A CASE STUDY OF LEARNING ANALYTICS WITHIN A STATISTICS COURSE FOR UNDERGRADUATE STUDENTS IN ECONOMICS Catherine Dehon, Philippe Emplit and Emma Van Lierde Université libre de Bruxelles cdehon@ulb.ac.be Higher education institutions globally face a continuous expansion of their enrolment in which learner success constitutes a major challenge. Therefore, there is growing interest in the analysis of data linked to student learning engagement. Indeed, large amounts of learning-related student data are currently not being fully exploited, while their aggregation and quantitative analysis would definitely be elements valuable to support teachers and students, to optimize students’ learning experience. In this global context, we have applied, in a public university without any academic filter for enrolment, such analysis to virtually tutor first-year undergraduate students in a statistics course. By supporting them in the form of voluntary online self-assessing tests, we examined what were the personal profiles of the students who were using available tests and how they exploited this help. Finally, using econometric models we tried to determine if there was a link between student success and the use of this help. INTRODUCTION Belgium is a federal state where the language-based communities (French-, Dutch- and German-speaking) are competent for the educational system. The French-speaking community (FWB) offers a unique framework for the analysis of success at university. The main highlighted features of this system are the openness of enrolment at any university degree to (almost) all socioeconomic and educational backgrounds, the non-existent grade publishing requirements and very low, common tuition fees (an important part of higher education is financed through public funds). As a result, almost 70 percent of the student population that finishes the general high school system enrols at university. However, during the first year, very high rates of failure and drop-out are observed (Arias & Dehon, 2013). Moreover, statistical courses are present in many curricula, including in the humanities and social sciences, where the skills and prerequisites in mathematics as well as student motivation for basic courses in statistics are extremely variable. It is therefore very important for this type of course to be able to best help students to success. In this specific community and institutional context, we would like to assess, by the way of learning analytics, how some virtual help, especially designed for early-stage learners by their educators and posted online on the institutional learning management system (LMS), the university’s educational platform, is used by students and whether or not the use of these tools has an impact on their success. Indeed, there has been since 2011 an increasing interest in learning analytics (LA), i.e. the acquisition and the quantitative analysis of data linked to learners’ academic activity, enabling higher education institutions to exploit large amounts of student data that were previously not used to their full potential, with among other sources the Learning Management System (LMS) (Leitner, Kahlil, & Ebner, 2017; Siemens & Long, 2011). This aggregation and analysis of these data enables the support of institutions’ main stakeholders, namely learners, instructors and the administrative staff, for improving student experience and eases the understanding of the current situation and actions to be taken to achieve such improvement (Siemens & Long, 2011). Examples of learning analytics use has been seen worldwide, with tools such as “Ma Réussite” at the Université Laval in Québec enabling all stakeholders, i.e. learners, educators or academic officials, to take appropriate actions, on the basis of some student’s performance coloured indicators. In this case, those are related to the student individual use of online resources compared to the aggregation of the use of his/her peers, and they are generally considered as adequate predictors of final grades (Pothier, 2016). A similar tool was used at Purdue University and seems to have yielded positive results, with higher retention and higher performance for students who used the tool (Sclater, Peasgood, & Mullan, 2016). By using a unique data set containing the entire enrolled undergraduate student population in economics at the Université libre de Bruxelles (ULB), this case study aims to be the first complete analysis to investigate which online elements might be considered, be useful and be consistently developed in the context of the creation of a predictive model for student success, based on the use by IASE 2019 Satellite Paper – Refereed Dehon, Emplit & Van Lierde In: S Budgett (Ed.), Decision Making Based on Data Proceedings of the Satellite conference of the International Association for Statistical Education (IASE), August 2019, Kuala Lumpur, Malaysia. ©2019 ISI/IASE iase-web.org/Conference_Proceedings.php