<p>Social presence positively influences the motivation, satisfaction, retention, and learning outcomes of online students. Although it is crucial for successful online learning experiences, little work has thoroughly examined the evolution of social presence over time and the influence of social presence on peer interaction. In other words, if social presence can be learned by interacting with others. This three-article dissertation study elucidates this gap by answering the overarching question: How does online students' social presence evolve over time to shape their online learning behaviors? Using stochastic-actor oriented models to reflect the dependence among learners in online collaborative learning communities, this dissertation investigated how learners' social presence evolved in learner-learner interaction resulting in two empirical studies and one conceptual framework. The first study explored social presence through clickstream interaction (e.g., number of replies received/sent in an online discussion) of 382 learners enrolled in a Massive Open Online Course. Three key findings from the study were: 1) dropout rates could be lowered if social presence affordances are used purposefully; 2) adding social media characteristics to online discussion boards, for example, "like" buttons, inhibits conversational behavior, and eventually, decreases achievement of learning outcomes; and 3) the "rich-get-richer" effect also applies to social presence, reinforcing highly active students' behavior and risking inactive online students to experience isolation. The second study used peer-nomination data (i.e., asking students who they interact with) and a scale to investigate the spread of social presence perceptions in online networks of students over three consecutive courses (n = 197). Although there was no evidence of social influence, online learners who nominated more peers are more likely to report higher social presence perceptions over time. Students were not more likely to share with those who showed similar levels of social presence. The "rich-get-richer effect" was observed in the incoming nominations of learners. The third study is a conceptual framework that integrates network theory and the online learning literature into a new perspective to analyze learners' online behaviors and interactions under the light of social presence theory. The proposed framework includes four main steps: 1) interaction, 2) social presence alignment, 3) unit of analysis definition, and 3) network statistics and inferential analysis selection. The findings of this dissertation improve educational practice by identifying behaviors that harm online social presence and providing specific actions for online instructors and instructional designers to promote social presence in online learning.  </p>

Recent research have been focusing on supporting teachers in the classroom. Such support has been shown to benefit from the development and employment of teacher-facing analytic dashboards to help them to make fast and effective decisions in regard to the in-class student learning activities. The evolving interest in this field has facilitated the emergence of the Teaching Analytics area of practice and research. However, current research efforts have indicated that the use of such dashboards usually adds another layer to the already dynamic and complex situation for teachers, which can divert their attention and can often be experienced as a disturbing factor in the class. Therefore, it is highly important to examine how such teacher-facing dashboards can be improved from the user experience perspective, in a way that would allow teachers to grasp student learning activities easily and with good perceived usability. The aim of this study is to understand how we can better design teacher-facing dashboards to more adequately support K-12 teachers in their decisions that would provide relevant in-time and student support. The study applies Nielsen's three-round iterative design approach to understand the existing usability problems and further develop the dashboard, originally designed by the company. In order to investigate users’ perceived attitude towards the redesigned dashboard, the final prototype has been evaluated through a Technology Acceptance Model questionnaire and semi-structured interviews with nine participants. As a result, the redesigned teacher-faced dashboard was proven to have a high potential to support teachers’ decisions. The efficiency of the Technology Acceptance Model was verified and put into general context on how tools for teachers should be designed for the usage in the classroom. Additionally, some major challenges for teachers with using external tools during class were discovered and are discussed in the context of a newly designed dashboard. / Befintlig forskning stödjer lärare i klassrummet genom att utveckla analytiska visualiseringsverktyg (a.k.a. dashboards) som lärare kan använda för att fatta snabba och effektiva beslut med avseende på elevernas läraktiviteter. Det växande intresset för detta område har lett till framväxten av Teaching Analytics-fältet inom praktik och forskning. Forskning har dock visat att användandet av dessa verktyg vanligtvis lägger till ytterligare ett lager till den redan dynamiska och komplexa situationen för lärare, vilket kan avleda deras uppmärksamhet och ofta fungera som en störande faktor i klassrummet. Därför är det mycket viktigt att undersöka hur sådana visualiseringsverktyg för lärare kan förbättras ur användarperspektiv, på ett sätt som skulle göra det möjligt för lärare att förstå elevernas läraktiviteter enkelt och med god upplevd användbarhet. Syftet med denna studie är att förbättra användargränssnittet för ett befintligt, så att det på ett mer adekvat sätt kan stödja lärare i sina beslut och erbjuda relevant stöd till eleverna. Studien tillämpar Nielsens tre-rundors iterativa designmetod för att förstå de befintliga användbarhetsproblemen och vidareutveckla en existerande dashboard, ursprungligen utvecklad av företaget. För att undersöka användarnas inställning till det omdesignade verktyget har den slutliga prototypen utvärderats genom ett frågeformulär och semistrukturerade intervjuer med nio deltagare. Resultat visar att det omdesignade de verktyget har en stor potential för att stödja lärarnas beslut i klassrummet. Effektiviteten för Teknik Acceptant Modellen (TAM) verifieras och sattes i allmän kontext för hur olika verktyg för lärare bör utformas för användning i klassrummet. Dessutom diskuteras lärarnas stora utmaningar med att använda externa verktyg under lektioner i samband med ny verktyget.

A Digital Education: AI A.1:1 Learning Analytics in der Hochschullehre 2 Das Projekt ‚Learning Analytics – Studierende im Fokus‘ 3 Herausforderungen und Chancen der Integration von Learning Analytics in die Lehre 4 Zusammenfassung und Ausblick / Bereits seit Mitte der 1990er Jahre hält Digitalisierung Eingang in den Bereich der Hochschullehre, wobei Bildungstechnologien wie Artifcial Intelligence, Blended and Hybrid Course Models, Microcredentialing, Open Educational Resources, Quality Online Learning und Learning Analytics eine Vielzahl an Einsatz- und Nutzungsmöglichkeiten für innovative Lehr- und Lernszenarien bieten. Diese Möglichkeiten werden begleitet von Fragen im Hinblick auf die pädagogische und didaktische Verwendung digitaler Elemente in der Lehre.:1 Learning Analytics in der Hochschullehre 2 Das Projekt ‚Learning Analytics – Studierende im Fokus‘ 3 Herausforderungen und Chancen der Integration von Learning Analytics in die Lehre 4 Zusammenfassung und Ausblick

Diese Studie untersucht umfassend Barrieren bei der Veröffentlichung von Forschungsdaten im Bereich Learning Analytics (LA) mithilfe eines Mixed-Methods-Ansatzes. Methodologisch gegliedert in vier Phasen – Systematic Literature Review (SLR), Leitfrageninterviews, eine weltweite Online-Umfrage und adaptive Workshops – zeigt die Forschung eine Vielzahl interdisziplinärer und internationaler Perspektiven auf. Das SLR bildet die Grundlage, indem es rechtliche, ethische und ressourcenbezogene Hindernisse für die Datenveröffentlichung identifiziert. Durch die Integration dieser Erkenntnisse in Interviews zeigt sich ein vertieftes Verständnis kultureller und institutioneller Unterschiede, die die Datenpublikation beeinflussen. Eine globale Umfrage verdeutlicht zudem eine Diskrepanz zwischen der Bereitschaft von Forschenden, Daten zu teilen, und ihrer Bewertung der Vorteile geteilten Wissens. Dies weist auf Vertrauensthemen und den geringen wahrgenommenen Nutzen gemeinsamer Daten in der Forschung hin, trotz zunehmender Infrastrukturen und Förderungen für Open Data. Adaptive Workshops beleuchten die Lücke zwischen der Anerkennung der Bedeutung von Datenfreigabe und der Fähigkeit der Forschenden, diese effektiv umzusetzen. Insbesondere Datenschutzbedenken, etwa zur DSGVO, und der Verlust von Kontrolle über geteilte Daten erweisen sich als große Hürden. Die Ergebnisse dieser Studie verdeutlichen, wie Barrieren der Datenpublikation je nach Disziplin und Region variieren und tief in kulturellen und institutionellen Rahmen eingebettet sind. / This study investigates barriers to research data publication in Learning Analytics (LA) through a mixed-method approach encompassing a Systematic Literature Review (SLR), semi-structured interviews, a global survey, and adaptive workshops. The SLR establishes a foundation by identifying legal, ethical, and resource-related barriers to data publication across disciplines. Findings from the SLR integrate in the subsequent interviews, which reveal cultural and institutional nuances affecting researchers' motivations and capabilities for data sharing. A global survey uncovers a discrepancy between researchers' willingness to share data and their perceived benefits from accessing others' data, highlighting trust issues within the scientific community despite growing support for open data. Adaptive workshops underscore the gap between researchers' recognition of data sharing importance and their practical ability to implement it, with data protection concerns, particularly related to GDPR compliance, emerging as major barriers alongside fears of losing data control. The findings from this study illustrate how barriers to data publication vary by discipline and region, being deeply embedded within cultural and institutional frameworks.

Submitted by William Justo Figueiro (williamjf) on 2015-07-28T20:32:18Z No. of bitstreams: 1 27d.pdf: 4180188 bytes, checksum: 617cda1d8cedaa910bb66676e96c33d1 (MD5) / Made available in DSpace on 2015-07-28T20:32:18Z (GMT). No. of bitstreams: 1 27d.pdf: 4180188 bytes, checksum: 617cda1d8cedaa910bb66676e96c33d1 (MD5) Previous issue date: 2014-04-15 / Nenhuma / Aplicações que fazem uso de tecnologias como Mineração de Dados Educacionais (MDE) e Learning Analytics (LA) vêm sendo adotadas na mitigação da evasão escolar, disponibilizando informações sobre os alunos que são utilizadas em intervenções pedagógicas. Os trabalhos estudados sobre a implementação destas aplicações priorizam a descrição das técnicas empregadas e existem poucas avaliações da sua utilização em larga escala, além da falta de detalhamento sobre as causas da evasão. Este trabalho apresenta um estudo de fatores envolvidos no fenô- meno de evasão escolar e descreve a utilização de um sistema para MDE e LA durante 18 meses em cursos de graduação na modalidade de Educação a Distância. É ampliada a análise dos fatores tradicionalmente monitorados e utilizados nos sistemas de MDA e LA, com a inclusão de elementos associados ao papel exercido pelos docentes e pelo conjunto de aspectos metodológicos de cada instituição. O sistema possui como diferencial a flexibilidade na integração e utilização dos dados gerados no processo de mediação digital, o que permite que necessidades de diferentes ferramentas de apoio sejam disponibilizadas. Resultados positivos destacados são a identificação de perfis de alunos evasores e a realização de intervenções pedagógicas, com redução das médias da evasão. / Educational Data mining (EDM) and Learning Analytics (LA) applications have been adopted in mitigation of dropout, providing information about students who are employed in pedagogical interventions. The most papers about the implementation of these systems describe the techniques employed, there are few evaluations of their large-scale use, apart from the lack of detail about the causes of dropout. This work presents a study of factors involved in dropout and describes the use of a system for EDM and LA during 18 months for undergraduate courses in distance education. The analysis of the factors traditionally monitored and used in EDM and LA systems is extended, with the inclusion of elements associated with the role exercised by the teachers and by institutional methodological aspects. The system has flexibility in integration and use of data generated in the process of digital mediation, which allows different support tools to be available. Some results are the identification of evaders students profiles and the realization of pedagogical actions with reducing evasion.

abstract: With the advent of Massive Open Online Courses (MOOCs) educators have the opportunity to collect data from students and use it to derive insightful information about the students. Specifically, for programming based courses the ability to identify the specific areas or topics that need more attention from the students can be of immense help. But the majority of traditional, non-virtual classes lack the ability to uncover such information that can serve as a feedback to the effectiveness of teaching. In majority of the schools paper exams and assignments provide the only form of assessment to measure the success of the students in achieving the course objectives. The overall grade obtained in paper exams and assignments need not present a complete picture of a student’s strengths and weaknesses. In part, this can be addressed by incorporating research-based technology into the classrooms to obtain real-time updates on students' progress. But introducing technology to provide real-time, class-wide engagement involves a considerable investment both academically and financially. This prevents the adoption of such technology thereby preventing the ideal, technology-enabled classrooms. With increasing class sizes, it is becoming impossible for teachers to keep a persistent track of their students progress and to provide personalized feedback. What if we can we provide technology support without adding more burden to the existing pedagogical approach? How can we enable semantic enrichment of exams that can translate to students' understanding of the topics taught in the class? Can we provide feedback to students that goes beyond only numbers and reveal areas that need their focus. In this research I focus on bringing the capability of conducting insightful analysis to paper exams with a less intrusive learning analytics approach that taps into the generic classrooms with minimum technology introduction. Specifically, the work focuses on automatic indexing of programming exam questions with ontological semantics. The thesis also focuses on designing and evaluating a novel semantic visual analytics suite for in-depth course monitoring. By visualizing the semantic information to illustrate the areas that need a student’s focus and enable teachers to visualize class level progress, the system provides a richer feedback to both sides for improvement. / Dissertation/Thesis / Masters Thesis Computer Science 2016

abstract: While predicting completion in Massive Open Online Courses (MOOCs) has been an active area of research in recent years, predicting completion in self-paced MOOCS, the fastest growing segment of open online courses, has largely been ignored. Using learning analytics and educational data mining techniques, this study examined data generated by over 4,600 individuals working in a self-paced, open enrollment college algebra MOOC over a period of eight months. Although just 4% of these students completed the course, models were developed that could predict correctly nearly 80% of the time which students would complete the course and which would not, based on each student’s first day of work in the online course. Logistic regression was used as the primary tool to predict completion and focused on variables associated with self-regulated learning (SRL) and demographic variables available from survey information gathered as students begin edX courses (the MOOC platform employed). The strongest SRL predictor was the amount of time students spent in the course on their first day. The number of math skills obtained the first day and the pace at which these skills were gained were also predictors, although pace was negatively correlated with completion. Prediction models using only SRL data obtained on the first day in the course correctly predicted course completion 70% of the time, whereas models based on first-day SRL and demographic data made correct predictions 79% of the time. / Dissertation/Thesis / Doctoral Dissertation Educational Technology 2017

Submitted by Maria Medeiros (maria.dilva1@ufcg.edu.br) on 2018-05-24T14:28:20Z No. of bitstreams: 1 ALLAN SALES DA COSTA MELO - DISSERTAÇÃO (PPGCC) 2016.pdf: 1339540 bytes, checksum: b3bbc96d990dd77ee344c4f5434a4d00 (MD5) / Made available in DSpace on 2018-05-24T14:28:20Z (GMT). No. of bitstreams: 1 ALLAN SALES DA COSTA MELO - DISSERTAÇÃO (PPGCC) 2016.pdf: 1339540 bytes, checksum: b3bbc96d990dd77ee344c4f5434a4d00 (MD5) Previous issue date: 2016 / A evasão estudantil é uma das maiores preocupações dos institutos de ensino superior brasileiros já que ela pode ser uma das causas de desperdício de recursos da Universidade. A previsão dos estudantes com alta probabilidade de evasão, assim como o entendimento das causas que os levaram a evadir, são fatores cruciais para a definição mais efetiva de ações preventivas para o problema. Nesta dissertação, o problema da detecção de evasão foi abordado como um problema de aprendizagem de máquina supervisionada. Utilizou-se uma amostra de registros acadêmicos de estudantes considerando-se todos os 76 cursos da Universidade Federal de Campina Grande com o objetivo de obter e selecionar atributos informativos para os modelos de classificação e foram criados dois tipos de modelos, um que separa os estudantes por cursos e outro que não faz distinção de cursos. Os dois modelos criados foram comparados e pôde-se concluir que não fazer distinção de alunos por curso resulta em melhores resultados que fazer distinção de alunos por curso. / Students’ dropout is a major concern of the Brazilian higher education institutions as it may cause waste of resources. The early detection of students with high probability of dropping out, as well as understanding the underlying causes, are crucial for defining more effective actions toward preventing this problem. In this paper, we cast the dropout detection problem as a supervised learning problem. We use a large sample of academic records of students across 76 courses from a public university in Brazil in order to derive and select informative features for the employed classifiers. We create two classification models that either consider the course to which the target student is formally committed or not consider it, respectively. We contrast both models and show that not considering the course leads to better results.

Social Academic Analytics (SAA) is proposed as a new scientific approach toward developing suitable instruments to promote virtual collaboration among participants in the higher education field. SAA refers to the process of extracting relational data for the purpose of exploring organizational structures within virtual learning organizations and knowledge networks. Implementation of SAA provides opportunities for organizers and instructors to optimize socio-technological infrastructures within (virtual) knowledge networks so as to encourage collaborative work, while offering significant potential for quality assurance. SAA combines theories and models from both informatics and the social sciences at the macro level in order to formulate data analysis for the field of (web-based) educational research. In this paper we introduce SAA and its constituent activities. Finally we select case studies and applications to compare analytical concepts from diverse disciplines and conclude with further suggestions as to how SAA concepts can be applied in educational data management.

Twenty percent of the 2013-2014 sophomore class at a Washington high school was failing high-stakes tests, making these students ineligible to graduate. In an attempt to help students identify their academic proficiency with respect to the Common Core Curricular Standards 9 months before the high-stakes exam, the high school recently introduced the adaptive diagnostic software i-Ready. Cognitive learning theories comprised the framework for this study, which posit that learning is dependent on previous knowledge and central to measuring performance levels. The purpose of this quantitative correlational project study was to examine whether 10th grade students' achievement on i-Ready math scores (N = 220) could predict the subsequent high-stakes mathematics scores on the End of Course Exam while controlling for gender, ethnicity, and socioeconomic status. The i-Ready emerged as a statistically significant predictor of the End of Course Exam scores with β = .64 (p < .001), explaining R2 = .43 of the criterion variance. Gender, ethnicity, and socioeconomic status had no significant moderating influence. The project deliverable as a result of this study was a position paper advising the use of the i-Ready as a predictor for the End of Course Exam at the high school under study. The implications for positive social change include allowing educators to use the i-Ready as an early warning system for students in danger of failing high-stakes exams. This study may help identify students at risk of not graduating who could benefit from instructional support.