This paper surveys the state of the art on prediction in MOOCs through a Systematic Literature Review (SLR). The main objectives are: (1) to identify the characteristics of the MOOCs used for prediction, (2) to describe the prediction outcomes, (3) to classify the prediction features, (4) to determine the techniques used to predict the variables, and (5) to identify the metrics used to evaluate the predictive models. Results show there is strong interest in predicting drop-outs in MOOCs. A variety of predictive models are used, though regression and Support Vector Machines stand out. There is also wide variety in the choice of prediction features, but clickstream data about platform use stands out. Future research should focus on developing and applying predictive models that can be used in more heterogeneous contexts (in terms of platforms, thematic areas, and course durations), on predicting new outcomes and making connections among them (e.g., predicting learners' expectancies), on enhancing the predictive power of current models by improving algorithms or adding novel higher-order features (e.g., efficiency, constancy, etc.).

No abstract

No abstract

No abstract

Learning Analytics (LA) is perceived to be a promising strategy to tackle persisting educational challenges in Latin America, such as quality disparities and high dropout rates. However, Latin American universities have fallen behind in LA adoption compared to institutions in other regions. To understand stakeholders' needs for LA services, this study used mixed methods to collect data in four Latin American Universities. Qualitative data was obtained from 37 interviews with managers and 16 focus groups with 51 teaching staff and 45 students, whereas quantitative data was obtained from surveys answered by 1,884 students and 368 teaching staff. According to the triangulation of both types of evidence, we found that (1) students need quality feedback and timely support; (2) teaching staff need timely alerts and meaningful performance evaluations, and (3) managers need quality information to implement support interventions. Thus, LA offers an opportunity to integrate data-driven decision-making in existing tasks.

a b s t r a c tPresent MOOC and SPOC platforms do not provide teachers with precise metrics that represent the effectiveness of students with educational resources and activities. This work proposes and illustrates the application of the Precise Effectiveness Strategy (PES). PES is a generic methodology for defining precise metrics that enable calculation of the effectiveness of students when interacting with educational resources and activities in MOOCs and SPOCs, taking into account the particular aspects of the learning context. PES has been applied in a case study, calculating the effectiveness of students when watching video lectures and solving parametric exercises in four SPOCs deployed in the Khan Academy platform. Different visualizations within and between courses are presented combining the metrics defined following PES. We show how these visualizations can help teachers make quick and informed decisions in our case study, enabling the whole comparison of a large number of students at a glance, and a quick comparison of the four SPOCs divided by videos and exercises. Also, the metrics can help teachers know the relationship of effectiveness with different behavioral patterns. Results from using PES in the case study revealed that the effectiveness metrics proposed had a moderate negative correlation with some behavioral patterns like recommendation listener or video avoider.

The advancement of learning analytics has enabled the development of predictive models to forecast learners' behaviors and outcomes (e.g., performance). However, many of these models are only applicable to specific learning environments and it is usually difficult to know which factors influence prediction results, including the predictor variables as well as the type of prediction outcome. Knowing these factors would be relevant to generalize to other contexts, compare approaches, improve the predictive models and enhance the possible interventions. In this direction, this work aims to analyze how several factors can make an influence on the prediction of students' performance. These factors include the effect of previous grades, forum variables, variables related to exercises, clickstream data, course duration, type of assignments, data collection procedure, question format in an exam, and the prediction outcome (considering intermediate assignment grades, including the final exam, and the final grade). Results show that variables related to exercises are the best predictors, unlike variables about forum, which are useless. Clickstream data can be acceptable predictors when exercises are not available, but they do not add prediction power if variables related to exercises are present. Predictive power was also better for concept-oriented assignments and best models usually contained only the last interactions. In addition, results showed that multiple-choice questions were easier to predict than coding questions, and the final exam grade (actual knowledge at a specific moment) was harder to predict than the final grade (average knowledge in the long term), based on different assignments during the course.

The use of Massive Open Online Courses (MOOCs) is increasing worldwide and brings a revolution in education. The application of MOOCs has technological but also pedagogical implications. MOOCs are usually driven by short video lessons, automatic correction exercises, and the technological platforms can implement gamification or learning analytics techniques. However, much more analysis is required about the success or failure of these initiatives in order to know if this new MOOCs paradigm is appropriate for different learning situations. This work aims at analyzing and reporting whether the introduction of MOOCs technology was good or not in a case study with the Khan Academy platform at our university with students in a remedial Physics course in engineering education. Results show that students improved their grades significantly when using MOOCs technology, student satisfaction was high regarding the experience and for most of the different provided features, and there were good levels of interaction with the platform (e.g., number of completed videos or proficient exercises), and also the activity distribution for the different topics and types of activities was appropriate. ß 2016 Wiley Periodicals, Inc. Comput Appl Eng Educ 25:15-25, 2017; View this article online at wileyonlinelibrary.com/ journal/cae;