Thesis (M.Tech.: Information technology)- Dept. of Information Technology, Durban University of Technology, 2008. vii, 131 leaves. / This study aimed to explore the impact of cognitive ability on the understanding of computer programming by students enrolled for a programming course at the University of KwaZulu-Natal. The rationale for this study is provided by the general perception held by the academic community that computer programming is a difficult faculty to master. This assertion is corroborated by reports of high failure rates in computer programming courses at tertiary institutes. A literature review was undertaken to investigate the contribution of other factors on the ability to achieve competence in computer programmer. Based on the outcome of the literature review, this study argues that cognitive ability warrants a higher priority relative to the other factors. As a strategy, cognitive science theory was consulted to establish a framework to quantify competency in computer programming. On the basis of this endeavour, two protocols were identified to facilitate the quantification process. The first was the “deep and surface” protocol used in previous studies to ascertain students’ cognitive style of understanding for computer programming. The second was an error analysis framework which was developed as part of the current study. These protocols were used as frameworks to underpin the data collection phase of the study. This study found that at least 50% of the students enrolled in a computer programming course adopt a superficial approach to the understanding of computer programming. In order to explain this phenomenon, a cognitive ability test was administered. Here it was established that at least 39% of these students have not reached a level of cognitive development that will enable the invocation of abstract thought. The study also found that this inability to handle abstractionism, an essential requirement for success in computer programming, is reflected in the severity of errors made in computer programming assessment tasks.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

Software for Moessbauer spectroscopy has been developed for use on the Apple computer. This software is capable of collecting and fitting Moessbauer data when the Apple computer is interfaced with the ND2200 series multichannel analyzer. The software was tested in a study of the feasibility of using Moessbauer spectroscopy to study glass. Results indicate that iron in brown commercial glass and the volcanic glasses behaves similarly. This study concludes that Moessbauer spectroscopy is a promising method for the study of natural and archeological glasses and quality control in the manufacture of commercial glass. / Department of Physics and Astronomy

Metrics calculated during the design phase of a software life-cycle can be used to predict the errors in the software project at an early stage, improve the overall software quality, and increase the efficiency of the software life-cycle.In this thesis, a design metric D(G) for the structure design G is presented. The need and importance of the design metric is shown, the metric is explained in detail, results are given and the solutions are presented to improve the design quality based on the results. A strategy is explained to implement the design metric into a company-wide program. The limitations of the metrics model are also given. A complete model of the software development life-cycle, incorporating the metrics is also presented. / Department of Computer Science