods in collecting feedback are too formal, restricted, and in many cases, too late. Many students are finding refuge in open social media to group and share their experiences and to support each other throughout their studies. OUSocial is a platform that monitors behaviour, sentiment, and topics, in open social media groups that are set up by, and for, Open University students. It captures anonymous feedback from students towards their courses, and tracks the evolution of engagement and sentiment within, and across, those groups.

Making the most of ‘business data’ in open and informal learning – Andrew Law, Director of Open Media Unit
After more than 40 years of working with the public in the open – partly to meet our social mission and partly to support our business interests – we have collated vast amounts of data about impact. Much of this data is rather simple, traditional web traffic data. However we can show how even this simple data can help us tune our operating models to support both our social mission and our business models – in order to gain the most from being open