Information Technology Reference
In-Depth Information
In addition, the participants had to give a rating to “My overall grade
for the program” on a seven-point scale (1=worst grade, 7=best grade).
Fig. 11.11 illustrates ratings given by participants as pairs of boxplots.
In the case of each pair, the left boxplot represents ratings given after the
interview, and the right boxplot represents ratings given after longer-term
use. After the interview (Fig. 11.11), participants felt that the prototype
looked very good (median 6.0), was easy and fun to use (medians 6.0 and
5.0, respectively), and that it presented music in a novel way (median 5.5).
The findings are well in line with participants' first impressions. These
results (or other results of the study) did not correlate with participants'
age or gender. We also experimented with Chi-Square and t-tests, but the
conditions for the test were not met in any of the cases. Ratings for the
other statements (see the left boxplots on the lower picture of Figure
11.11) were also on the “positive” side, i.e., the median values were
greater than four. In other words, the prototype looked quite promising
based on participants' first experiences. However, the situation changed
for the worse after participants had tested the prototype at home for
longer-term use. While the idea was still considered to be novel, the
median of looks decreased from 6.0 to 5.0, and participants also
considered the application to be less fun. Using Wilcoxon's signed ranks
test, the Z values for looks and fun were -3.47 and -3.16 with p<0.01 (2-
tailed), meaning the changes were also statistically significant. In longer-
term use, the prototype was seen to be too simple and to lack important
features (see below for ideas for future development). For example, the
lack of a text based search function was a major shortcoming to 68% of
the participants (56% after the interview). This is in line with the findings
of Karapanos et al. (2008), where it was concluded that “Overall, while
early experiences seemed to relate to hedonic aspects of product use,
prolonged experiences became increasingly tied to aspects reflecting how
the product becomes meaningful in one's life.” Logically, participants who
felt that the prototype had an extensive set of well-implemented features
also believed that the appeal of the program will last and that it was fun to
use (Spearman's rho=0.49 and rho=0.77, significant at the p<0.01 level (2-
tailed)).
The median value of lasting appeal decreased dramatically from 5.0 to
3.0, and only 45% of participants stated that they would like to use the
prototype for a longer time (73% after the interview). Using Wilcoxon's
signed ranks test, the Z value for lasting appeal was -2.87 with p<0.01 (2-
tailed), meaning that the change was also statistically significant. In
addition to the lack of certain important features such as the textual search,
other reasons for this included technical problems (e.g., program crashes
Search WWH ::
Custom Search