A2E 2022 Abstracts


Area 1 - Analytics in Educational Environments 

Full Papers
Paper Nr: 3
Title:

Assess Performance Prediction Systems: Beyond Precision Indicators

Authors:

Amal Ben Soussia, Chahrazed Labba, Azim Roussanaly and Anne Boyer

Abstract: The high failure rate is a major concern in distance online education. In recent years, Performance Prediction Systems (PPS) based on different analytical methods have been proposed to predict at-risk of failure learners. One of the main studied characteristics of these systems is its ability to provide accurate early predictions. However, these systems are usually assessed using a set of evaluation measures (e.g. accuracy, precision) that do not reflect the precocity, continuity and evolution of the predictions over time. In this paper, we propose to enrich the existing indicators with time-dependent ones including earliness and stability. Further, we use the Harmonic Mean to illustrate the trade-off between the predictions earliness and the accuracy. In order to validate the relevance of our indicators, we used them to compare four different PPS for predicting at-risk of failure learners. These systems are applied on real data of K-12 learners enrolled in an online physics-chemistry module.
Download

Paper Nr: 4
Title:

Examining the Effectiveness of Different Assessments and Forecasts for Accurate Judgments of Learning in Engineering Classes

Authors:

Christopher Cischke and Shane T. Mueller

Abstract: Research and anecdotal evidence suggests that students are generally poor at predicting or forecasting how they will do at high-stakes testing in final exams. We hypothesize that better judgments of learning may allow students to choose more efficient study habits and practices that will improve their learning and their test scores. In order to inform interventions that might provide better forecasts, we examined student data from several university engineering courses. We examined how well three types of assessments predict final exam performance: performance on ungraded exercises; student forecasts prior to exams, and performance on graded material prior to exams. Results showed that ungraded exercises provided poor forecasts of exam performance, student predictions provided marginal forecasts, as did prior graded work in the course. The best predictor was found to be performance on previous high-stakes exams (i.e., midterms), whether or not these exams covered the same material as the later exam. Results suggest that interventions based on prior exam results may help students generate a better and more accurate forecast of their final exam scores.
Download

Paper Nr: 6
Title:

Design Guidelines for a Team Formation and Analytics Software

Authors:

Bowen Hui, Opey Adeyemi, Mathew de Vin, Callum Takasaka and Brianna Marshinew

Abstract: Many researchers over the past several decades studied the success factors of a team. Despite much research efforts, there is still no consensus on how a team should ideally be formed. Consequently, how one decides to form teams in a class depends on the domain, classroom context, and pedagogical objectives. Therefore, software used to support an instructor in forming teams must be flexible enough to accommodate a variety of use case scenarios. In this work, we review the general team formation process and summarize our development efforts in building a team formation and analytics software over the past four years. We found two advantages of using learning analytics in our software: (i) to gain the user’s trust in the system and (ii) to help the user assess whether the suggested teams are balanced and which modifications should be made if any. Based on our experience, we present design recommendations for developing team formation software that reveals challenges and opportunities, especially in combination with learning analytics.
Download

Short Papers
Paper Nr: 1
Title:

Modelling the Effect of Academic Performance on National Achievement Test (NAT)

Authors:

Nathalie Joy G. Casildo

Abstract: The title of this study is Predicting the Effect of Academic Performance on the National Achievement Test Using Data Mining. Students from Central Mindanao University's Senior High School (SHS) provided the data for this study. The goal of this study is to develop a model that can predict how academic performance affects the National Achievement Test. Its specific objectives are to extract predictive features of subjects that affect the National Achievement Test (NAT), determine the effects of various academic subjects on the National Achievement Test, identify the subjects that have a strong effect on the National Achievement Test, and finally determine whether academic performance per quarter affects the National Achievement Test.
Download

Paper Nr: 2
Title:

Teachers' and Students' Perception Regarding the Use of Moodle

Authors:

Douglas Legramante, Ana Azevedo and José Manuel Azevedo

Abstract: This paper seeks to analyse the perceptions of students and teachers regarding the factors that influence satisfaction and the intention to continue using Moodle. An approach that integrates DeLone and McLean's Information Systems Success Model to Davis' Technology Acceptance Model is used. The two models are widely used in research related to the context of e-learning. A quantitative methodological approach was assumed, based on the post-positivist paradigm. Data collection was carried out employing a self-administered questionnaire developed in Google Forms. Descriptive analysis techniques were applied for the data analysis of 144 valid questionnaires. The results showed that teachers and students have a positive perception of the ease of use and usefulness of Moodle, besides evidencing those users are satisfied and intend to continue using Moodle. This research contributes to the formation of knowledge about the perception of Moodle users as support for classroom teaching, providing helpful information for educational institutions, researchers, managers, administrators, and designers of e-learning systems.
Download

Paper Nr: 5
Title:

A Moodle Component for Data Visualization with Applied Learning Analytics for Students

Authors:

M. Barbachano-Chiu, V. Menéndez-Domínguez and L. Curi-Quintal

Abstract: One of the main characteristics of e-learning platforms, such as Moodle, is the registration, monitoring, and storage of the interaction of its users with the published resources, access, and times of entry and exit to them. This information certainly occupies a lot of space and, since its analysis requires prior knowledge by teachers, this information is often forgotten and even eliminated. For this, through Learning Analytics, whose main objective is to apply an intelligent use to data produced by students to predict, evaluate and optimize their learning, we have proposed the development of a Moodle component that allows, with the help of data visualization, to present different results obtained through different learning analytics techniques to evaluate the academic performance of a student or a group enrolled in a course through parameterization, visual and quantitative support by the tool.
Download