This thesis presents two separate and complementary tools for understanding and analyzing design patterns. The first tool, the High-Level Pattern Representation (HiLPR), exposes the fundamental characteristics hidden within a design pattern's solution. This tool combines the information in parallel patterns' solutions and forces, and integrates information that is critical for pattern implementation. The second tool, the Dynamic Pattern Categorization (DPC), works between all of the patterns in an entire pattern language, and groups patterns of similar characteristics to support analysis and selection. Possible categories are presented and discussed, and further work can combine the exposure of characteristics from HiLPR into categorization by the DPC. The evaluation of these tools highlights a hidden weakness of current design pattern languages and practices. The conclusions raised by this work suggest that there are methods that will support pattern language construction. / Graduate

Das Ziel der vorliegenden Studie war es, zwei neuartige elektrische Zahnbürsten (Braun Oral-B 3D Excel & Cybersonic 2) mit einer Handzahnbürste (Elmex 29) in ihrer klinischen Effizienz bei der Reduktion von Plaque und Gingivitis zu vergleichen. 120 Probanden, deren Mundhygiene Verbesserungspotenzial aufwies, nahmen an der Studie teil, die einem untersucherblinden Parallel-Design folgte. Die Probanden wurden den Testzahnbürsten entsprechend randomisiert auf drei Gruppen verteilt. Nach der Screening-Untersuchung erhielt jeder Teilnehmer eine professionelle Zahnreinigung. Während der Baseline-Untersuchung erfolgte die Objektivierung des Mundhygieneniveaus der Probanden mittels Indizes für den Plaquebefall (QHI, API) und den Entzündungsgrad der Gingiva (PBI). Anschließend verwendeten die Probanden die jeweilige Testbürste acht Wochen lang zweimal täglich für zwei Minuten im Hausgebrauch. Das Mundhygieneniveau der Teilnehmer wurde in einer Zwischenuntersuchung nach vier Wochen und einer Abschlussuntersuchung nach acht Wochen mittels der drei Indizes (QHI, API, PBI) reevaluiert. Die statistische Auswertung zeigte, dass nach vier und acht Wochen, alle Indizes betreffend, die Anwendung beider elektrischer Bürsten in Verbesserungen resultierte, die den Ergebnissen der Handzahnbürste statistisch signifikant überlegen waren (p < 0,001). Die Cybersonic 2 war der Braun Oral-B 3D Excel im Hinblick auf die approximale Plaquereduktion nach acht Wochen (p < 0,05) sowie die Gingivititsreduktion nach vier und acht Wochen unterlegen (p < 0,01). Man kann schlussfolgern, dass Braun Oral-B 3D Excel und Cybersonic 2 Plaque und Gingivitis effektiver als eine manuelle Zahnbürste reduzieren können. / It was the objective of the present study to evaluate the efficacy of two new power toothbrushes (Braun Oral-B 3D Excel & Cybersonic 2) and one manual toothbrush (Elmex 29) in reducing plaque and preventing gingivitis. 120 subjects, whose dental hygiene was open to improvement, took part in the examiner-blind parallel study. Corresponding to the testbrushes they were randomly assigned to three groups. After screening, each participant got a professional tooth cleaning. Four weeks later, at baseline, two plaque indices (QHI, API) and one gingivitis index (PBI) were recorded. Thereafter the subjects used the assigned toothbrushes eight weeks two times a day for two minutes each. The level of dental hygiene was re-evaluated after four and eight weeks. Statistical analysis revealed that all indices showed statistically significant reductions for both power toothbrushes which were superior to the manual brush (p < 0,001). Braun Oral-B 3D Excel was superior to Cybersonic 2 with respect to reduction of approximal plaque after eight weeks (p < 0.05) and reduction of gingivitis after four and eight weeks (p < 0,01). One can conclude that Braun Oral-B 3D Excel and Cybersonic 2 may be more efficacious than a manual toothbrush in removing plaque and preventing gingivitis.

South Africans borrow more and save less than other nations (Discovery Bank, 2018). One reason is a lack of financial knowledge. If a mobile application could guide individuals to modify their financial habits slightly by spending less and saving more, they could dramatically improve their financial future. When designing visualisation systems such as a mobile application interface, users' qualitative design feedback and quantitative usability evaluation are both important and complementary. The benefit of usability feedback in software development is undisputed. The importance of qualitative design feedback from users however, seems to be controversial in Science. Gathering users' qualitative design feedback, ahead of usability evaluation, can have a substantial impact on downstream development costs. The researcher used design as a tool for thinking (imagining new possibilities) and communicating (sharing ideas). The purpose was to clarify ways in which a mobile application interface could support users in making better financial decisions and creating better financial futures for themselves and consequently for society. A user centred design (UCD) approach was followed, emphasising design before development, with a strong focus on user involvement in all three phases, namely requirements gathering, design and evaluation. A primary client archetype for mid-career professionals was developed, split into two personas, Alan and Zoe, based on personality and self-rated motivational attributes which were used in an unconventional way to inspire two parallel, diverse designs. In early design stages, before an idea is well formed, producing multiple contrasting designs in parallel and qualitative design feedback from users is beneficial to establishing utility (solving the right problem), tapping into users' domain knowledge, improving the quality of the design and reducing fixation on one idea. Once the concept has been socialised and evolved sufficiently with users' input, converging on one final design and testing usability (solving the problem in the right way) become more important. This research offers two refinements of the UCD process guidelines for the benefit of researchers and practitioners.

To date, there has been little, if any, holistic evaluation of one of the BankSETA programmes that bank employees are nominated to attend, the Certificate in Management Development (CMD). The programme began in 2010 with the aim of increasing the pool of skills in entry-level management occupations in the banking and micro-finance sector (BankSETA, 2016). In order to verify whether the CMD programme has achieved the purpose for which it was instituted by the BankSETA, it is important that a programme evaluation be conducted. The main aim of this study was to undertake a holistic evaluation of the CMD programme to determine students’ perceptions, their increase in knowledge, their ability to apply on-the-job learning and the impact of the programme on the organisation. It also aimed to identify any barriers or obstacles and to provide recommendations to the BankSETA. Kirkpatrick’s (1996) four-level model of reaction, learning, behaviour and results was used as the theoretical framework for this study. A convergent parallel design was used to gather and analyse the data. The convergent design allowed the researcher to use quantitative and qualitative data-collection methods simultaneously, prioritising the methods equally and keeping each phase independent during analysis, and to then mix the results during the overall interpretation of the findings. The findings of this study highlight the importance of programme evaluation and how indispensable this practice is to the success of any programme. Even though positive results were conveyed at every phase of the study, there are certain areas that can be improved upon in order to maximise the outcomes of the CMD programme. / Business Management / M. Com. (Business Management)