In the last few years, most educational institutions have adopted different solutions based on e-learning in order to both supplement classroom teaching or offer entirely online training. Other institutions like the Open University of Catalonia (UOC) have been online educational communities from their inception. In both cases there is a virtual space in which students interact with other students and teachers, as well as the resources and services offered by the institution. In this setting, which goes beyond areas outside the traditional classroom, it is a challenge to analyse the complexity of the learning process, from the student and teacher’s perspective, taking into account the different spaces of interaction with quantitative data collected automatically, including surveys and qualitative interviews.
Learning Analytics is an emerging discipline that includes a set of techniques and methodologies to capture, measure, analyse and visualize information relating to the learning process of students in virtual environments, with the aim of extracting useful knowledge. That knowledge is applied to understand and optimize the learning process, reintroducing it in the form of action. Learning Analytics is based on an interdisciplinary evidence-based research which combines research from two different angles: the underlying conception of learning or the educational context (Learning); and the application of new computational, statistical and visualization methods (Analytics) to their understanding. And finally, our challenge is to measure the real impact of each proposed solution in a continuous cycle of observation, analysis, and improvement action.
To sum up, there are four key elements which shape LAIKA as a singular research group:
- Interdisciplinarity: research problems require an interdisciplinary approach and a team with complementary profiles. LAIKA aggregates four different disciplines.
- Innovation in new scenarios for teaching and learning: LAIKA is involved with several innovation projects concerning from e-assessment to MOOCs.
- Network of top Open Universities in Europe: LAIKA is collaborating actively, and leading initiatives, with Open University UK, Open University Nl and UNED.
- Evidence-based research: an excellent position to develop data-driven research based on analytics using data from UOC students.
Doctoral research lines
1. Learning Analytics for action: Intervention design and analysis in programming and math courses
This research line uses data analysis to design an intervention at course level and find a relationship between that and (improved?) retention or student performance and satisfaction. We also analyse the influence of the educational personal feedback in the learning process and the relevance, for instance, of the emotional factors in this process. This analysis also includes the study of appropriate personalised feedback tools to formative assessment, specifically in the context of learning by competence like rubrics and portfolios, among others.
2. Design and evaluation of interventions for repeaters in introductory programming courses
It is well known that learning to program is not an easy task. Many students struggle with algorithmics and coding during their first semester and fail, having to repeat the programming course in a later semester. However, when they have to take the course for a second time, they find that they can take advantage of practically nothing of what they did and learned the first time they took it. In this research line we want to design and evaluate interventions for repeaters in an introductory programming course, trying to understand what happened the first time they took the course, the reasons that led them to drop out or fail, and their optimal starting point and the necessary support when they repeat it. In order to do so, we will combine available data from thousands of students taking an introductory programming course with questionnaires and interviews with repeating students, following a learning analytics approach.
3. Multilevel analysis of student trajectories in STEM grades: relationships between core subjects, academic results, mentorship, dropout and re-enrolment
Students enrolled in STEM degrees (like Computer Science or Data Science) have to overcome a number of obstacles in the form of core courses during their first semesters, such as math or programming courses. Knowing how students progress in each course in which they are enrolled and the mentoring actions, their academic results and their relationship to dropping out or re-enrolment the next semester, is important to understand which factors contribute to their long-term success. Our goal is to address some well-known problems of online/distance students, such as difficulties in following the planned schedule of activities, learning abstract concepts, how to measure the difficulty of each activity / subject, the impact of the received feedback in their academic results, and the relationship between such results, their enrolment and dropout, among others. In order to do so, we will combine available data from thousands of students taking STEM degrees with questionnaires and interviews, following a learning analytics approach, using multilevel analysis and other advanced statistical and data mining techniques.
Education and ICT (e-learning). Responsive Teaching and Learning Processes and Outcomes in Online Education
4. Tools for supporting the teaching-learning process in fully online programming courses
Learning introductory programming is considered difficult for novice students. As a result, drop-out rates in programming courses are usually high. This fact is worse when the learning environment is fully online. Therefore, teaching programming online is a great challenge.
The main goal of this research is to design, develop and test e-learning tools that support students and instructors throughout teaching-learning process in fully online programming courses. The topics of interest of this research include, but are not limited to, the following:
- To generate automatic feedback in order to support students during their learning process. Feedback can include design, functionality and quality aspects, among others.
- Give teachers mechanisms that allow them be able to provide effective proof of student authenticity and authorship of the programming activities in a cost efficient manner.
- Tools that allow teachers and students to do programming assignments/activities in the cloud, e.g. Web IDE, collaborative tools, etc.
- Automatic or manual assessment tools that support teachers while they grade programming assignments, e.g. dashboards, static analysers, rubrics, etc.
- Tools that help students to understand programming concepts more easily, e.g. tracing/debugging tools, compiler with easy messages, contextual hints, text-based screencasts, etc.
- A new instructional design (i.e. schedule, activities, assignments, tools, etc.) that helps students to acquire programming skills.
5. Visual learning analytics for virtual learning environments
Virtual learning environments generate huge amounts of interaction data that can be analysed and visualised in order to better understand both the teaching/learning process and users’ behaviour. This analysis can be done at different levels of detail, combining data from multiple sources (services, learners’ profiles, etc.) coming from one or more educational scenarios (a virtual classroom, blog, repository, etc.). Research on this topic is meant to build robust models that can be used to help learners, teachers and managers to fulfil their goals, and to detect and resolve bottlenecks in virtual learning environments, as well as identifying and explaining the most relevant reasons for these, by means of visual learning analytics (both methodologies and tools).
6. Interactive recommendation systems for higher education enrolment
Higher education students at open / distance universities enjoy from a high degree of flexibility during enrolment, which allows them to choose from a long list of subjects to complete their degree. Although this can be seen as a success of enrolment flexibility measures, it may be also the source of one of the most well-known problems in open / distance education: high dropout rates, partly caused by inadequate enrolment. In this research line, we will analyse and adapt state-of-the-art recommendation systems to the particularities of the enrolment procedure, taking into account enrolment data and academic results from previous semesters but also students’ preferences and personal interests. Our goal is to design and evaluate interactive recommendation systems that provide students and their mentors with support during enrolment, following a user-centred design approach.