Investigación

Actas de la European MOOC Stakeholder Summit 2017, incluyendo los artículos presentados en las sesiones de investigación y experiencias.

lytics_banner

El propósito del laboratorio Lytics de Stanford está enfocado en una dimensión del entorno de aprendizaje interactivo o infraestructura tecnológica de los MOOCs.

Carlos Alario Hoyos, Iria Manuela Estévez Ayres, Carlos Delgado Kloos Árbol académico, Julio Villena Román, Pedro J. Muñoz Merino, Enrique Llorente Pérez

MOOCs have made it possiblenot only to provide quality open education for any learnerworldwide, but also to rethink theway on-campusteachingis delivered. Thematerials produced for a MOOC canbeconsumed byon-campusstudents beforearriving to the classroom, using class time to do activities that promote active learning, following this way a flippedclassroom strategy. This paper presents the experience of redesigning a first-year engineering course with a large number of students (over 400 each year), in which MOOCs are reused, and a flipped classroom strategy is implemented, dedicatingmost of traditional lecture time to do hands-on, interactive activities. The results show an increase in students’ motivation,both in the use of MOOC content outside the classroom, and in the realization of hands-on, interactive activities inside theclassroom. In relation to the teacher, having information on students’ previous work outside the classroom, and onstudents’ work in the hands-on, interactive activities carried out inside the classroom, allows understanding better thedifferences between groups, tailoring the explanations during class time, and providing proper reinforcement activities tobe done after class.
 

M. Elena Alonso-Mencía, Carlos Alario-Hoyos, Jorge Maldonado-Mahauad, Iria Estévez-Ayres, Mar Pérez-Sanagustín & Carlos Delgado Kloos

Learners in massive open online courses (MOOCs) are required to be autonomous during their learning process, and thus they need to self-regulate their learning to achieve their goals. According to existing literature, self-regulated learning (SRL) research in MOOCs is still scarce. More studies which build on past works regarding SRL in MOOCs are required, as well as literature reviews that help to identify the main challenges and future research directions in relation to this area. In this paper, the authors present the results of a systematic literature review on SRL in MOOCs, covering all the related papers published until the end of 2017. The papers considered in this review include real experiences with at least a MOOC (other learning scenarios sometimes claimed as MOOCs, such as blended courses, or online courses with access restrictions, are out of the scope of this analysis). Most studies on SRL in MOOCs share some common features: they are generally exploratory, based on one single MOOC and tend not to specify in which SRL model they are grounded. The results reveal that high self-regulators engage in non-linear navigation and approach MOOCs as an informal learning opportunity. In general, they prefer setting specific goals based on knowledge development and control their learning through assignments.
 

Pedro Manuel Moreno-Marcos, Carlos Alario-Hoyos, Pedro J Munoz-Merino, Iria Estevez-Ayres, Carlos Delgado Kloos

One of the characteristics of MOOCs (Massive Open Online Courses) is that the overall number of social interactions tend to be higher than in traditional courses, hindering the analysis of social learning. Learners typically ask or answer questions using the forum. This makes messages a rich source of information, which can be used to infer learners behaviour and outcomes. It is not feasible for teachers to process all forum messages and automated tools and analysis are required. Although there are some tools for analysing learners interactions, there is a need for methodologies and integrated tools that help to interpret the learning process based on social interactions in the forum. This work presents the 3S (Social, Sentiments, Skills) learning analytics methodology for analysing forum interactions in MOOCs. This methodology considers a temporal analysis combining the social, sentiments and skill dimensions …
 

Armin Weinberger, Carlos Alario-Hoyos, Poline Bala, Dennis Batangan, Carlos Delgado Kloos, Narayanan Kulathuramaiyer, John Carlo Navera, Josenh Palis, Alwin Melkie Sambul, Peter Sy, Tat-Chee Wan

This Paper explores the current use of Massive Open Online Courses (MOOCs) as a means of educational outreach among identified remote populations in Southeast Asia. Often excluded from traditional educational outreach, these groups are targeted through the COMPETEN-SEA Project, a Capacity Building in Higher Education project funded by the Erasmus+ programme of the European Commission and implemented in partnership with European and Southeast Asian universities. It is hoped that the Project will aid participating Southeast Asian countries address societal needs and attain national development goals.
 

Cristina Catalán Aguirre, Carlos Delgado Kloos, Carlos Alario-Hoyos, Pedro J Muñoz-Merino

This paper presents an initial design of a conversational agent for educational purposes built for Google Assistant and its first prototype. Recent studies suggest that people will get more and more attached to voice assistant because they can easily use technology without being forced to learn it. Speech recognition might facilitate a more efficient work environment without being overly rigid and overly domineering. Modern frameworks allow harnessing natural language understanding as well a machine learning tools, thereby making it easy to build conversational agents. In this paper, we present first design decisions and a prototype for building an agent for learning Java that complements a MOOC for programming with Java.
 

Jorge Maldonado-Mahauad, Mar Pérez-Sanagustín, Pedro Manuel Moreno-Marcos, Carlos Alario-Hoyos, Pedro J Muñoz-Merino, Carlos Delgado-Kloos

In the past years, predictive models in Massive Open Online Courses (MOOCs) have focused on forecasting learners’ success through their grades. The prediction of these grades is useful to identify problems that might lead to dropouts. However, most models in prior work predict categorical and continuous variables using low-level data. This paper contributes to extend current predictive models in the literature by considering coarse-grained variables related to Self-Regulated Learning (SRL). That is, using learners’ self-reported SRL strategies and MOOC activity sequence patterns as predictors. Lineal and logistic regression modelling were used as a first approach of prediction with data collected from N = 2,035 learners who took a self-paced MOOC in Coursera. We identified two groups of learners: (1) Comprehensive, who follow the course path designed by the teacher; and (2) Targeting, who seek …
 

Pedro Manuel Moreno-Marcos, Carlos Alario-Hoyos, Pedro J Muñoz-Merino, Carlos Delgado Kloos

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 dropouts 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 …
 

Christian M Stracke, Rocael Hernández, Carlos Delgado Kloos, Mar Pérez Sanagustín, António Moreira Teixeira

Presentation at OE Global 2018, Delft, The Netherlands, by Stracke, C. M., et al. (2018, 24 April) on: “How to make MOOCs better for specific target groups and developing countries?”
 

Pedro Manuel Moreno-Marcos, Carlos Alario-Hoyos, Pedro J Muñoz-Merino, Iria Estévez-Ayres, Carlos Delgado Kloos

Forum messages in MOOCs (Massive Open Online Courses) are the most important source of information about the social interactions happening in these courses. Forum messages can be analyzed to detect patterns and learners’ behaviors. Particularly, sentiment analysis (e.g., classification in positive and negative messages) can be used as a first step for identifying complex emotions, such as excitement, frustration or boredom. The aim of this work is to compare different machine learning algorithms for sentiment analysis, using a real case study to check how the results can provide information about learners’ emotions or patterns in the MOOC. Both supervised and unsupervised (lexicon-based) algorithms were used for the sentiment analysis. The best approaches found were Random Forest and one lexicon based method, which used dictionaries of words. The analysis of the case study also showed an …
 

Carlos Alario-Hoyos, Iria Estévez-Ayres, Jesús M Gallego-Romero, Carlos Delgado Kloos, Carmen Fernández-Panadero, Raquel M Crespo-García, Florina Almenares, María Blanca Ibáñez, Julio Villena-Román, Jorge Ruiz-Magaña, Jorge Blasco

Many MOOCs are being designed replicating traditional passive teaching approaches but using video lectures as the means of transmitting information. However, it is well known that learning-by-doing increases retention rates and, thus, allows achieving a more effective learning. To this end, it is worth exploring which tools fit best in the context of each MOOC to enrich learners’ experience, including built-in tools already available in the MOOC platform, and third-party external tools which can be integrated in the MOOC platform. This paper presents an example of the integration of a software development tool, called Codeboard, in three MOOCs which serve as an introduction to programming with Java. We analyze the effect this tool has on learners’ interaction and engagement when running the MOOCs in synchronous (instructor-paced) or asynchronous (self-paced) modes. Results show that the overall use of the …
 

José A Ruipérez-Valiente, Pedro J Muñoz-Merino, José A Gascón-Pinedo, C Delgado Kloos

The emergence of massive open online courses (MOOCs) has caused a major impact on online education. However, learning analytics support for MOOCs still needs to improve to fulfill requirements of instructors and students. In addition, MOOCs pose challenges for learning analytics tools due to the number of learners, such as scalability in terms of computing time and visualizations. In this work, we present different visualizations of our “Add-on of the learNing AnaLYtics Support for open Edx” (ANALYSE), which is a learning analytics tool that we have designed and implemented for Open edX, based on MOOC features, teacher feedback, and pedagogical foundations. In addition, we provide a technical solution that addresses scalability at two levels: first, in terms of performance scalability, where we propose an architecture for handling massive amounts of data within educational settings; and, second, regarding.
 

Alario-Hoyos, C., Estévez-Ayres, I., Delgado Kloos, C., Villena-Román, J.

The concept of SPOCs (Small Private Online Courses) emerged as a way of describing the reuse of MOOCs (Massive Open Online Courses) for complementing traditional on-campus teaching. But SPOCs can also drive an entire methodological change to make a better use of face-to-face time between students and teachers in the classroom. This paper presents the redesign and evaluation of a first-year programming course in several engineering degrees, with over 400 students overall, through the reuse of MOOCs as SPOCs on campus, combined with a flipped classroom strategy aimed at promoting active learning. Results from a students’ self-reported questionnaire show a very positive acceptance of the SPOC, which includes both videos and complementary formative activities, and an increase of motivation through the combination of the SPOC and activities implemented in lectures to flip the classroom.

Ruipérez-Valiente, J. A., Muñoz-Merino, P. J., Gascón-Pinedo, J. A., & Kloos, C. D.

The emergence of massive open online courses (MOOCs) has caused a major impact on online education. However, learning analytics support for MOOCs still needs to improve to fulfill requirements of instructors and students. In addition, MOOCs pose challenges for learning analytics tools due to the number of learners, such as scalability in terms of computing time and visualizations. In this work, we present different visualizations of our “Add-on of the learNing AnaLYtics Support for open Edx” (ANALYSE), which is a learning analytics tool that we have designed and implemented for Open edX, based on MOOC features, teacher feedback, and pedagogical foundations. In addition, we provide a technical solution that addresses scalability at two levels: first, in terms of performance scalability, where we propose an architecture for handling massive amounts of data within educational settings; and, second, regarding the representation of visualizations under massiveness conditions, as well as advice on color usage and plot types. Finally, we provide some examples on how to use these visualizations to evaluate student performance and detect problems in resources.

Ruipérez-Valiente, J. A., Muñoz-Merino, P. J., Kloos, C. D., Niemann, K., Scheffel, M., & Wolpers, M

Presentación del Profesor Carlos Delgado Kloos en la Jornada de Educación Abierta celebrada el 11 de marzo de 2013 en la Universidad Carlos III de Madrid en el marco de la Open Education Week promovida por el Consorcio OpenCourseWare.