Search

The rapid development of Massive Online Open Courses (MOOCs) has attracted significant attention. Despite the publicity, one of the major problems is the high dropout of MOOC learners. Many MOOCs attract thousands of learners; thus, it is inevitable that they will have diverse motivations for studying. Therefore, it is important that MOOC instructors design courses that can address varieties of motivations to sustain online learning. The emergence, and widespread use of social media presents MOOC designers with opportunities to use social media to improve interactions. However, there are only few researches on how these engagements on social media affect learners' motivation and retention in MOOCs. This research aims to close these gaps and answer the question "What is the impact of social media engagement on learners' motivation in MOOCs?" Mixed methods approaches were used in the study, gathering data via the course platform, questionnaires, social media forum posts and a focus group study on the mobile messenger app, WhatsApp. The research used a MOOC entitled "Entrepreneurship and Innovation" that had been previously offered by the University of Greenwich, London. This was significantly redesigned and bore little relationship to the new version of the same name which ran for five weeks on the Canvas Network platform. The redesigned MOOC used three social media groups. These were advertised on the course platform and students who consented, registered on either Facebook, Google Hangout via webpage links. Those interested in WhatsApp were given a mobile number to join. At the end of the MOOC, the statistical analysis showed that of the 450 students who registered for the MOOC, 289 (64%) participants started the course, and 161 (35.8%) did not start the course; 94 (32.5%) participated in the social media platforms; 195 (67.5%) did not participate in social media and engaged solely via the Canvas platform. The results show that 92.5% of those who engaged in social media started the course while 70.7% who did not engage in social media, started. Completion rate was 24.5% in social media learners and 10.8% in non-social media learners. The qualitative data was subjected to thematic analyses. The qualitative results and the focus group study further revealed that engagement with social media motivated learners to start the course, network, share information, and obtain quick responses. However, others complained of distractions. Thus, this study provides statistical evidence, which shows that using social media to engage MOOC students before the course starts, could enhance course start-up and positively impact on course engagement, retention and completion.

The motivation for this study is the under-representation of females in computer science education and careers, which can be seen as both a waste of talent and the loss of educational and career opportunities for girls and women. This problem, commonly referred to as the "shrinking pipeline", acknowledges the progressive reduction in the number of females in computer science education and careers over a period of years. Strategies and initiatives implemented over the years to solve this problem include attempts to dispel computing career myths, promote the use of mentors and provide accurate information about computing to key influencers of girls. Technological initiatives have also explored the use of digital educational games to engage more girls with computer science. The limitation with these technological initiatives lie in their inability to engage as the many girls as boys, which is reflective of the situation that we see with digital entertainment games. Educational software designed for students and children has been described as exemplifying the same problems as experienced with games designed for boys, because the characteristics often promote negative gender stereotypic constructs about computer science. Considering the ubiquity of digital games, their impact is far reaching across all age groups, but it is particularly important for 11-14 year olds, as this is the age group where a gender divide over computer science appears in schools. To resolve this problem, gender neutral and gendered games were designed to make digital educational games more appealing to girls. The outcomes from these design techniques have resulted in suggestions to including game characteristics that appeal to girls in digital educational game design solutions. Consequently, the research questions addressed by this study are: - (1) "Can we understand what makes some digital entertainment games appealing to 11-14 year old girls?" - (2) "Can we use this knowledge to create computer science learning games that appeal to this audience?" The approach of the research was to conduct an exploratory study of digital entertainment games with a view to identifying the significant game characteristics that make them motivationally appealing to the 11-14 year old girls. The findings of the exploratory study were used to create two digital experimental educational games, one designed around the characteristics with the most positive appeal, and the other designed around the antithetical values of those game characteristics. The experimental games were designed using an adaptation of the Aleven et al. (2010) framework and based on the Google Blockly maze game which facilitated the combination of the learning objectives, game mechanics, dynamics & aesthetics (MDA) and instructional principles. The main study which followed thereafter involved 304 participants (girls=152 and boys=152), from five different locations in southeast England. The participants engaged with both games and online questionnaires were used to collect their views and opinion of the games. The boys of a similar age group were included in the study to provide comparative data for the study. The evidence from the main study identified the significant game characteristics which can make digital educational games appealing to the target group, namely 11-14 year old girls. The evidence indicated that the significant game characteristics are a key criterion in making the experimental games appealing to girls. This knowledge can be used to inform the design of gender-specific digital educational games and gender inclusive games. This can be achieved through the application of design frameworks, and the customisation of the significant game characteristics that appeal to this target group.

Owing to their easy engagement and motivational nature, games predominantly in young age groups, have been omnipresent in education since ancient times. More recently, computer video games have become widely used, particularly in secondary and tertiary education, as a method of enhancing the understanding of some subject areas (especially in English language education, geography, history and health) and also used as an aid to attracting and retaining students. Many academics have proposed a number of approaches using video game-based learning (GBL), to impart theoretical and applied knowledge, especially in the Computer Science discipline. Despite several years of considerable effort, the empirical evidence in the GBL literature is still missing, specifically that which identifies what students learn from a serious game regarding programming constructs, and whether or not they acquire additional skills after they have been introduced to a GBL approach. Much of the existing work in this area explores the motivational aspect of video games and does not necessarily focus on what people can learn or which cognitive skills they can acquire that would be beneficial to support their learning in introductory computer programming. Hence, this research is concerned with the design, and determining the educational effectiveness, of a game model focused on the development of computational thinking (CT) skills through the medium of learning introductory programming constructs. The research is aimed at designing, developing and evaluating a serious game through a series of empirical studies in order to identify whether or not this serious game can be an educationally effective tool for learning computer programming at the CT level. The game model and its implementation are created to achieve two main purposes. Firstly, to develop a model that would allow students to practise a series of cognitive abilities that characterise CT, regardless of their programming background. Secondly, to support the learning of applied knowledge in introductory programming by demonstrating how a limited number of key introductory computer programming constructs which introductory programming students often find challenging and/or difficult to understand. In order to measure the impact of the serious game and its underlying game model, a pilot-study and a series of rigorous empirical studies have been designed. The pilot study was conducted as a freeform evaluation to obtain initial feedback on the game’s usability. A group of students following Computer Science and related degree programmes with diverse backgrounds and experience participated in the pilot-study and confirmed that they found the game enjoyable. The feedback obtained also showed that the majority of students believed the game would be beneficial in helping introductory programming students learn computational thinking skills. Having incorporated the feedback into a revised version of the game, a further series of rigorous studies were conducted, analysed and evaluated. In order to accurately measure the effect of the game, the findings of the studies were statistically analysed using parametric or non-parametric measures depending on the distribution of data gathered. Moreover, the correlations between how well students did in the game, the knowledge gain students felt, and the skills they felt they acquired after their game-play are thoroughly investigated. It was found that intrinsic motivation, attitude towards learning through game-play, students’ perception of their programming knowledge, how well students visualise programming constructs and their problem solving abilities were significantly enhanced after playing the game. The correlations of the studies provided evidence that there is no strong and significant relationship between the progress of students in the game and the computational thinking skills they felt they gained from it. It was concluded that students developed their computational thinking skills regardless of whether or not they reached the higher levels in the game. In addition to this, it was found that there are no strong and significant correlations between the key computer programming constructs and the computational thinking skills, which provides strong evidence that learning how introductory computer programming constructs work and developing computational thinking skills, are not directly connected to each other in the game environment. It was also found that students felt that their conditional logic, algorithmic thinking and simulation abilities had significantly developed after playing the game. As a result, this research concludes that the designed serious game is an educationally effective tool for a) learning how key introductory computer programming constructs work and b) developing cognitive skills in computational thinking.

The work in this thesis presents an approach towards the effective monitoring of business processes using Case-Based Reasoning (CBR). The rationale behind this research was that business processes constitute a fundamental concept of the modern world and there is a constantly emerging need for their efficient control. They can be efficiently represented but not necessarily monitored and diagnosed effectively via an appropriate platform. Motivated by the above observation this research pursued to which extent there can be efficient monitoring, diagnosis and explanation of the workflows. Workflows and their effective representation in terms of CBR were investigated as well as how similarity measures among them could be established appropriately. The monitoring results and their following explanation to users were questioned as well as which should be an appropriate software architecture to allow monitoring of workflow executions. Throughout the progress of this research, several sets of experiments have been conducted using existing enterprise systems which are coordinated via a predefined workflow business process. Past data produced over several years have been used for the needs of the conducted experiments. Based on those the necessary knowledge repositories were built and used afterwards in order to evaluate the suggesting approach towards the effective monitoring and diagnosis of business processes. The produced results show to which extent a business process can be monitored and diagnosed effectively. The results also provide hints on possible changes that would maximize the accuracy of the actual monitoring, diagnosis and explanation. Moreover the presented approach can be generalised and expanded further to enterprise systems that have as common characteristics a possible workflow representation and the presence of uncertainty. Further work motivated by this thesis could investigate how the knowledge acquisition can be transferred over workflow systems and be of benefit to large-scale multidimensional enterprises. Additionally the temporal uncertainty could be investigated further, in an attempt to address it while reasoning. Finally the provenance of cases and their solutions could be explored further, identifying correlations with the process of reasoning.

This research project is interested in the area of personalised and adaptable learning and in particular within an e-learning context. Brusilovsky (1996) and Santally (2005) stress the importance of adaptive systems within e-learning. Karagiannikis and Sampson et al. (2004) argue that personalised learning systems can be defined by their capability to adapt automatically to the changing attitudes of the “learning experience” which can, in turn, be defined by the individual learner characteristics, for example the type of learning material. The project evolved to cover areas including personalised learning, e-learning environments, authoring tools, tagging, learning objects, learning theories and learning styles. The main focus at the start of the project was to provide a personalised and adaptable learning environment for students based on their learning style. During the research, this led to a specific interest about how an academic can create, tag and author learning objects to provide the capability of personalised adaptable e-learning for a learner. Research undertaken was designed to gain an understanding of personalised and adaptive learning techniques, e-learning tools and learning styles. Important findings of this research showed that e-learning platforms do not offer much in the way of a personalised learning experience for a learner. Additionally, the research showed that general adaptive systems and adaptive systems incorporating learning styles are not commonly used or available due to issues with flexibility, reuse and integration. The concept of tagging was investigated during the research and it was found that tagging is underused within e-learning, although the research shows that it could be a good ‘fit’ within e-learning. This therefore led to the decision to create a general purpose discriminatory tagging methodology to allow authors to tag learning objects for personalisation and reuse. The main focus for the evaluation of this tagging methodology was the authoring side of the tagging. It was found that other research projects have evaluated the personalisation of learning content based on a learner’s learning style (see Graf and Kinshuk (2007)). It was therefore felt that there was a sufficient body of existing evidence in this area whereas there was limited research available on the authoring side. The evaluation of the discriminatory tagging methodology demonstrated that the methodology could allow for any discrimination between learners to be used. The example demonstrated within this thesis includes discriminating according to a learner’s learning style and accessibility type. This type of platform independent flexible discriminatory methodology does not exist within current e-learning platforms or other e-learning systems. Therefore, the main contribution of this thesis is therefore a platform independent general-purpose discriminatory tagging methodology.

Recognising the powerful role that technology plays in the lives of people, researchers are increasingly focusing on the most effective uses of technology to support learning and teaching. Technology Enhanced Learning (TEL) has the potential to support and transform student learning and provides the flexibility of when, where and how to learn. At the same time, it promises to be an effective educational method (Wei and Yan 2009). One of the hottest topics in this field is adaptive learning (Mylonas, Tzouveli and Kollias 2004). Today, with the ability of advanced technologies to capture, store and use student data, it is possible to deliver adaptive learning based on student preferences. TEL can also put students at the centre of the learning process, which allows them to take more responsibility for their own learning. However, this requires students to be metacognitive so they can manage and monitor their learning progress. This thesis investigates the impact of student metacognitive skills on their learning outcomes in terms of recalling and retaining information within a formally designed and TEL environment. The learning outcomes of students who study a subject consistent with their learning styles and another group of students who study the same subject in contrast to their learning styles are then compared to determine which group performs better. Based on this approach, a TEL environment is designed for undergraduate students to use for the purpose of collecting the required experimental data. The results of this study suggest that effective use of metacognitive skills by students has a direct bearing on their learning performance and ability to recall information. The outcomes reveal that successful students use effective metacognitive skills to complete their studies and achieve their learning goals in a TEL environment. Therefore, it clear that metacognition can play a critical role in successful learning, and, furthermore, this approach can assist educationalists in understanding the importance of metacognition in learning and in considering how technology can be used to better to allow students to apply metacognitive skills. The designed TEL environment for this study can be utilised as a precursor to implement TEL environments that can be adapted to individual learning styles, and to support the development of metacognitive skills.

The purpose of this thesis is to investigate the impact of learners‟ measure of control over their learning, while working in different online learning environments, and how this, in combination with a structured learning material selection according to their learning preferences, can affect their learning performance. A qualitative study was carried out on the understanding of different learning philosophies, different learning environments and different learning preferences, in correlation with learners‟ measure of control over their learning environments, in terms of their influence on their learning performance. The research commenced with a survey of UK Higher Educational institutions, to determine the usage of adaptive e-learning systems in UK HE and the type and nature of the systems in use, which in combination with the literature review enabled the clarification of the research hypothesis and objectives. Since a measurement of learners‟ learning performance was needed, an adaptable personalised e-learning system (ALPELS) was developed to create an environment where a qualitative measurement could be done. Experimental data was then gathered from two cohorts of MSc students over two semesters, who used the newly designed and developed online learning environment. The successful implementation of the project has produced a large amount of data, which demonstrates a correlation between i) adaptable and personalised e-learning systems, and ii) learners‟ learning styles (which in itself supports the behaviouristic approach towards this type of online learning environment – ALPELS). The study indicates a dependency between an online controlled learning environment and learners‟ learning performances, showing that a personalised e-learning system (PELS) would be supportive of recall (R) and understanding (U) types of content materials (with an indication of 4.89%), but also demonstrating an increase in student learning performance in an adaptable e-learning system (ALELS) while using competency (C) types of content materials (with an indication of 5.43%). These outcomes provide a basis for future design of e-learning systems, utilising different models of learner control based on underpinning educational philosophies, in combination with learning preferences, to structure and present learning content according to type.