Geoscience Reference
In-Depth Information
These Likert type statements (a quantitative assessment of satisfaction) were
complemented by open-ended (qualitative) questions on advantages, drawbacks,
suggestions for future development and any other comments on the environments.
Statistical analysis was applied to the results. The Chi Square (satisfaction) and
ANOVA tests (efficiency, effectiveness) in particular were run to ascertain if there
was any significant difference between results for GIS, CAD and VR, in terms of
the three usability dimensions. The three technologies were also compared on the
basis of spread (standard deviation).
3 Results
3.1 Participant Experience
There were 16 participants in this study, either working in the urban planning
and design field or senior students, having completed the urban design paper sup-
plemented by practical experience. Most had used CAD software for their urban
design tasks, a couple GIS and a couple more pen and paper. Only five had already
applied 3D functionality in their role.
CAD was rated as being powerful for tasks such as precise drawing and
structuring (e.g. for road design), but was not too user-friendly. Despite not many
using GIS for urban design, all had some experience of GIS in other contexts, used
for cartography, analysis and spatial data management. The learning curve was an
issue here, however, as well as speed. Although all participants had heard of VR,
only two had used it, one for public briefings i.e. as a communication tool.
3.2 Efficiency
Table
1
and the plot in Fig.
10
show the comparative results for efficiency. The
times for each of the four specific tasks are shown. On the whole, completion of
the tasks in the VE was on average quicker than in CAD and GIS environments.
The ANOVA test marked the difference between VR and each of the other types
as being statistically significant (
p
< 0.05). There was no significant difference
between performance in GIS and CAD.
3.3 Effectiveness
Figure
11
shows the comparative results for effectiveness. This was measured in
terms of percentage of tasks completed without error or verbal instruction. The
plot shows that there is visually not much difference between the three technolo-
gies, which is reinforced by non-significance (
p
< 0.05) in the ANOVA test results.
