An Introduction to the October-December issue

WELCOME to the last issue of 2017, the second extended issue of this year which brings together 12 papers. We start the issue with two survey papers that offer deep analyses of popular research areas. Robert Bodily and Katrien Verbert present a “Review of Research on Student-Facing Learning Analytics Dashboards and Educational Recommender Systems”. Research into systems that track learning analytics data and report it directly to students is becoming increasingly popular in areas such as learning analytics, intelligent tutoring, and educational recommender systems. The authors examined nearly 1,000 articles on the topic, retrieved from databases and journals, and selected 93 articles to include in the analysis. The analysis reveals subtopics that have not received sufficient attention and offers recommendations for further research. A survey of Augmented Reality Learning Experiences (ARLEs) is provided by Neven Drljevi c, Lung Hsiang Wong, and Ivica Boti ci in “Where Does My Augmented Reality Learning Experience (ARLE) Belong? A Student and Teacher Perspective to Positioning ARLEs”. The survey adopts the dual criteria of enabling meaningful learning and reducing teaching effort. It finds that some ARLEs support one or more of active, constructive, authentic, intentional, and cooperative learning. But only two papers from the survey describe ARLEs that offer effective support for a teacher to guide the activities. The next group of papers in this issue focus on problems of student modelling. The paper “Design and Evaluation of a Case-Based System for Modelling Exploratory Learning Behavior of Math Generalization” by Mihaela Cocea and George D. Magoulas investigate the problems of learner modelling in exploratory learning environments. The authors propose an interesting strategy-focused mechanism that is able to assess a learner’s progress by monitoring learner interaction with the system. The approach was evaluated in a learning environment for mathematical generalization. “Dynamic Bayesian Networks for Student Modeling” by Tanja K€aser, Severin Klingler, Alexander G. Schwing, and Markus Gross attempts to advance the state of the art in probabilistic student modelling. The paper suggests a student modelling approach based on dynamic Bayesian networks, a more powerful technology than Hidden Markov Models used in traditional Bayesian knowledge tracing. To demonstrate the power of this new approach, the paper shows the benefits of new models on five data sets from various learning domains. In “Predicting Student Actions in a Procedural Training Environment”, Diego Riofr ıo-Luzcando, Jaime Ram ırez, and Marta Berrocal-Lobo suggest an approach to build a collective student model by processing past student logs. This model is then used to predict the actions of new students, to improve tutoring feedback provided by an intelligent tutoring system. The proposed model has been validated using student logs collected in a virtual laboratory for teaching biotechnology. The remaining part of the issue introduces papers on a broad range of learning technology topics. The paper “How Learners’ Interactions Sustain Engagement: A MOOC Case Study” by Ayse Saliha Sunar, Su White, Nor Aniza Abdullah, and Hugh C. Davis examines social behavior of learners on FutureLearn MOOC courses. It finds that learners who ‘follow’ other participants are more likely to complete a course. The paper proposes social following alongside posting to forums as predictors of course completion. The results of trials of software tools to prompt reflection in the workplace are presented in “In-App Reflection Guidance: Lessons Learned Across Four Field Trials at the Workplace” by Angela Fessl, Gudrun Wesiak, Ver onica RiveraPelayo, Sandra Feyertag, and Viktoria Pammer. The trials showed that the tools to track mood and activity were adopted into the workplace, though the depth of reflection they prompted was generally low. “Automated Essay Feedback Generation and Its Impact on Revision” by Ming Liu, Yi Li, Weiwei Xu, and Li Liu describes an analysis of teacher comments on student essays that provides data to train a machine learning system for classifying types of response to students. The system generates feedback on weaknesses in the language and organization of essays. A study in two English writing classes shows it can be effective in helping students improve aspects of essay content and structure. Design of effective educational games is the topic of “Let Them Play: The Impact of Mechanics and Dynamics of a Serious Game on Student Perceptions of Learning Engagement” by Yichuan Wang, Pramod Rajan, Chetan S. Sankar, and P.K. Raju. In an educational game, designed for the research study, clear goals and perceived usefulness were found to be