| dc.description.abstract | Start-ups and new product development projects are fuzziness and ambiguity set in motion. In the chaos of these early phase projects, more and more research is turning activities like prototyping, marketing, funding and production into a science, but there is still much left to chance. This master thesis is an attempt to add to the science of early stage development activities and take them out of the black box, trying to understand how to successfully create communication type prototypes for technically trained audiences vs. financially trained audiences. During the work on this problem, I discovered that a more interesting relationship is between those who possess knowledge about the technical aspects of a prototype, and those who do not. This is due to the fact that one s education and work experience does not provide a complete picture of one s knowledge. Also, in order to understand what determines an evaluation one has to understand what aspects of a prototype affects the judgement, and in what degree. Thus, in addition to background, I wanted to understand how design resolution and function affects the judgement of the audience.
To understand the evaluations of prototypes I designed an experiment using real start-up prototypes: Wiral, a lightweight cable cam system and MovieMask, a face accessory for mobile devices to simulate a cinematic experience. The experiment was used to gather data on the effects of background, design resolution and function on prototype evaluations. The experiment was then run with 18 participants with different professional backgrounds at different locations in Norway and the US. I then performed a statistical analysis on the data registered from the experiments. The experiment setup, participant requirements and time restrictions resulted in a low sample size, reducing the impact of the statistical analysis in this thesis. However, the statistical analysis did reveal some statistically significant results for background, design resolution and function. There was not enough evidence to conclude the effect of these variables, but there was an interesting correlation between the tests showing that the evaluations of MovieMask and Wiral had large differences in distributions, when testing for the effects of background, design resolution and function indicating that the concepts themselves are affected differently. Lastly I present interpretations of results, implications, limitations of the study and a recommendation for further work on the subject. | |
