Kullanılabilirlik testlerinde prototip uygunluğu ve kullanıcı uzmanlığı: Taşınabilir navigasyon cihazı ile bir çalışma

Usability of a product, interface or service is a crucial issue in terms of success of design. Several evaluation methods were introduced in the usability literature. Usability testing is the most reliable assessment method compared to other methods such as usability inquiry and usability inspection. Usability tests are must-have parts of the design process. They are more practical to cover the real-world experience and can be applicable to any phase of the design process of physical (consumer products; from cars to chair, computers to lamps…etc.) and digital products (such as; websites, apps, any business software…etc.). Even if the main purpose and outcomes of the usability testing differ according the studied object, in general, the main aim is to identify usability problems. Several factors can change the data provided by usability tests. Based on the Four- Factor Framework of Contextual Fidelity Model (Sauer et al, 2010), these factors are system prototype, user characteristics, test environment and task scenarios. In recent years, studies have been done to find the effects of these factors individually, especially with prototyping fidelity and user expertise. Beside these, quite a few studies in the literature were found for the effect of the combination of these factors. Only one research with a physical product was conducted for both prototype fidelity and user expertise effects on usability testing outputs (Sauer et.al 2010). Although the results of the related research provide some insight about the effects of these two factors, it is not adequate to predict the possible effects especially on digital interfaces. The purpose of this thesis is to contribute to the literature by looking at the influence of prototype fidelity and user expertise on usability testing outputs of a digital interface and interaction between these factors if any. The main contribution with these revelations will be providing knowhow for those who want to design specific usability tests. To simulate the realistic experience and let participants to complete the task without any interruption, "Retrospective Think Aloud Protocol" and "Performance Measurements" were used to gather data to achieve the desired goals. Twenty participants were used in total divided into four groups with five participants each; 5 novice and 5 expert participants worked with the low fidelity prototype (paper) while another 5 novice and 5 expert participants worked with the high fidelity prototype (the device itself). By doing this evaluation, usability problems were analyzed in terms of number, severity and variety of types. In addition, performance data (time on task, success rate) is also evaluated. Accordingly, it becomes possible to comment on which user groups takes part actively in what kind of prototypes. Although the sample size of five participants per each group has little statistical power, the results showed that, novice users found significantly more usability problems than experts in total and the difference between novices and experts was higher under low-fidelity prototype rather than high fidelity one. The study also showed that in order to understand the differences between the tested groups (novices and experts) and interfaces (low and high fidelity), usability problems must be isolated from each other. By doing this, it is important to consider on the cause of the problem to define types and subtypes for the classification with descriptions in detail. According to the result of the study, significant differences were found between user groups with the categories; use flow, menu categorization, interactive components and aesthetic & visual. In addition, content related problems were found significantly more in low fidelity prototypes and the interaction between expertise and fidelity is found significant with in the category of "aesthetic and visual". Lastly, experts showed better performance with the time spent on test.