Digital Assistance for Goal-Setting and Goal Pursuit in Higher Education

Please use this identifier to cite or link to this item:
Open Access logo originally created by the Public Library of Science (PLoS)
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorDr. Tobias Thelenger
dc.creatorWeber, Felix-
dc.description.abstractThis doctoral thesis investigates how data and AI-driven digital assistants can support university students in goal setting, goal pursuit, and achievement. The first chapter introduces contextual information about higher education, human-machine interaction, self-regulated learning, digital study assistants, and constructivism. The first chapter also clarifies the aims and motivation, states the main research questions, and concludes with an outlook on the content and structure of the text. In the second chapter, goals are introduced as a concept in the Cognitive Sciences, ranging from motor control, human-machine interaction, AI algorithms, planning, games, navigation, and human motivation. The second chapter also disambiguates the goal construct from related terms and constructs. The third chapter summarizes two approaches to measuring goal characteristics that have been taken during the dissertation research: An external approach was based on extensive tagging of goals by six raters, while an internal approach was based on self-assessment with a Likert-scale questionnaire. The fourth chapter centers around the Hierarchical Goal Systems (HGS) concept. It starts with a theoretical foundation, including a review of hierarchical goal structures in the literature, formal and functional definitions of Hierarchical Goal Systems, and potential advantages and disadvantages of such representations. The central part of chapter four describes the development process and a row of formative field studies with a hierarchical goal-setting assistant called 'GoalTrees', publicly available as open-source software under an MIT license. In the productive database of the field study server, a significant amount of hierarchical goal system data and goal characteristics scores has been accumulated. Chapter five outlines how this data can be utilized to reproduce previous findings and increase knowledge about goal types, based on the theoretical concept of 'Conceptual Spaces', combined with goal data in goal characteristics space. Clustering in a high-dimensional conceptual space of goal characteristics can potentially work as a data-driven, bottom-up process in the proposed approach. Chapter six summarizes the findings and insights from this line of research on an ontological and epistemological level, reflects on the applied methods and scientific practice, and concludes with an outlook on future research and possible next steps. Due to the high interaction costs for users to answer questionnaires to measure goal characteristics, a reliable prediction procedure for characteristics based on goal formulations in natural language, for instance, a pre-trained and fine-tuned BERT neural network, could significantly improve the usability and user experience of 'GoalTrees' in the future.ger
dc.rightsAttribution-NonCommercial-NoDerivs 3.0 Germany*
dc.subjectDigital Study Assistantseng
dc.subjectArtificial Intelligenceeng
dc.subjectConceptual Spaceseng
dc.subject.ddc150 - Psychologieger
dc.subject.ddc500 - Naturwissenschaftenger
dc.subject.ddc004 - Informatikger
dc.titleDigital Assistance for Goal-Setting and Goal Pursuit in Higher Educationeng
dc.typeDissertation oder Habilitation [doctoralThesis]-
thesis.typeDissertation [thesis.doctoral]-
dc.contributor.refereeProf. Dr. Kai-Uwe Kühnbergerger
dc.contributor.refereeJun.-Prof. Dr. Maria Wirzbergerger
Appears in Collections:FB08 - E-Dissertationen

Files in This Item:
File Description SizeFormat 
thesis_weber.pdfPräsentationsformat35,68 MBAdobe PDF

This item is licensed under a Creative Commons License Creative Commons