Repository logo
 

Faculty of Accounting and Informatics

Permanent URI for this communityhttp://ir-dev.dut.ac.za/handle/10321/1

Browse

Has files: YesSubject: search.filters.subject.Computer-assisted instruction--South Africa

Search Results

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    Early prediction of students at risk in a virtual learning environment using ensemble machine learning techniques
    (2021-12-13) Soobramoney, Ranjin; Singh, Alveen
    Students at risk (SAR) are those students who are considered to have a higher probability of failing academically or dropping out of an academic programme. The literature reveals that SAR is a global problem at Higher Education Institutions (HEIs). A high failure rate can not only harm the reputation of the HEIs, but if left unchecked, can be detrimental to these HEIs. The problem of identifying SAR is a pervasive and persistent one. However, early identification of SAR will allow for timely and focused interventions, thereby reducing the problem. Various techniques have been used by HEIs to identify SAR. The traditional statistical approach is one such technique. One of the key challenges with this technique however, is that it often requires a large amount of manual analysis of the data to predict SAR, which in turn also makes early predictions of SAR more computationally challenging. To overcome some of the challenges of the traditional statistical approach, machine learning-based techniques have been proffered to predict SAR. Since machine learning (ML) models are based on the input data rather than the underlying problem, they are expected to have better predictive capabilities than traditional statistical models. Several ML-based techniques have been applied to predict SAR with varying degrees of success. This study proposes the use of ensemble ML techniques for early and accurate prediction of SAR using students’ demographic and weekly online Virtual Learning Environment (VLE) data. Aggregating the predictions of a group of ML classifiers is expected to provide a better generalization performance than each of the individual classifiers on their own. The use of ensemble ML techniques for this study will provide an improved solution to the problem of predicting SAR. To this end, this study focused on training forty different ML predictive models, one for each week of the semester, using twenty-five different ML classifiers. Each model was trained using students’ demographic data combined with data from their weekly interactions with a VLE. Based on the training results, four classifiers, namely AdaBoostClassifier, LGBMClassifier, RandomForestClassifier, and XGBClassifier were selected as base learners for the ensemble classifier. Hyperparameter optimization was performed using Random Search on each of the four classifiers. These classifiers were then used to create a voting classifier ensemble for each of the forty weeks, with 10-fold cross validation being used to evaluate the predictive models. The results show that the voting classifier ensemble method outperformed the individual classifiers overall over forty weeks and can thus provide an improved solution to the problem of predicting SAR.
  • Thumbnail Image
    Item
    Technologically disadvantaged students’ perceptions of blended learning in a higher education institution : the case of students at Walter Sisulu University
    (2021-11-10) Gqokonqana, Onke; Cloete, Melanie Bernice
    Students at Walter Sisulu University come from rural areas and are generally unfamiliar with technology as a learning tool. Similarly, Walter Sisulu University is categorised as a historically disadvantaged institution of higher learning, with face-to-face instruction being the preferred approach. Under apartheid, historically disadvantaged institutions were founded to meet the educational “needs” of the former “rural homelands,” which were marked by demographic areas and market variation in comparison to their affluent equivalents, resulting in social hierarchies. The conventional chalk-and-talk technique of teaching and learning has become less effective as more educational institutions integrate technology in teaching and learning. Walter Sisulu University is attempting to incorporate technology into teaching and learning by implementing blended learning, which is the addition of online components to the traditional face-to-face form of instruction. The learning management system was not completely utilized despite the university purchasing a license for Blackboard customised as ‘Wise-up' at Walter Sisulu University seven years ago. It was implemented three years ago in the Accounting and Finance department. The purpose of this study was to examine the technologically disadvantaged students' perspectives of blended learning particularly for Cost Accounting 2 with students from a technologically disadvantaged background and the use of blended learning. This is critical for determining the learning management system modifications that must be made as well as the course design on its own. The data was collected using a quantitative technique from all second-year National Higher Certificate: Accountancy students at Walter Sisulu University. This census approach was chosen because it allowed the study to reduce sampling error by allowing all registered students to participate in the survey. According to the quota of registered students, 400 students were supposed to fill out the survey, however, only 119 (n=119) did. The data acquired through the use of the 'Question pro' application was analysed using a statistical package for social science version 25. The study found that blended learning is an effective model to learn Cost Accounting 2 because the institution gave enough information on how to use the system. Students benefit from blended learning because it allows them to be involved in their studies. To be fully matched with blended learning standards, changes were made to the Cost Accounting 2 module. A revamp of the course guide could be beneficial because it will make it apparent to students what will be covered in face-to-face mode and what will be covered through the usage of the learning management system. Due to connectivity concerns, some students expressed dissatisfaction with the use of the learning management system. As a result of the Coronavirus epidemic, the study used an online questionnaire instead of faceto-face as lectures were in suspension. The study was confined to National Higher Certificate: Accountancy students because the goal was to learn about the students’ perceptions of Cost Accounting 2 through blended learning. Future research could look into the perceptions of blended learning among the entire Accounting Department’s students, as accounting-related disciplines differ at times. Other methodology could also be used to explore students’ perceptions of blended learning.
  • Thumbnail Image
    Item
    Developing a web based blended learning technique to improve computer programming competence of information technology students
    (2017) Jackson, Priyalushinee; Olugbara, Oludayo O.
    Computer-programming dexterity is an essential skill for students of computer science, information technology and engineering who are intrinsically expected to be able to do programming. However, teaching and learning computer-programming concepts and skills have been recognised as a great challenge for both teachers and students, for many reasons. Computer-programming requires new ideas in thinking and conceptualising practical solutions. It requires creative skills in solving practical, but often difficult problems. Moreover, computer-programming students, generally lack problem-solving skills and self-efficacy. They typically find it difficult to use artificial programming languages to solve challenging problems. There is the problem of poor background in science and mathematics that would help students to rapidly understand the intricacies of computer-programming. Students are not motivated to overcome the fear of the bizarre syntax of computer-programming codes. These challenges, coupled with the huge potential of computing applications in the society have made the development of effective pedagogies and environments for computer-programming courses, an important issue. To address this issue in a unique way, this study proposes to explore a web-based, blended learning technique with minimal instructor intervention, to improve the computer-programming competence of information technology students. These students are expected to have developed an acceptable level of computer-programming competence at university to be job ready and to be self-reliant in their future careers. The technique being proposed in this study was implemented in a blackboard ™/®/© web-based environment. The effectiveness of the technique was demonstrated using experimentation coupled with the data analysis method that is based on the three-parameter item response theory and retrospective pretest. The method used in this study to evaluate computer- programming competence of students reflects the perspective of the students in the evaluation process. The results of the study indeed show that using the proposed technique, information technology students dynamically collaborate with their peers with minimal instructor intervention towards improving their computer-programming competence.
  • Thumbnail Image
    Item
    A study of factors shaping learners' perceptions of ICT-based teaching and learning by applying personality and technology adoption theories on indigenous knowledge students
    (2016) Ngcobo, Khumbuzile Mornica; Eyono Obono, Seraphin Desire
    Existing literature indicates that the use of Information and Communication Technologies (ICTs) and the inclusion of Indigenous Knowledge Systems (IKSs) in the school curriculum have the potential to increase academic performance. However, formal education is still unable to integrate ICTs into the teaching and learning of school subjects, especially, those that are related to IKSs. This research therefore aims to construct a model of the factors shaping learners’ perceptions on the usefulness of ICTs for the teaching and learning of IKSs. This aim is sub-divided into four research objectives: (a) to identify appropriate technology diffusion theories for the investigation of the factors shaping learners’ perceptions on the usefulness of ICTs for the teaching and learning of IKSs, (b) to construct a theoretical model of the factors shaping learners’ perceptions on the usefulness of ICTs for the teaching and learning of IKSs, (c) to perform an empirical confirmation of the above announced theoretical model of the factors shaping the perceptions of learners on the usefulness of ICTs for the teaching and learning of IKSs, and (d) to suggest new ideas for future research on learners’ perceptions on the usefulness of ICTs for the teaching and learning of IKSs. A review of existing literature on eLearning adoption by students and learners was conducted in order to achieve objectives a, b, and d. As for objective c, it was achieved through the survey of 115 Hospitality studies learners from the ILembe and UMgungundlovu municipality districts in the KwaZulu-Natal (KZN) province of South Africa. The study’s findings can be summarized as follows: (a) The Technology Adoption Model (TAM) is the backbone of the model designed by this study on the factors affecting learners’ perceptions of the usefulness of ICTs for the teaching and learning of IKSs; (b) Learners’ perceived usefulness of ICTs is hypothetically affected by the following factors: demographics, computer self-efficacy, trust in ICTs, and level of conscientiousness, (c) All these factors were empirically confirmed through a survey conducted by the current study, except that the only validated demographics were : school location, cell phone access, class grade and preferred subject; (d) This research recommends further investigation on the factors affecting learners’ perceived usefulness of ICTs for the teaching and learning of IKSs, mainly because of the insufficient literature on this subject.
  • Thumbnail Image
    Item
    A SCORM compliant e-learning content prototype for the training of OBE mathematics educators in the context of developing countries
    (2010) Piyose, Xolisa; Eyono Obono, Seraphin Desire
    The main purpose of this study was to examine how e-learning can help resolve some of the most acute problems that are specific to the nature of the outcomes-based education (OBE) system in developing countries. This was accomplished by investigating the relevant literature on OBE and by designing an e-learning content prototype for the South African version of OBE, with the focus on the training of Mathematics and Mathematical Literacy educators. OBE is an education system centred on the theory of mastery of learning introduced by Bloom in the 1950s. It has been implemented worldwide in primary and secondary schools and also in tertiary institutions. Some studies have shown that OBE is problematic, and that both educators and learners are opposed to this system of education. Existing research also reports that the quality of education in OBE is very poor as compared to that of the traditional education system. This study was an attempt to identify the most acute problems experienced by educators and learners in the OBE context and to design a prototype for e-learning content that can be used in courses in an attempt to solve these problems. The research population of the perceptions survey consisted of Mathematics and Mathematical Literacy educators and learners from the 6 000 primary and secondary schools of the KwaZulu- Natal (KZN) province of South Africa. Eighteen schools formed part of the research sample, with an average of two schools per region (the KZN province is divided into nine regions). Both private and public schools were included in the study. Data for the study were collected from March 2007 to August 2007 in the form of a perceptions survey of 104 educators and 288 learners, yielding an average of six educators and sixteen learners for each randomly selected school. Results from the perceptions survey show that educators and learners do not understand OBE terminology. In addition, educators claim that they are not sufficiently trained for OBE. Research iii participants also report that schools’ basic infrastructure is unsatisfactory, and that their classrooms are overcrowded. Mathematics is perceived as the most difficult subject by both educators and learners. The survey also reveals that most public schools do not have computers and that neither educators nor learners have access to computers in their public libraries. A SCORM- (Sharable Content Object Reference Model) compliant e-learning course was developed in this study to address the most acute problems identified by the survey, based on the Software Engineering Unified Model. The designed e-course contains OBE terminology such as learning outcomes, OBE principles, assessment standards, assessment methods, national curriculum statement and learning fields. The e-learning course content also contains the Mathematics and Mathematical Literacy curriculum for grades 11 and 12. It was constructed using 16 documents extracted from the National Department of Education’s website: 7 documents under Further Education and Training, 4 under Teacher Guide, and 5 under General Education and Training. The evaluation of the e-learning content prototype was conducted through a survey among 36 educators from different primary and secondary schools of the Mnquma Municipality of the Eastern Cape province of South Africa. They were trained in the SCORM-compliant e-learning course content at the Walter Sisulu University’s Butterworth campus. The training took place from 10 to 13 November 2009. After the training, educators filled out a questionnaire on their perceptions of the effectiveness of the proposed e-learning content prototype with regard to the practice of OBE. Results from the SCORM e-course evaluation survey showed that the proposed SCORM software artefacts allow educators to have a better understanding of OBE terminology. The proposed software artefact is user-friendly and educators recommended its use not only for Mathematics but for all subjects.