Closing the loop is an automated technique defined as completing “the cycle of system design, deployment, data analysis, and discovery leading back to design” (Liu & Koedinger, 2017). It enables automated detection and intervention if needed, which can be considered a form of feedback. It also allows educators to modify the course content or even create a personalized curriculum or course plan based on the student activity and performance. Consequently, this improves the learning outcomes (Liu & Koedinger, 2017). For instance, Lawson, Beer, Rossi, Moore, and Fleming (2016) elaborated on how CQUniversity Australia implemented an automated engagement system called Early Alert Student Indicators (EASI), which approximates students’ success rate, thereby, identifying students who are “at risk” of failure or dropping out of the doctoral program in order to develop early intervention strategies. The student data used to calculate the success rate was gathered from various university systems and combined with students’ learning activity on Moodle, an online learning management system (Lawson, Beer, Rossi, Fleming, 2016). The data analyzed consisted of the students’ name, gender, study load, enrollment history, and grade-point average (GPA) history (Lawson, Beer, Rossi, Fleming, 2016). As for the early intervention strategies, a built-in mail-merge facility to send a personalized email to students, offering academic assistance and support (Lawson, Beer, Rossi, Fleming, 2016). An email is quite a common automated intervention strategy.
Nonetheless, closing the loop was also used to improve an automated cognitive modelling in order to redesign the intelligent tutoring system. Cognitive models map the tasks of the intelligent tutoring system with the underlying knowledge components (Q-matrices) (Liu & Koedinger, 2017). Liu and Koedinger (2017) created the Learning Factors Analysis (LFA) to automate the improvement of cognitive modelling, which permits automated discovery and is based on the data-driven method. The results revealed that there were statistically significant learning gains with an automated cognitive model refinement (Liu & Koedinger, 2017).
Overall, closing the loop is a useful technique in self-directed learning as it provides automated feedback to learners’ performance in order to reflect on, intervene, and guide them. It is just as useful in traditional educational institutions as is shown by Lawson, Beer, Rossi, Moore, and Fleming (2016). Furthermore, it can be used to improve students’ attendance by notifying them and the instructors if the number of absences exceeds a certain threshold. It can also be applied to MOOCs by helping to decrease the attrition rates due to early intervention strategies such as curriculum revision and email reminders of students’ enrollment (Whitehill, Williams, Lopez, Coleman, & Reich, 2015). In other words, it enables personalized self-directed learning.
Lawson, C., Beer, C., Rossi, D., Moore, T., & Fleming, J. (2016). Identification of “At Risk” Students Using Learning Analytics: The Ethical Dilemmas of Intervention Strategies in a Higher Education Institution. Educational Technology Research and Development, 64
Liu, R., & Koedinger, K. R. (2017). Closing the Loop: Automated Data-Driven Cognitive Model Discoveries Lead to Improved Instruction and Learning Gains. Journal of Educational Data Mining, 9
Whitehill, J., Williams, J. J., Lopez, G., Coleman, C. A., & Reich, J. (2015). Beyond prediction: First steps toward automatic intervention in MOOC student stopout.