Artificiell intelligens (AI) och i synnerhet maskininlärning (ML) har inom UX-design visat potential att förbättra designprocessen genom att exempelvis identifiera användargrupper från stora datamängder, effektivisera idégenerering och automatisera repetitiva uppgifter. Det råder dock oenighet kring hur tekniken kan integreras i designprocessen. En viktig del av en designers arbete är att konsekvent prioritera användarbehov och därigenom förbättra användartillfredsställelsen. Därför försöker designers empatiskt sätta sig in i användarnas situation genom att identifiera deras behov och noggrant utforska potentiella problemområden. En visualiseringsteknik som ofta används av designers för att utveckla och förstå användarupplevelsen är journey mapping (JM). JM är dock en mycket resurskrävande process då den förutsätter nära samarbete mellan olika avdelningar och team. Trots detta visar forskning att de flesta designers inte integrerar element relaterade till insikter i sina kartor, vilka ofta är avgörande för att omvandla JM från visuell berättelse till handlingsplan. Integreringen av ML i JM-processen skulle potentiellt kunna möjliggöra en snabbare, mer datadriven och anpassningsbar designprocess som fokuserar mer på användarnas behov och önskemål. Utmaningen ligger i att smidigt integrera tekniken utan att förlora mänskliga perspektiv och tolkningar, vilka är grundläggande för användarcentrerad design. Denna studie fokuserar på att utforska sätt att integrera ML i delar av designprocessen för att möta utmaningarna som uppstår vid strävan efter användarcentrerad design i en resurskrävande miljö. För att uppnå en djupgående och strukturerad förståelse av samspelet mellan designprocessen, empatiskapande samt ML valdes en systematisk litteraturgranskning som den primära datainsamlingsmetoden samt tematisk analys som dataanalysmetod för denna studie. Målet är att utifrån detta resultat presentera praktiska rekommendationer, baserat på sammanställning av befintlig forskning, för hur JM-processen kan omformas och göras mer ML-driven. Genom en omfattande litteratursökning inhämtades data, vilken genomgick kvalitetsbedömning. Därefter genomfördes en inkludering- och exkluderingsprocess i enlighet med förutbestämda kriterier. Detta resulterade i 13 utvalda dokument, från vilka mönster och teman extraherades genom en tematisk analys. Resultatet visade på vikten av samskapande mellan människa och maskin för att möjliggöra en ansvarsfull designprocess. I resultaten presenteras även tekniska möjligheter och dilemman, där ML exempelvis kan automatisera repetitiva uppgifter och möjliggöra kontinuerlig övervakning och utvärdering av användarupplevelsen över tid. Dock kräver detta omfattande och representativa träningsdataset för att konstruera ML-modeller med hög generaliseringsförmåga gentemot nya och komplexa data. Resultaten belyser också bristen på förståelse för ML bland UX-designers och betonar vikten av samarbete med datavetare för att skapa en effektiv designprocess. Utifrån dessa resultat presenteras rekommendationer för en ML-driven JM-process. Även om UX-rollen troligtvis behöver omdefinieras, framhåller resultaten att kulturen inom UX-designprocessen snarare är "dataaktiverad" än helt datadriven. Detta perspektiv understryker det fortsatta behovet av mänsklig intuition och förståelse trots den ökade inriktningen på datadrivna metoder. Slutligen diskuteras hur framtida forskning skulle kunna inrikta sig på att utvärdera rekommendationernas praktiska tillämpbarhet. / Artificial Intelligence (AI), particularly Machine Learning (ML), has demonstrated potential within UX design to enhance the design process by identifying user groups from extensive datasets, streamlining idea generation, and automating repetitive tasks. However, there is ongoing debate about the integration of this technology in the design process. A crucial aspect of a designer's work is consistently prioritizing user needs to enhance user satisfaction. Designers strive to empathetically understand user situations by identifying their needs and meticulously exploring potential problem areas. One commonly used visualization technique employed by designers to develop and understand user experiences is Journey Mapping (JM). However, JM is a resource-intensive process, requiring close collaboration across different departments and teams for extensive data collection. Despite this, research indicates that most designers do not integrate elements related to insights into their maps, crucial for transforming JM from a visual narrative to an actionable plan. Integrating ML into the JM process could potentially enable a faster, more data-driven, and adaptable design process, effectively focusing on user needs and desires. The challenge lies in seamlessly integrating the technology without losing the human perspectives and interpretations fundamental to user-centered design. This study focuses on exploring ways to integrate ML into parts of the design process to address challenges in pursuing user-centered design in a resource-intensive environment. A systematic literature review was chosen as the primary data collection method, with thematic analysis employed as the data analysis method. The aim was to present practical recommendations based on a compilation of existing research on the design process, making the JM process more MLdriven. Through an extensive literature search, data was gathered and subjected to quality assessment. Subsequently, an inclusion and exclusion process were conducted according to predetermined criteria, resulting in 13 selected documents. Patterns and themes were extracted through thematic analysis. The results emphasize the importance of co-creation between humans and machines to enable a responsible design process. Technical possibilities and dilemmas are also presented, highlighting ML's ability to automate repetitive tasks and facilitate continuous monitoring and evaluation of user experience over time. However, this requires comprehensive and representative training datasets to construct ML models with high generalization ability to new and complex data. The results also underscore the lack of ML understanding among UX designers and emphasize the importance of collaboration with data scientists for an efficient design process. Recommendations for an ML-driven JM process are presented based on these findings. While the UX role may need redefinition, the results emphasize a "data-activated" culture within the UX design process rather than being entirely data-driven. This perspective underscores the ongoing need for human intuition and understanding despite the increased focus on data-driven methods. Finally, the discussion explores how future research could evaluate the practical applicability of the recommendations.

Mobile learning (m-learning) is still in its infancy, and great efforts should be made so as to investigate the potentials of an educational paradigm shift from the traditional one-size-fits-all teaching approaches to an adaptive learning that can be delivered via mobile devices. Thus, the next challenge has been identified from this implication: How learning design can be implemented so as to benefit from the m-learning characteristics and achieve adaptation and personalization of the individual learning process in different contexts? An important factor for achieving personalized and adaptive m-learning has been the pedagogically meaningful and technically feasible processing of learners’ contextual information. Therefore in this work, design and delivery of personalized educational scenarios are suggested to be re-thought so as to benefit from the affordances of mobile technologies and the learners’ context / El aprendizaje móvil (m-learning) se encuentra todavía en su infancia y grandes esfuerzos se deben hacer para investigar el cambio de paradigma educativo, desde la forma de enseñanza tradicional de 'un modelo único para todos' a un aprendizaje adaptativo que se pueda entregar a través de dispositivos móviles. De esta manera, el siguiente desafío ha sido identificado por esta implicación: ¿Cómo se puede implementar el diseño instruccional con el fin de beneficiarse de las características del m-learning y lograr la adaptación y personalización del proceso de aprendizaje personal en diferentes contextos? Un importante factor para lograr un m-learning personalizado y adaptable ha sido el procesamiento de la información contextual de los estudiantes. Por lo tanto, en este trabajo se sugiere que sean re-pensados el diseño y la entrega de escenarios educativos orientados a la personalización del aprendizaje y que se beneficien de las potencialidades de las tecnologías móviles y el contexto de los estudiantes

This study investigates the terms, dialogue, structure and active learning in Open Distance Learning texts in aa constructivist and problem-solving approach. In-text activities are imbedded in selfstudy texts. Their construct is validated against appropriate instructional design and learning theories for active learning. A course team developed, implemented and piloted a Unisa course for mathematics teachers in South Africa. The results describe their professional development through action learning. The Tornado-approach affects teams in organisations. A team used participatory action research and action learning with rich descriptions to document the longitudinal case study of four action research cycles over eight years. Qualitative research involved a literature survey, notes, portfolios, mindmaps, letters, interviews, document analyses, and assessment results. Our action research demonstrates ways to implement a post-modern learning design in an industrialized institution. The researcher applies action learning in professional development, in team management, and suggests models for successful/unsuccessful teams (tornado-effect). / Educational Studies / M. Ed. (Didactics)

[Abstract]: Research has suggested that, despite support through policy and resource provision,information and communications technologies (ICTs) have made little impact on the practiceof education and that limited teacher preparation for the use of ICTs represents a partialexplanation. The purpose of this study was to investigate what form of professionaleducation might be effective in preparing pre-service teachers to integrate ICTs into theirteaching. Self-efficacy beliefs about teaching with computers were identified as a potentiallysignificant source of influence on teachers' use of ICTs for teaching. It was proposed thatinteractive multimedia using a problem-based learning design (IMM-PBL) should be aneffective tool for increasing self-efficacy. Principles for the design of IMM-PBL were derivedfrom the relevant literature.An IMM-PBL package was designed and developed for delivery in a web browser formatusing content relevant to the integration of ICTs into teaching. Interviews with and sampleresponses prepared by computer-using teachers provided the basis for ensuring therelevance of content.The completed materials were evaluated in use with a group of 24 final year pre-serviceteachers in a Queensland university. Participants in the trials reported that the materialswere engaging and assisted their learning about integrating computers in their teaching. Astatistically significant increase in self-efficacy for teaching with computers was found forusers who had initially low self-efficacy for teaching with computers.The principles proposed for IMM-PBL design were found to offer a practical basis for thedevelopment of effective learning materials. With further development, IMM-PBL promisesto be a powerful and flexible approach to supporting learning for teachers and otherprofessionals.

The thesis presents a model, suitable for the design of any type of qualification in integrative eGovernment education. The integrative approach combines education of adult learners and students and promotes international cooperation.

This thesis concerns the changing predispositions and conditions for contemporary meaning-making in school education. From a socio-cultural perspective, multimodal theory formation is used to find suitable tools and concepts for developing teaching and learning. The overall aims are to investigate and conceptualize meaning-making in school in the frame ofmultimodal theory. Firstly, the research questions are concerned with how teachers work with written; paper-based, expository texts, and secondly, with students' meaning-making, working with meaning-offerings from different modes and media. This is followed by questions surrounding the predispositions for a multimodal view in the Swedish curriculum outline. Finally, the consequences for the role of fiction in education, using multimodal theory formation as a framework are addressed. The thesis presents two empirical studies which investigate meaning-making in upper secondary education, followed by critical discussions of the cmTiculum outline and the role of fiction. The empirical data was collected using methods inspired by ethnography in classes taking social sciences and media courses. The analyses were inspired by multimodal research, and the main analytical tools consist of a discourse framework and model inspired by Roz IvaniC, the Leaming Design Sequence developed by Staffon Selander, the wheel of multimodality and the pedagogy of multiliteracies, both developed by the New London Group and Bill Cope and :Mary Kalantzis. The first study focuses on the teachers' perspective in trying to develop students' meaning-making through written, paper-based expository texts. Analyses within the discourse framework and design layer model are used to describe the teachers' practical theory. The wheel ofmultimodality is used to differentiate the meaning-offerings used in class, and the pedagogy of multiliteracies is used to describe and analyze the discussions in groups and with the teacher. Results highlight three major possibilities for working with written, paper-based expository texts: a vvider view on meaning-making, meaning-offerings encompassing several modes and media, and the teacher's modeling ofthe reading through discussion. The second study describes and analyzes meaning-making and design in learning \vith meaning-offerings from different modes and media from the students' perspective. The analytical tools are the wheel of multimodality, the Learning Design Sequence and the further-developed pedagogy of multiliteracies. Results show a similarity in meaning-making regardless of mode and media, staiiing with the visual mode and with the students focusing their efforts on comprehending the meaning-offering. This can be explained by lack of clarity and lack of guidance which are seen as obstacles for learning. The discussions surrounding the curriculum outline and the role of fiction show that, in using a multimodal theory formation frame, the curriculum does not explicitly support a multimodal view on meaning-making and that fiction can not be seen as unique due to neither mode nor media. The results suggest that multimodal theory formation gives access to tools that are useful in developing students' meaning-making according to the predispositions and conditions oftoday, in which reading development is viewed as part of developing meaning-making as a who lei and that meaning-making in school should be based on a non-hierarchical and inclusive view on modes and media to create a readiness and a flexibility to meet demands of a rapidly-changing society. As a consequence, the curriculum outline needs to be reworded and the role of fiction in education needs to be problematized.

This study investigates the terms, dialogue, structure and active learning in Open Distance Learning texts in aa constructivist and problem-solving approach. In-text activities are imbedded in selfstudy texts. Their construct is validated against appropriate instructional design and learning theories for active learning. A course team developed, implemented and piloted a Unisa course for mathematics teachers in South Africa. The results describe their professional development through action learning. The Tornado-approach affects teams in organisations. A team used participatory action research and action learning with rich descriptions to document the longitudinal case study of four action research cycles over eight years. Qualitative research involved a literature survey, notes, portfolios, mindmaps, letters, interviews, document analyses, and assessment results. Our action research demonstrates ways to implement a post-modern learning design in an industrialized institution. The researcher applies action learning in professional development, in team management, and suggests models for successful/unsuccessful teams (tornado-effect). / Educational Studies / M. Ed. (Didactics)

The thesis presents a model, suitable for the design of any type of qualification in integrative eGovernment education. The integrative approach combines education of adult learners and students and promotes international cooperation.

The goal of learning analytics is to optimize learning and the environments in which it occurs. Since 2011, when learning analytics was defined as a separate and distinct area of academic inquiry, the literature has identified a need for research that presents evidence of effective learning analytics, as well as, learning analytics research that is conducted in conjunction with learning theory. This study uses Efklides' metacognitive and affective model of self-regulated learning (MASRL) to define cognitive, metacognitive, and affective variables that can explain students' learning outcomes in hybrid/online sections of Calculus I in the 2020-21 academic year. Cognitive variables were measured according to the cognitive operational framework for analytics (COPA). Metacognitive variables were defined according to the ways in which students interacted with the course content in the learning management system (LMS) and supplemental instruction, and affective variables were measured by ways students gave evidence of their affective states, such as in discussion board posts. All variables were compared across the course learning design, activities, and outcomes. Binary logistic regression revealed five significant variables: two cognitive, one metacognitive, and two affective. Thus, this study provided a learning analytics, evidence-based link between self-regulated learning theory and learning design, activities, and outcomes. In addition, implications for students, instructors, and learning theory were explored, as well as, the qualifications of this study as evidence of effective learning analytics.

Im folgenden Artikel werden die Anfänge des Reallabors Kiel als hochschuldidaktisches Lehrformat an der Christian-Albrechts-Universität zu Kiel dargestellt. Im Rahmen des Projekts erfolgreiches Lehren und Lernen (PerLe) an der Universität Kiel wird seit dem Sommersemester 2017 jedes Semester ein Lehrformat im Reallabor Kiel angeboten. Die Formate sollen transdisziplinäres Lernen im Sinne einer Bildung für nachhaltige Entwicklung fördern. Der Beitrag beschreibt die Grundzüge der Lehrformate, ordnet zentrale Aspekte in aktuelle hochschuldidaktische Diskussionen ein und formuliert Fragen, die für erfolgversprechende Umsetzungen von Lehrformaten mit Reallabor-Bezügen im bisherigen Prozess deutlich wurden.