Almost all of the
beta test subjects participated in this follow-up assessment. Therefore,
it can be considered a complete picture of the test process without sampling
bias. Overall, the participants found PBCAT to be utilitarian, easy to
understand and simple to use. All would readily use it if they had a need
to analyze pedestrian or bicycle crashes with motor vehicles.
The ability to extend
the beta test results beyond the test population could be limited by several
factors. First, the subject test participants did not include people who
were already familiar with and using pedestrian and bicycle crash types.
This led to novelty and learning factors which could have increased interest
in and acceptance of PBCAT. Second, there were no police officers or police
support personnel in the test. These are groups that traditionally do
extensive work with crash data. Third, the participants were quite high
in computer literacy. Fourth, the beta test only covered the entry and
crash typing modules. There were no structured exercises for the other
modules, although some of the subjects did, in fact, use them.
Although all four
of the foregoing factors could potentially limit the ability to generalize
the results of the beta test, it is fair to conclude that their effects
were not debilitating. Although there was a clear novelty and learning
effect from the test, the participants volunteered that they considered
PBCAT to be a useful long-term tool. They also were convinced that PBCAT
was a time and work saver. This assessment of effectiveness should readily
transfer to police or other groups that would have need of the functionality
It should be noted
that law enforcement personnel were also recruited to participate in this
effort, since they would be another group that may be a candidate user
for the software. Unfortunately, none of the individuals contacted were
able to participate. The absence of police personnel in the beta test
group appears to be largely an issue of face validity. While police departments
are regular users of crash data, they do not typically conduct analyses
that would differ markedly from those of interest to any other groups
concerned with pedestrian or bicycle safety. On the contrary, it is reasonable
to postulate that planners and pedestrian/bicycle coordinators involved
in safety efforts would have interests that are quite similar to those
of police personnel.
The high computer
literacy of the test group certainly assisted them in getting beyond the
few program bugs that surfaced in the test version of the software. Beyond
that advantage, the above average computer skills of the participants
appeared to have little impact on the beta test. Moreover, even though
these people were expert, they clearly understood the limitations of novice
computer users and stated the belief that even a beginner would have no
problem with PBCAT, particularly after it was installed.
Since the test only covered the entry and typing modules, it could not shed light on the functionality of the other parts of PBCAT. The user interface of the program, however, was well liked by all participants and is uniform across all of the modules. It is therefore fair to conclude that the participants would also have appreciated the operational ease of the balance of PBCAT.