Search

I MÅNGA LÄNDER världen runt har digitalisering och programmering införts i läroplaner för förskola och grundskola. Under senare tid har flera studier om undervisning i programmering i olika sammanhang genomförts. Dessa studier har oftast haft fokus på elever från 7 år och äldre och det finns få studier som rapporterar om införandet av programmering för barn i förskola och yngre elever i förskoleklass. Denna klyfta hanterar vi i denna studie som fokuserar på programmering och utveckling av elevers datalogiska tänkande i förskoleklass. Hösten 2017 startade Ifous (Innovation, forskning och utveckling i skola och förskola) ett forsknings- och utvecklingsprogram med fokus på utveckling av didaktiska modeller för programmering i utbildningen från förskola till och med grundskolan. Programmet avslutades våren 2020. I detta program deltog skolor från fem skolhuvudmän, Tyresö kommun, Åstorps kommun, Simrishamns kommun, Stockholms stad och Freinetskolan Hugin i Norrtälje. Programmet som sådant innebar viss kompetensutveckling och främjade forskning om programmeringsundervisning från förskoleklass till och med årskurs 9 i grundskolan. I denna artikel presenterar vi hur en lärare arbetade med programmering för att främja datalogiskt tänkande hos elever i förskoleklass under treårsperioden och den erfarenhet som hon förvärvat. I artikeln drar vi slutsatsen att förskoleklasselever med systematisk och tankeväckande didaktisk modellering fullt ut kan utveckla ett antal grundläggande datalogiska färdigheter., Programmering i ämnesundervisningen

Educational concepts are the core of teaching and learning. From the perspective of educational technology, concepts are essential meta-data, represen- tative terms that can connect different learning materials, and are the foundation for many downstream tasks. Some studies on automatic concept extraction have been conducted, but there are no studies looking at the K-12 level and focused on the Swedish language. In this paper, we use a state-of-the-art Swedish BERT model to build an automatic concept extractor for the Biology subject using fine- annotated digital textbook data that cover all content for K-12. The model gives a recall measure of 72% and has the potential to be used in real-world settings for use cases that require high recall. Meanwhile, we investigate how input data fea- tures influence model performance and provide guidance on how to effectively use text data to achieve the optimal results when building a named entity recognition (NER) model.

Automated Essay Scoring (AES) automatically assigns scores to essays at scale and may help to support teachers' grading activities. Recently, AES methods based on deep neural networks (DNN) have significantly improved upon the state-of-the-art performance by learning relations between holistic essay scores and student essays. However, DNN-based AES methods function like black-box, negatively affecting the ability to provide automated writing evaluation (AWE). In this work, we proposed a new method, topic-aware BERT, based on fine-tuning the pre-trained language model to learn relations between essay scores and text representations of student essays as well as topical information in essay writing instructions. Moreover, we propose an approach to automatically retrieve key topical sentences in student essays by probing self-attention maps in intermediate layers of topic-aware BERT. We evaluate the performance of topic-aware BERT to (i) perform AES and (ii) retrieve key topical sentences using the open dataset Automated Student Assessment Prize and a manually annotated dataset, respectively. Our model achieves a strong AES performance compared with previous state-of-the-art DNN-based methods and shows effectiveness in identifying key topical sentences in argumentative essays.

Many national curricula have incorporated computational thinking (CT) into compulsory education. Teacher ability to deliver the revised curriculum determines whether these new skills can be successfully integrated into teaching. Therefore, it is crucial to examine teacher readiness. This study measured Swedish K-9 teacher CT skills through a CT test validated by an expert review panel and a principal component analysis. Additionally, we engaged statistical analyses to examine the relationship between the teachers' background and their CT test scores, as well as their self-reported ability to teach CT. The result demonstrated the teachers' proficiency in different types of CT skills. Another finding revealed that the type of programming language mastered by teachers was associated with both their CT test score and self-reported ability to teach CT. This CT test can support teachers to identify specific areas for professional development and may facilitate the school management to plan teachers' competence training strategically.

Formative feedback has long been recognised as an effective tool for student learning, and researchers have investigated the subject for decades. However, the actual implementation of formative feedback practices is associated with significant challenges because it is highly time-consuming for teachers to analyse students’ behaviours and to formulate and deliver effective feedback and action recommendations to support students’ regulation of learning. This paper proposes a novel approach that employs learning analytics techniques combined with explainable machine learning to provide automatic and intelligent feedback and action recommendations that support student’s self-regulation in a data-driven manner, aiming to improve their performance in courses. Prior studies within the field of learning analytics have predicted students’ performance and have used the prediction status as feedback without explaining the reasons behind the prediction. Our proposed method, which has been developed based on LMS data from a university course, extends this approach by explaining the root causes of the predictions and by automatically providing data-driven intelligent recommendations for action. Based on the proposed explainable machine learning-based approach, a dashboard that provides data-driven feedback and intelligent course action recommendations to students is developed, tested and evaluated. Based on such an evaluation, we identify and discuss the utility and limitations of the developed dashboard. According to the findings of the conducted evaluation, the dashboard improved students’ learning outcomes, assisted them in self-regulation and had a positive effect on their motivation.

It has been argued that participation as involvement as well as taking part in decision making is a fundamental part of learning. The first large arena for experiencing this for many young people is the school. This paper investigates how 16 pupils at grade eight view their own and their teacher’s role when it comes to facilitating participation. The pupils are interviewed the semester after a long researcher led participation project and the interviews are analyzed using thematic analysis. The analysis identifies four different types of participation from the pupils’ responses. These four types can be connected to passive and active participation as well as participation as involvement and participation as influence

Knowledge tracing—where a machine models the students’ knowledge as they interact with coursework—is a well-established area in the field of Artificial Intelligence in Education. In this paper, an ensemble approach is proposed that addresses existing limitations in question-centric knowledge tracing and achieves the goal of predicting future question correctness. The proposed approach consists of two models; one is Light Gradient Boosting Machine (LightGBM) built by incorporating all relevant key features engineered from the data. The second model is a Multiheaded-Self-Attention Knowledge Tracing model (MSAKT) that extracts historical student knowledge of future question by calculating their contextual similarity with previously attempted questions. The proposed model’s effectiveness is evaluated by conducting experiments on a big Kaggle dataset achieving an Area Under ROC Curve (AUC) score of 0.84 with 84% accuracy using 10fold cross-validation.

In this paper, we propose a counterfactual explanations-based approach to provide an automatic and intelligent recommendation that supports student's self-regulation of learning in a data-driven manner, aiming to improve their performance in courses. Existing work under the fields of learning analytics and AI in education predict students' performance and use the prediction outcome as feedback without explaining the reasons behind the prediction. Our proposed approach developed an algorithm that explains the root causes behind student's performance decline and generates data-driven recommendations for action. The effectiveness of the proposed predictive model that constitutes the intelligent recommendations is evaluated, with results demonstrating high accuracy.

QC 20210526

Ingress: 23 forskare inom it- och utbildningsområdet: Regeringens utredare borde inte lyfta fram läsning på skärm som något negativt. Forskning ­visar att både tryckta och digitala läromedel behövs i skolan., DN Debatt.

QC 20201208

The bachelor thesis is commonly a necessary last step towards the first graduation in higher education and constitutes a central key to both further studies in higher education and employment that requires higher education degrees. Thus, completion of the thesis is a desirable outcome for individual students, academic institutions and society, and non-completion is a significant cost. Unfortunately, many academic institutions around the world experience that many thesis projects are not completed and that students struggle with the thesis process. This paper addresses this issue with the aim to, on the one hand, identify and explain why thesis projects are completed or not, and on the other hand, to predict non-completion and completion of thesis projects using machine learning algorithms. The sample for this study consisted of bachelor students’ thesis projects (n=2436) that have been started between 2010 and 2017. Data were extracted from two different data systems used to record data about thesis projects. From these systems, thesis project data were collected including variables related to both students and supervisors. Traditional statistical analysis (correlation tests, t-tests and factor analysis) was conducted in order to identify factors that influence non-completion and completion of thesis projects and several machine learning algorithms were applied in order to create a model that predicts completion and non-completion. When taking all the analysis mentioned above into account, it can be concluded with confidence that supervisors’ ability and experience play a significant role in determining the success of thesis projects, which, on the one hand, corroborates previous research. On the other hand, this study extends previous research by pointing out additional specific factors, such as the time supervisors take to complete thesis projects and the ratio of previously unfinished thesis projects. It can also be concluded that the academic title of the superv, QCR 20210510

There has been significant efforts in studying collaborative and social learning using aggregate networks. Such efforts have demonstrated the worth of the approach by providing insights about the interactions, student and teacher roles, and predictability of performance. However, using an aggregated network discounts the fine resolution of temporal interactions. By doing so, we might overlook the regularities/irregularities of students' interactions, the process of learning regulation, and how and when different actors influence each other. Thus, compressing a complex temporal process such as learning may be oversimplifying and reductionist. Through a temporal network analysis of 54 students interactions (in total 3134 interactions) in an online medical education course, this study contributes with a methodological approach to building, visualizing and quantitatively analyzing temporal networks, that could help educational practitioners understand important temporal aspects of collaborative learning that might need attention and action. Furthermore, the analysis conducted emphasize the importance of considering the time characteristics of the data that should be used when attempting to, for instance, implement early predictions of performance and early detection of students and groups that need support and attention., QC 20210203

Problem and goal. Computational thinking has been introduced in many countries around the world and teachers are working intensely to incorporate programming activities in the classroom. However, teachers are faced with several challenges due to the fact that there is still little research conducted focusing on programming education for younger children, that programming didactics is a rather new phenomenon for the K-9 educational system, and that K-9 teachers have little training with regards to programming. In Sweden for instance, programming has been introduced in several subjects and not as a subject in its own, which create a pressure on teachers to utilize programming as an instrument to teach and enhance learning of different subjects such as mathematics. Methodology. In this paper, we report on a larger lesson study conducted in a primary school (sixth grade) in Sweden with a total of 155 participating pupils. The aim of the lesson was to study whether the visual programming languages, Scratch, in particular, can be used to teach computational thinking, mathematics and social science in an interdisciplinary way. Results. Thus, the paper more specifically presents findings related to: 1) reflections of the use of lesson study methodology to develop programming education; 2) how programming can be utilized as an instrument to teach mathematics as well as social sciences in an interdisciplinary way; and 3) the didactical strategies employed by the teachers. Conclusion. The evidence from this study suggests that the interdisciplinary character of the lesson which incorporates learning goals of mathematics, social science and programmingwas highly beneficial. The pupils gained a better understanding of learning material by drawing, digitalizing and animating their ideas in Scratch.

Student engagement is critical for learning. However, little is known about engagement and disengagement and particular social groups. Recent research has alerted that engagement in technology-enhanced learning (TEL) settings may manifest differently than engagement in analogue learning settings. This study explores how different social groups of upper secondary school students (n= 410) engage and disengage when learning with digital technologies. We used an instrument to approach dimensions of engagement and disengagement in TEL. Using thematic analysis, we identified cognitive, emotional, behavioural and social aspects of engagement and disengagement in eight-student interviews which together with theory, informed a questionnaire. Using statistical methods, we explored the relationship between engagement, disengagement and the social categories: gamers, gender and non-native speakers. We found significant differences between the groups. For example: that high-frequency gaming students were not as easily distracted as students reporting low-frequency gaming, that female students engaged in TEL in different ways than male students, and that non-native speakers displayed significantly fewer tendencies to engage in unauthorised uses of digital technologies than native speakers. Identifying indicators reflecting engagement and disengagement in TEL in social groups can inform successful practices that stimulate student engagement and can be used to avoid, or redeem, group-specific challenges that trigger disengagement.

Although Computational Thinking and Programming have become obligatory in many national curricula, the majority of teachers in practice are currently in dire need of support from both the research and teaching community. A national research and teacher development project was initiated in Sweden to address this issue. To develop their pedagogical competence in these new subjects, 31 in-service teachers from the project performed lesson studies at their schools. This study collected and analyzed the documentation of 12 lesson studies regarding how computational thinking skills are taught and assessed in K--9, with block-based programming. The study applies a validated CT framework to identify the involved computational thinking skills. The result presents a progression stage scheme for these skills in compulsory schools. The paper also discusses the problems and challenges in the assessment of computational thinking skills.

The association between mathematics and programming in an educational context is not new. Today, programming has been introduced into curricula worldwide for younger children. In the Swedish case, primary school teachers are expected to integrate programming in mathematics education from autumn 2018. However, Swedish teachers' knowledge of programming and programming didactics is limited. Meanwhile, there is little research on K-9 programming education. This has led to the dilemma that the mathematics teachers have limited support in didactic knowledge and good examples. This study reports on a teacher professional development project in programming. More specifically, teachers used Lesson Study to plan, execute, and evaluate lessons that integrated programming into various school subjects in elementary school. This study analyzed the didactic strategies developed in 10 lesson studies, as well as mapped the opportunities and challenges of pupils' learning in the mathematics subject. The result was the identification of three didactic strategies, which were analog programming, robot programming and block programming, as well as 11 didactic methods applied within these strategies. The paper contributes with examples of the didactic methods that teachers have developed and evaluated using lesson study. The paper further provides insights on how teachers can take progression into account by applying the three didactic strategies. At last but not least, the study shows a great need for teachers to develop computational thinking abilities.

Teachers around the world have started teaching programming at the K-9 level, some due to the formal introduction of programming in the national curriculum, others without such pressure and on their own initiative. In this study, we attempted to understand which skills – both CT-related and general – are developed among pupils in the process of working with programming in schools. To do so, we interviewed 19 Swedish teachers who had been teaching programming for a couple of years on their own initiative. The teachers were selected based on their experience in teaching programming. Our thematic analysis of these interviews shed light on what skills teachers perceive pupils develop when programming. This led us to identify three themes related to CT skills and five themes related to general skills. The CT skills identified corresponded well with and were thus thematically structured according to the dimensions of CT proposed in the framework of Brennan and Resnick, namely computational concepts, computational practices and computational perspectives. In addition to the CT skills, our thematic analysis also resulted in the identification of general skills related to digital competency and 21st century skills, namely cognitive skills and attitudes, language skills, collaborative skills and attitudes and creative problem-solving skills and attitudes.

QC 20190903

We know that employing collaborative learning strategies does not lead to productive collaborative learning per se. In fact, some groups are dysfunctional and might have a detrimental influence on group members. This issue has methodologically been studied through self-reported surveys, transcripts coding, and observational methods. Although these methods are informative, they are also time intensive, exhausting and not practical to be applied in real practice. Social network analysis (SNA) and learning analytics, on the other hand, open the door for using automatic and non-intrusive methods that can help us monitor group interactions. Here we study if SNA combined with machine learning techniques can be employed in order to better understand and predict how different type and quantities of collaborative interactions in online environments affect individual and group performance. More specifically, we study the correlation between group interaction parameters as measured by SNA and groups and individual’s performance. Using interaction parameters and machine learning methods, we identify the indicators that best predict groups that will perform and gain and groups that will not, as well as individuals who gain in performance and those who do not. The article provides support for the idea that learning analytics can help automatically monitor group performance and offer an opportunity for educators and learners to support productive collaborative learning., QC 20201208

In this learning analytics study, we attempt to understand the role of temporality measures for the prediction of academic performance. The study included four online courses over a full-year duration. Temporality was studied on daily, weekly, course-wise and year-wise. Visualising the activities has highlighted certain patterns. On the week level, early participation was a consistent predictor of high achievement. This finding was consistent from course to course and during most periods of the year. On course level, high achievers were also likely to participate early and consistently. With a focus on temporal measures, we were able to predict high achievers with reasonable accuracy in each course. These findings highlight the idea that temporality dimension is a significant source of information about learning patterns and has the potential to inform educators about students’ activities and to improve the accuracy and reproducibility of predicting students’ performance., QC 20201207

Learning analytics (LA) is a relatively new research discipline that uses data to try to improve learning, optimizing the learning process and develop the environment in which learning occurs. One of the objectives of LA is to monitor students activities and early predict performance to improve retention, offer personalized feedback and facilitate the provision of support to the students. Flipped classroom is one of the pedagogical methods that find strength in the combination of physical and digital environments i.e. blended learning environments. Flipped classroom often make use of learning management systems in which video-recorded lectures and digital material is made available, which thus generates data about students interactions with these materials. In this paper, we report on a study conducted with focus on a flipped learning course in research methodology. Based on data regarding how students interact with course material (video recorded lectures and reading material), how they interact with teachers and other peers in discussion forums, and how they perform on a digital assessment (digital quiz), we apply machine learning methods (i.e. Neural Networks, Nave Bayes, Random Forest, kNN, and Logistic regression) in order to predict students overall performance on the course. The final predictive model that we present in this paper could with fairly high accuracy predict low- and high achievers in the course based on activity and early assessment data. Using this approach, we are given opportunities to develop learning management systems that provide automatic datadriven formative feedback that can help students to selfregulate as well as inform teachers where and how to intervene and scaffold students., QC 20201207

The bachelor thesis is commonly a necessary last step towards the first graduation in higher education and constitutes a central key to both further studies in higher education and employment that requires higher education degrees. Thus, completion of the thesis is a desirable outcome for individual students, academic institutions and society, and non-completion is a significant cost. Unfortunately, many academic institutions around the world experience that many thesis projects are not completed and that students struggle with the thesis process. This paper addresses this issue with the aim to, on the one hand, identify and explain why thesis projects are completed or not, and on the other hand, to predict non-completion and completion of thesis projects using machine learning algorithms. The sample for this study consisted of bachelor students’ thesis projects (n=2436) that have been started between 2010 and 2017. Data were extracted from two different data systems used to record data about thesis projects. From these systems, thesis project data were collected including variables related to both students and supervisors. Traditional statistical analysis (correlation tests, t-tests and factor analysis) was conducted in order to identify factors that influence non-completion and completion of thesis projects and several machine learning algorithms were applied in order to create a model that predicts completion and non-completion. When taking all the analysis mentioned above into account, it can be concluded with confidence that supervisors’ ability and experience play a significant role in determining the success of thesis projects, which, on the one hand, corroborates previous research. On the other hand, this study extends previous research by pointing out additional specific factors, such as the time supervisors take to complete thesis projects and the ratio of previously unfinished thesis projects. It can also be concluded that the academic title of the superv, QC 20201207

In 2019, the mobile learning body of knowledge is extensive and much is known about the technology impacts and affordances of mobile devices in educational settings. A particular focus has now shifted toward specific tech- nologies in specific subjects. Mathematics is one such subject and tablets are one such technology that is gaining attention. This systematic review represent- ing the latest generation of tablet technology within the tablet-mediated learning in mathematics body of knowledge sought to derive evidence that supported questions into (a) what math sub-disciplines were covered, (b) what technology (application/hardware) was utilized, and (c) what pedagogical approaches were deployed in educational settings. This included analysis of the (d) advantages and (e) disadvantages present in those elements. Thirty-nine relevant articles were collected from various academic technology and educational databases. The results demonstrate that tablets are being predominantly deployed in vari- ous sub-disciplines such as Arithmetic, Computation, and Geometry with the iPad as the dominant choice for tablet hardware/applications. Pedagogical ap- proaches lean heavily on game-based learning, environment interaction, and special needs support. Technological advantages include increased collabora- tion and mathematics engagement enabled by tablet mobility and a high poten- tial for customization of solutions. Developers, teachers, and researchers need to be informed of potential challenges in designing content for tablet technology deployments in mathematics., QC 20190819

Many national curricula have incorporated Computational Thinking (CT) in compulsory education. Teachers' ability to deliver the revised curriculum determines whether these new skills can be successfully integrated into teaching. Therefore, it is crucial to examine teachers' readiness. This study measured Swedish K-9 teachers' CT skills through a CT test validated by an expert review panel and a principal component analysis. Additionally, we engaged statistical analyses to examine the relationship between the teachers' background and their CT test scores, as well as their self-reported ability to teach CT. The result demonstrated the teachers’ proficiency in different types of CT skills. Another finding revealed that the type of programming language mastered by teachers was associated with both their CT test score and self-reported ability to teach CT. This CT test can support teachers to identify specific areas for professional development and may facilitate the school management to plan teachers’ competence training strategically.

Learning outdoors with mobile devices is associated with distinct challenges and constraints that needs to be taken into account when orchestrating formal mobile learning activities. In order to design pedagogically meaningful activities, we need to consider students scaffolding needs and have an understanding of the aspects that should be orchestrated for meeting those needs. This paper proposes an orchestration model for formal mobile learning activities across contexts that take such scaffolding needs into account. The model has been interactively developed based on empirical research conducted in three case-studies and have theoretical basis in sociocultural perspectives on learning, particularly resting on the concept of scaffolding and on the learning design sequence model of Selander. The model takes the orchestration of six scaffolding aspects into account, namely: the social (collaborative) aspects, the teachers, the technology, the physical context, the learning processes and tasks, and the modes and representations.

The student experience with different aspects of online instructional settings has been the focus of educational practitioners and researchers in many studies. However, concerning technology-enabled formative assessment, little is known about student satisfaction regarding different possible formative e-assessment strategies the students are involved in. Using a 5-point Likert scale questionnaire, a web-based survey was developed to examine students’ satisfaction with the formative e-assessment strategies within an enriched virtual blended course. The results show that, in general, the students were satisfied with the quality of their engagement and the quality of feedback across all the formative e-assessment activities offered. The results also show that the student satisfaction varied between and within the formative e-assessment strategies. However, the gap between the student satisfaction mean ratings across all formative e-assessment strategies was marginal and could not help researchers decide upon which formative e-assessment strategy that stood out as the most preferred one. Learner satisfaction with different formative e-assessment strategies was positively correlated to each other at various levels but no relationship was found between students’ scores on the final course exam and learner satisfaction with formative e-assessment strategies. In the end, the study recommends a sustained and integrated use of the all three formative e-assessment strategies (online knowledge survey, online student-generated questions and peer-responses, and electronic reflective journals) in the context of hybrid courses. Further studies that would widen, diversify both the scope and the research instruments to investigate learner satisfaction with formative e-assessment strategies were also suggested.