Blended learning constitutes a key element of the NMMU’s Vision 2020; therefore, creating the expectation that academics from the NMMU will implement and utilise blended learning as part of their teaching pedagogy. Implementation reports drawn from Moodle (the learning management system utilised by NMMU) indicates that there are still numerous departments that are not fully utilising blended learning. The aim of this study was thus to determine the factors that impact on the acceptance and implementation of blended learning by academics at the NMMU. A modified version of the Technology Acceptance Questionnaire was electronically distributed to academics from the various NMMU campuses in order to collect the empirical data, with purposeful non-probability sampling being employed. The results indicated that factors, such as perceived usefulness, the student-centered nature of blended learning, the appeal of such environments to students, accessibility, as well as administrative- and communication benefits, had a positive influence on blended learning adoption. Lack of time and lack of knowledge with regard to blended learning use and implementation, concerns regarding accessibility, and lack of capacity and technical support were cited as issues that hampered blended learning implementation.

An analysis of the literature showed that there is very little agreement on when and how a computer program should branch a student through a CAI program. This, together with the fact that research in the field of arithmetic has shown that drill should follow effective teaching of concepts, led the author to investigate whether students working on arithmetic drill-and-practice would do better on a COMPUTER-PACED program or a SELF-PACED program. COMPUTER-PACED was defined to be where the computer program determined when the students should be branched to more or lass difficult questions. SELF-PACED was defined to be where the students determined when they were presented more or less difficult questions by pushing one of the two marked keys on the computer terminal. The evaluation was done by comparing the achievement of the COMPUTER-PACED and the SELF-PACED groups. For the length of the study the two groups of grade six students had a daily arithmetic lesson followed by a session at a computer terminal to work on arithmetic drill-and-practice programs. The results of the post-test (adjusted by using a pre-test as a covariate) showed that there was no significant difference between the two selection mechanisms. Further analysis showed that there was no significant difference between the males and females performance and that there was no significant interaction (sex X groups) effect. The results of the study indicate that when working with arithmetic drill-and-practice, students will do as well if the computer program controls when to branch as they would if the students control when to branch to a different level of difficulty. / Education, Faculty of / Graduate

This study was motivated by the belief that teaching a student in a challenging way would increase his ability to apply what he had learned to new, though related, problems. A specific challenging teaching strategy was chosen, which attempted to challenge all students appropriately, and to give the minimum amount of help. It was administered by the computer, which considerably facilitated the use of such an individualised strategy. The evaluation was done by comparing the effects of the challenging teaching strategy with those of a linear program, also computer-administered. A linear program was considered to exemplify an unchallenging approach. Both programs taught elementary base five arithmetic to Grade Six students, the students being assigned to the programs at random. The effects of the two strategies were then measured by means of a post-test. This aimed at evaluating both the grasp of the basic material and the ability to extrapolate from it to solve new problems in the same general subject area. The results of the post-test showed that both strategies succeeded in teaching the basic material equally well, so that neither strategy gave the student an advantage in this respect. However, the challenged group of students showed far greater ability to extrapolate from the material than did the linear program group, with an average, score over 45% better. This was significant at the .007 level. These results suggest that further investigation of the merits and application of a challenging teaching strategy should be eminently worthwhile. / Education, Faculty of / Graduate

With the increased use of computerized databases, the ability to allow users to access information using natural language is becoming more desirable. There are many natural language systems in existence today. The main problem with these systems is the amount of expertise and effort required to adapt them to a new domain. The design of a portable natural language front-end to a relational database is described in this thesis. It is designed so that a typical Database Administrator can implement a new domain in a reasonable amount of time. Database portability has been achieved by separating the domain dependent natural language definitions from the domain independent definitions. These domain dependent definitions are specified in the database schema, which is structured to extract the semantics contained in the structure of the actual database. A rich supply of standard definitions are available to both aid in the development of the database schema and to help force consistency amongst database domains. / Science, Faculty of / Computer Science, Department of / Graduate

This study addresses the effectiveness of educational microcomputer software with hearing impaired students. A review of the literature revealed a large void in the empirical research on this topic and also indicated that a lack of suitable and appropriate software, was a major hurdle to the successful use of CAI with hearing impaired students. This study investigated the effectiveness of a specific educational microcomputer program (MECC : Trapezoids and Triangles Areas) with a group of hearing impaired students. The 18 students were randomly divided into two groups. One group viewed the tutorial portion of the software and then worked with the practise section. While the other group worked only with the practise section of the software. This was done to measure and compare the effectiveness of both portions of the program. The students were carefully observed as they interacted with the computer. Problems they encountered with the software, their achievement during exposure to the software and on pre and post tests, were all recorded by the experimenter. These data were used to measure the changes that took place due to microcomputer application, to evaluate the effectiveness of this software, to make recommendations regarding the characteristics required to improve this software, and to set "guidelines" for the development of future educationally effective software for the hearing impaired. The results revealed the need for software that contains a measured and controlled level of syntax, more advanced diagnostic and remedial capabilities, and a simplified more thorough presentation of the information. This study demonstrated the need for more research related to identification of the important factors and teaching strategies which make software more educationally effective for use with hearing impaired students. / Education, Faculty of / Educational and Counselling Psychology, and Special Education (ECPS), Department of / Graduate

Two methods for improving the quality of Computer Assisted Instruction are examined. They are: using Intelligent Computer Assisted Instruction techniques to make the CAI system more flexible, and using graphics to increase the efficacy of teaching. Two computer systems for teaching the Logic Programming language Prolog were developed. The first is an ICAI system which uses the prerequisite relationships of the course material to plan a course of study. It distinguishes between methods of instruction and topics of instruction, giving students a great deal of freedom in choosing either one. The second is an animated trace which graphically illustrates the execution of Prolog programs. Information is displayed in three windows — one for Prolog goals, one for the database, and one for output from the program being traced. Results indicate that ICAI and graphics can both be used effectively in the teaching of programming languages, particularly in combination. / Science, Faculty of / Computer Science, Department of / Graduate

Recent developments in interactive Computer-Aided -Instruction and in Artificial Intelligence have enabled teaching machines and programs to deal reasonably effectively with the subject matter to be taught. Presented herein is a proposal and design for an intelligent LISP teaching machine. It is expected that such a system would be used in conjunction with other conventional methods to teach students, with some prior programming knowledge, the LISP programming language and "correct" programming style. With the belief that procedural knowledge is best learned by 'doing', this CAI system will integrate instruction in concepts, LISP syntax and semantics; instruction in the design of LISP functions and code; and analysis of students' solutions and consequent error correction. The goal of this LISP tutor is simply to act like a human tutor - cognizant of what the student is doing all the time, and able to provide advice and give direction where necessary. / Science, Faculty of / Computer Science, Department of / Graduate

Please read the abstract (Summary) in the section 00front of this document / Thesis (PhD (Computer Aided Education))--University of Pretoria, 2007. / Humanities Education / PhD / Unrestricted

Low reading proficiency rates have been observed throughout the United States including a rural school in southwestern New York State. Although the Outstanding School District (pseudonym) purchased i-Ready, an adaptive diagnostic and instructional program, only 35% of students in Grades 3 through 8 demonstrated proficiency in reading in 2018. The problem is that i-Ready has not been implemented as intended. The purpose of this case study was to investigate teachers' concerns related to implementation and use of i-Ready. The guiding research questions examined the teacher's most common concerns and challenges faced during implementation. The Concerns-Based Adoption Model served as the theoretical framework, specifically using the Stages of Concern dimension to discover feelings and perceptions of teachers. Purposeful, criterion-based sampling methods were applied resulting in 8 teachers being selected. Quantitative data were collected using the Stages of Concern Questionnaire. Analysis included converting raw scores to percentile scores, plotting each, and visually representing the findings. The results will provide information needed to make decisions about the use of the program, the challenges encountered in implementation, if professional development is needed, or if the program should be discontinued. Data analysis indicates concerns primarily in the “self” category indicative of limited use of i-Ready and challenges during implementation include limited time and information necessary about how to effectively use the program. The implications for social change include the potential of adding to existing literature on effective innovation implementation and further developing the knowledge base on effective reading interventions, which will lead to enhanced academic success and the ability of students to become productive members in their communities and societies.

Computer-based tests facilitate the collection of problem-solving processes, also known as process data. Response processes recorded in computer log files provide a new venue for investigating and understanding human behaviors. This thesis focuses on the development of statistical learning methods for process data and considers the following three problems. The first problem is feature extraction. Response processes are noisy and of non-standard formats. To exploit information in process data, we propose two generic methods that summarize response processes to vectors so that standard statistical tools such as regression models are applicable. In Chapter 2, features are extracted using multidimensional scaling and a pairwise dissimilarity measure of response processes. Chapter 3 utilizes autoencoder and recurrent neural network to explore the latent structure of process data. For both methods, empirical studies show that the extracted features preserve a substantial amount of information in the observed processes and have greater predictive power for many variables than the traditional item responses. The second problem is assessment based on process data. We present a statistical procedure in Chapter 4 that incorporates process information to improve the latent trait estimation of item response theory models. The procedure is data-driven and can be easily implemented by means of regression models. Theoretical guarantee is established for the mean squared error reduction. Application of this new process-data-based estimator to a real dataset shows that it achieves higher reliability than the traditional item-response-theory-based estimator. The third problem is identification of problem-solving strategies for exploratory analysis. The approach presented in Chapter 5 segments individual process into a sequence of more homogeneous subprocesses using action predictability. Each subprocess is associated with a subtask whereby long and complex response process can be transformed into shorter and more interpretable subtask sequence. Using this approach, problem-solving strategies can be visualized and compared among groups of respondents and process information can be decomposed for further analysis.