Date of Award


Degree Type


Degree Name

Doctor of Philosophy


Faculty of Business and Public Management

First Advisor

Associate Professor Dieter Fink


Effective and efficient training is a key factor in determining the success of end user computing (EUC) in organisations. This study examines the influences of two application interfaces, namely icons and menus, on training outcomes. The training outcomes are measured in terms of effectiveness, efficiency and perceived ease of use. Effectiveness includes the keystrokes used to accomplish tasks, the accuracy of correct keystrokes, backtracks and errors committed. Efficiency includes the time taken to accomplish the given tasks. Perceived ease of use rates the ease of the training environment including training materials, operating system, application software and associated resources provided to users. In order to facilitate measurement, users were asked to nominate one of two approaches to training, instruction training and exploration training that focussed on two categories of users, basic and advanced. User category was determined based on two questionnaires that tested participants' level of knowledge and experience. Learning style preference was also included in the study. For example, to overcome the criticisms of prior studies, this study allowed users to nominate their preferred interfaces and training approaches soon after the training and prior to the experiment. To measure training outcomes, an experiment was conducted with 159 users. Training materials were produced and five questionnaires developed to meet the requirements of the training design. All the materials were peer reviewed and pilot tested in order to eliminate any subjective bias. All questionnaires were tested for statistical validity to ensure the applicability of instruments. Further, for measurement purposes, all keystrokes and time information such as start time and end time of tasks were extracted using automated tools. Prior to data analysis, any 'outliers' were eliminated to ensure that the data were of good quality. This study found that icon interfaces were effective for end user training for trivial tasks. This study also found that menu interfaces were easy to use in the given training environment. In terms of training approaches, exploration training was found to be effective. The user categorisation alone did not have any significant influence on training outcomes in this study. However, the combination of basic users and instruction training approach was found to be efficient and the combination of basic users and exploration training approach was found to be effective. This study also found out that learning style preference was significant in terms of effectiveness but not efficiency. The results of the study indicates that interfaces play a significant role in determining training outcomes and hence the need for training designers to treat application interfaces differently when addressing training accuracy and time constraints. Similarly, this study supports previous studies in that learning style preferences influence training outcomes. Therefore, training designers should consider users' learning style preferences in order to provide effective training. While categories of user did not show any significant influence on the outcomes of this study, the interaction between training approaches and categories of users was significant indicating that different categories of users respond to different training approaches. Therefore, training designers should consider the possibility of treating differently those with and without experience in EUC applications. For example, one possible approach to training design would be to hold separate training sessions. In summary, this study has found that interfaces, learning styles and the combination of training approaches and categories of users have varying significant impact on training outcomes. Thus the results reported in this study should help training designers to design training programs that would be effective, efficient and easy to use.