Au cours des dernières années, l'intérêt porté aux performances de la navigation internet n'a cessé de croître au sein de la communauté scientifique. Afin de comprendre la perception qu'a l'utilisateur des performances de la navigation Web, au cours de cette thèse nous abordons différents problèmes liés aux performances de la navigation Internet telle qu'elle est perçue par l'utilisateur final. Cette thèse se compose de trois parties: la première partie présente notre nouvelle plateforme basée sur des mesures faites au niveau du navigateur. Nous présentons les différents paramètres que nous obtenons facilement à partir du navigateur, ainsi que des indicateurs du ressenti de l'utilisateur final. Ensuite nous utilisons des techniques de partitionnement de données afin de trouver les corrélations existantes entre performances de haut niveau et de bas niveau. Dans une seconde partie, nous présentons notre outil de diagnostic appelé “Firelog”. Nous étudions tout d'abord les différentes causes qui peuvent affecter le rendu d'une page Web. Ensuite nous décrivons en détails les différents composant de notre outil et les mesures qu'il effectue. Sur la base des paramètres mesurés, nous illustrons notre modèle pour le diagnostic des performances d'une manière automatique. Dans la dernière partie, nous proposons une nouvelle méthodologie, “Critical Path Method” (ou Méthode du Chemin Critique) pour l'analyse des performances de la navigation Web. Nous expliquons d'abord en détails les caractéristiques intrinsèques du Navigateur lors du rendu d'une page Web, puis nous présentons formellement notre méthodologie. / In recent years, the interest of the research community in the performance of Web browsing has grown steadily. In order to reveal end-user perceived performance of Web browsing, in this thesis work, we address multiple issues of Web browsing performance from the perspective of the end-user. The thesis is composed by three parts: the first part introduces our initial platform which is based on browser-level measurements. We explain measurement metrics that can be easily acquired from the browser and indicators for end-user experience. Then, we use clustering techniques to correlate higher-level performance metrics with lower level metrics. In the second part, we present our diagnosis tool called FireLog. We first discuss different possible causes that can prevent a Web page to achieve fast rendering; then, we describe details of the tool's components and its measurements. Based on the measured metrics, we illustrate our model for the performance diagnosis in an automatic fashion. In the last part, we propose a new methodology named Critical Path Method for the Web performance analysis. We first explain details about Web browser's intrinsic features during page rendering and then we formalize our the methodology.

The purpose of this study was to investigate the use electronic portfolios (e-portfolios) in private tertiary institutions in South Africa to enhance the learning experience of scholars. The study adopted an interpretive ontological stance and subjective epistemological stance as the researcher sought to understand how-portfolios can be used to enhance learning. The focus was on understanding people in a social setting though employing field research to collect and interpret data. Action research, in this case employing a pre-test, intervention and post-test limited to a single cycle, was selected as strategy. The empirical part of the study was conducted during the first semester of 2014 (January to May) at a private tertiary university in Gauteng, South Africa. A pre-survey has been conducted before implementation and a post-survey after the implementation of e-portfolios to test the attitude of the participants, where after the results were compared. The pre-test was administered before creating e-portfolios and the post-test towards the end of the semester after developing and maintaining e-portfolios. Questionnaires were used to obtain information such as the attitude of the respondents on learning, assessment, reflection, collaboration and interpersonal communication through the use of e-portfolios. Questionnaires served as the main data collection tool throughout the survey. The sample was selected from first and second year students in the Information Technology faculty, with the sample size n=48. The researcher further collected data through observing subjects and conducting group interviews that were recorded to augment the questionnaire approach and verify the results through triangulation. The questionnaires were distributed to the participants who were given thirty minutes to complete. Responses were made using a five-point Likert scale and open-ended questions. The response rate was outstanding as all 48 subjects participated in the study. The researcher used a mixed-methods approach for this study; SPSS 22.0 was used for quantitative data analysis and thematic analysis was employed for the qualitative data. Reliability for quantitative data was determined using Cronbach’s alpha. Regression analysis was done as well as correlations to determine relationships and associations. To further cement relationships, t-tests and Spearman’s correlation coefficient were employed. Trustworthiness for qualitative data and justification for credibility, dependability, transferability and confirmability were discussed. vi The results revealed that e-portfolios could be used effectively at private tertiary institutions to enhance the learning experience of scholars as evidenced by the significant statistics obtained in this study.