However, The Observer XT has not been often used to conduct a sophisticated level of data collection and analysis in design studies, partly due to the expensive fee charged by the software.
In this HTH-01-015
study, we calculated the frequency and duration of participants\' design behaviors using The Observer XT10.0 and exporting the data to an event log, a behavior analysis log and a visualized behavior chart for each participant.
This study attempted to verify and further develop a similar investigation by Ozkan and Dogan (2013), who discovered that the expertise level significantly impacted the selection of source categories and the type of similarity established between source and target. In this study, observations were carried out concerning the field of product design for outdoor furniture. First-year and third-year design students without experience in outdoor furniture design and designers with 2-8 years of experience were recruited in this research. All these experienced designers were employed by outdoor furniture design companies and designing outdoor furniture was their daily work, in that case, we referred to these people as experts with regard to this design problem. On the other hand, as daily encounter, outdoor furniture should be familiar to the participants and student designers could also be able to produce design concepts.
According to the study by Ozkan and Dogan (2013), when asked to select a category, experts overwhelmingly selected near-domain source examples to identify or solve potential problems, whereas first-year students chose mostly distant domain source examples to add new ideas into novel designs. Following their findings, we expected significant difference
between expert designers and student designers in source example selection and reasons behind their selections. The differences in frequencies and durations of analogical behaviors among participants were also expected to be pronounced.
More specifically, the following hypotheses were relevant to this study:H1: Experts would primarily choose near source examples; students would primarily choose distant source examples.H2: The explanations for the selection of source examples would significantly differ among designers with different expertise levels.H3: The frequencies of different behaviors during the analogical reasoning process will significantly differ among groups.H4: The total duration of different behaviors during the analogical reasoning process will significantly differ among groups.
The experiment was devised to test Hypothesis 1 & 2. The images of source examples had three levels, as did the expertise levels of participants. Coding of the explanation for source example selections served as the dependent variables.
21 First-year students (4 male, 17 female) and 31 third-year students (17 male, 14 female) who all majored in industrial design, as well as 12 expert designers (10 male, 2 female) that also majored in industrial design were recruited in this experiment. As all these experienced designers were employed by outdoor furniture design companies and engaged in this field at least two years, we referred to these people as experts with regard to this task.
is often recommended to use random analogical stimuli to make wild connections even between ‘extremely far-field’; domains, which may result in expanded design space and high quality of design concepts. However, this proposal has been directly challenged by the findings of Fu et al. (2013) that random analogical stimuli may be too distant from the design problem to allow for analogy design to occur. Following the studies by Kalogerakis et al. (2010), we selected the source analogs from near, medium, and distant source examples, which correspond to furniture, product and nature categories, respectively.
The source examples in this study were selected in a two-phase process. Initially 300 pictures were chosen randomly from picture sites on the Internet, which consisted of furniture, products and nature. Then the number of pictures was reduced to 100 by rating the soundness, which was defined by Gent