Mid Sweden University
Not a member yet
15329 research outputs found
Sort by
AI and the Evolving Role of the Software Engineer: Evidencefrom Swedish Job Advertisements
No full textThe integration of artificial intelligence (AI), particularlygenerative tools such as ChatGPT and GitHub Copilot, isaltering software engineering (SE) practices by automatingdevelopment tasks. This study investigates how AIintegration influences skill requirements in the Swedish SEjob market. Using empirical methods to assess whether AIrelated skills, such as machine learning (ML) and naturallanguage processing (NLP), are becoming more substantialin SE job advertisements. A long-term analysis wasconducted on a dataset of Swedish job ads from the pastdecade. The methodology consisted of collecting archivedjob ads and identifying key terms to extract AI-related skillsand terminology to compare them over time.Results show a clear increase in both the frequency andvariety of AI-related concepts in recent years. This suggeststhat the competence of AI is becoming an integral part of themodern software engineer's expertise. Our findings provideempirical insight into the shifting expectations of SEspecialists, while also offering practical implications foreducational institutes and employers aiming to narrow thegap between academic curricula and employment
Mellan utbildningsideal och styrdokument : En multipel fallstudie om lärares agens i bedömningspraktiker
No full textThis dissertation presents a study of teachers' agency in assessment practices. In recent decades, assessment has become an increasingly prominent aspect of teachers' work. Governance ambitions through more explicit regulation and pedagogical trends informed by research on formative assessment, have aimed to enhance the quality of learning while promoting greater equity in teachers’ assessment practices. However, research shows that teachers differ in both how assessments are conducted and in what is being assessed. The research problem in this dissertation is therefore grounded in the question of how and why teachers’ assessment practices vary. The aim of the study is to describe and analyze teachers’ agency in assessment practices in grade six, which is accomplished through four case studies where the empirical collection consisted of observations and interviews. Teachers' experiences, attitudes, and approaches to the cross-pressure of expectations and conditions surrounding their pedagogical work are understood theoretically through an ecological perspective on agency. The findings demonstrate that teachers' assessment practices differ in terms of the dimensions of assessments, namely, the purposes of assessment, its use, and its relationship to teaching. The dissertation further shows that teachers’ approaches to the interaction between individual characteristics and structural and classroom conditions, analyzed as rationales for assessment, can help explain why and how assessment practices vary, as different rationales highlight different dimensions of assessment in varying ways. The dissertation argues that agency in assessment practices can be understood within the tension between teachers’ educational ideals and the regulations outlined in governing documents. In the tension between educational ideals and regulations, aspects related to the classroom context may become significant. These classroom-contextual aspects also differ in nature from more structural factors, in that teachers can exert influence over them. This tension is primarily expressed across four domains: learning objectives, student participation, fairness and reliability, and consequences of assessment. The variation in assessment practice, for instance, in how research findings on formative assessment are implemented, or how efforts to standardize teachers’ assessment work through policies and reform packages are realized in practice, is therefore proposed to be understood through the lens of teachers’ agency within the tension between educational ideals and policy directives
Arbete: Plikt, rättighet eller tvång? : En diskursanalys av konstruktionen av arbete i svensk politik, media och civilsamhälle
No full text2025-06-04</p
Suicidpreventivt arbete i glesbygd : En kvalitativ studie om suicidpreventivt arbete i glesbygd med fokus på utmaningar, män och hegemonisk maskulinitet
No full text2025-08-26</p
Långtidsarbetslöshet : En kvalitativ studie om upplevelser och erfarenheter av att vara långtidsarbetslös
No full text2025-08-26</p
Lärande genom konflikthantering- pedagogers förhållningssätt och strategier : En studie om fyra förskollärares förhållningssättoch agerande i konflikter
No full textDenna studie syftar till att analysera hur pedagogers förhållningssätt och agerande är en viktig faktor när det kommer till hanterandet av barns konflikthanteringar i den pedagogiska verksamheten, detta för att ta reda på vilka strategier och metoder som används i deras hanterande. I genomförandet av studien bygger vår empiri på semistrukturerade intervjuer som har ljudinspelats och senare transkriberas, där vi även har använt oss av observationer. Vi har under fyra olika arbetsdagar fått följa fyra förskollärare. Därefter har den insamlade datan analyserats och bearbetats utifrån den tematiska analysen. Studiens teoretiska utgångspunkt utgår utifrån det relationella perspektivet som Aspelin tolkar utifrån Thomas, J. Scheff relationsteori som hävdar att pedagogers relationella kompetens är en väsentlig del gällande kommunikation mellan barn och pedagog, vilket är en viktig aspekt när det kommer till pedagogernas förhållningssätt i att hantera konflikter. I resultaten får vi ta del av att samtliga förskollärare har liknande synsätt när det kommer till konflikter och hanteringar av dessa, men där hanteringar och synsätt i vissa delar kan variera, dels utifrån ålder men även utifrån deras olika syn och erfarenheter med konflikter. Där vi slutligen i vår slutsats framför pedagogernas betydande roll för att hantera alla slags konflikter som uppstår på ett sätt som gör det lärande för barnen, med utgångspunkt utifrån det relationella perspektivet för att även förstå vikten av relationer mellan barn och pedagog när en hantering av en konflikt sker.Betyg i Ladok 250825.</p
Integrating Static Analysis and LLMs to Refactor ComplexMethods
No full textCode complexity significantly affects the readability andmaintainability of large-scale software systems. Refactoringis a common practice to address this, and recentadvancements in Large Language Models (LLMs) showpromise for automating the process. However, purely LLMbased refactoring often introduces errors or fails to addressspecific complexity issues. To mitigate such problems,hybrid approaches that combine static analysis with LLMshave been proposed, yet their effectiveness for complexityfocused refactoring remains underexplored. This studyevaluates PMD-LLM-Refactor, a hybrid approach integratingstatic complexity analysis with LLM-driven refactoring,specifically targeting method-level code complexity. Weempirically compared it against an LLM-only baseline usingthree models (DeepSeek-R1 Distill Qwen 32B, Gemini 2.0Flash, Llama 4 Maverick) across 45 methods from five opensource Java projects. The comparison evaluated fourstructural complexity metrics (Cyclomatic, Cognitive, NPath,and Method Length) and verified functional correctnessthrough existing unit tests. The hybrid approachoutperformed the LLM-only baseline, yielding more effectiverefactorings, a higher average complexity reduction acrossall metrics (58.8% vs. 51.7%), and lower variability. However,no statistically significant performance difference wasobserved among the three LLMs within the hybridconfiguration. The findings suggest that combining staticanalysis with LLMs improves the precision, consistency, andoverall effectiveness of automated method-level refactoringto reduce code complexity. The structured guidance enablesLLMs to perform more targeted transformations while alsomitigating common failure modes such as syntax errors andhallucinated code
Childhood self-regulation and life outcomes in a 50-year follow-up study
No full textSelf-regulation is integral to human well-being. Previous research established associations between self-regulation and life outcomes, although mostly based on cross-sectional or longitudinal research with relatively short follow-up periods. We aimed to investigate associations between self-regulation during childhood and low education, psychiatric diagnosis, conviction for non-violent and violent crime, and premature mortality up to 50 years later. The sample consisted of boys born between 1943 and 1951 in Stockholm, Sweden, who were initially interviewed as part of the Clientele Study and followed up until they were 58-66 years of age as part of the Stockholm Life Course Project (N = 251). Our measure of self-regulation was indicated by five dimensions extracted from multiple sources and informants when study members were children. Information on life outcomes were extracted from Swedish national registry data available through 2009. Results from logistic regression and survival analyses showed that lower self-regulation during childhood was associated with a higher risk of conviction for non-violent and violent crime, receiving a psychiatric diagnosis and premature mortality up until late adulthood. Self-regulation was not associated with low education. Analyses were adjusted for intelligence and parental socioeconomic status. While based on a relatively small sample and measurement of self-regulation using data from the 1960s, our study supports the notion that self-regulation during childhood is central for well-being across the lifespan
Exploring the Usability of Retrieval Augmented Generation and Artificial Intelligence in Government : Toward High-Usability AI Systems in governance
No full textRecent advances of Artificial Intelligence (AI) and Retrieval-Augmented Generation (RAG) have facilitated interfacing with textual data in unprecedented ways. However, usability studies identifying factors for high-usability AI systems in governance and the employee perspective are lacking. The purpose of the study was to bridge the gap between artificial intelligence system capabilities and governance needs from the perspective of its end users, contributing to a foundation for designing generative AI and RAG systems with high usability in governance. The study had a two-pronged approach to data collection. A think aloud protocol and semi structured interview were used together. A RAG-model based on over 7000 documents from open government data was created for the study, served as the basis for data collection and represents a contribution in how RAG can be designed for the context of government and usability research methodology. Seven government employees from Region Skåne participated in the study: politicians, committee secretaries and a political secretary. Data was analyzed using thematic analysis. Results indicate several key factors for designing high usability AI systems in governance and exemplify many facets of usability that were valuable to the participants. Five overarching themes pertaining to the study purpose were revealed: areas of application, application handling, process of integration, quality and relevance and dangers of AI. The study provides guidance for developers and government stakeholders when designing AI and RAG systems and demonstrates that usability in the context of governance extends beyond interface design and requires considerations regarding safety and contextual fit.De senaste framstegen inom Artificiell Intelligens (AI) och Retrieval-Augmented Generation (RAG) har skapat stora möjligheter att söka och interagera med textdata. Det saknas dock studier som identifierar faktorer för att skapa hög användbarhet hos AI-system i offentlig förvaltning samt perspektivet från människor som arbetar där. Syftet med studien var att överbrygga gapet mellan AI-systems förmågor och de behov som slutanvändarna inom offentlig förvaltning har, samt att lägga en grund för att designa system med hög användbarhet i offentlig förvaltning. Studien hade en tvådelad metod för datainsamling med ett ”tänk-högt”-protokoll och en semistrukturerad intervju. En skräddarsydd RAG-modell baserad på över 7000 offentliga dokument skapades för studien, fungerade som grund för datainsamlingen och bidrar till kunskap om hur en RAG-modell kan designas inom offentlig förvaltning och metod inom användbarhetsforskning. Sju anställda i Region Skåne deltog i studien; politiker, utskottssekreterare och en politisk sekreterare. Data analyserades med en tematisk analys. Resultatet påvisar flera viktiga faktorer för att designa AI-system med hög användbarhet inom offentlig förvaltning. Studien belyste och exemplifierade också många aspekter av användbarhet som var värdefulla för deltagarna. Fem övergripande teman som hör samman med studiens syfte framkom: tillämpningsområden, applikationsstyrning, integrationsprocess, kvalitet och relevans samt farorna med AI. Studien bidrar med vägledning för utvecklare och intressenter inom offentlig förvaltning att designa AI och RAG system samt visar att användbarhet inom kontexten av offentlig förvaltning går bortom gränssnitt och kräver överväganden gällande säkerhet samt anpassning till kontext
