Abstract:
A part of many digital system is the display of error messages. The research aims to create
a set of guidelines for error message design in digital systems. These guidelines will enable
designers and developers to create better error messages that convey the right information
at the right time and in the right way. In other words, to create error messages that are
necessary and effective.
The first step in the process of generating this set of guidelines was to perform a literature
review in order to find existing theory that is applicable to the design of error messages. The
literature review also includes research on warning design theory. The assumption is that
warnings are, to some extent, similar to error messages. Therefore, the research surrounding
it is also expected to be applicable to error messages. The use of warning design theory
stems from the lack of research on error messages and the comparable richness of the body
of knowledge on warnings.
From this literature review it was possible to propose a set of guidelines for error message
design. The initially proposed guidelines were evaluated by performing two usability studies
on an existing Internet banking website. The first usability study involved a heuristic evaluation
of some of the error messages in the website, using the guidelines as heuristics. The
second usability study entailed conducting individual interviews with representative users
where the same error messages used in the heuristic evaluation was evaluated.
The results of the heuristic evaluation were then used to determine whether the guidelines
are effective. The effectiveness of a guideline is an indication of whether experts can easily
use it to analyse error messages and detect possible usability problems. The results of
the individual interviews were used to determine whether the proposed guidelines are valid.
The validity of the guidelines is a measure of how well the guidelines and the suggestions
raised by using them, reflect the pain points and concerns of users. The results of both
these usability studies were also compared to one another to get a further indication of the
effectiveness and validity of the guidelines.
From this analysis some changes and additions were made to the initially proposed guidelines.
These updates are expected to increase the effectiveness and validity of the guidelines
compared to the initially proposed guidelines. In other words, the updates are expected to make the guidelines easier to use and to enable experts to find usability problems that are a
closer match to the concerns of users.
The research followed the design science research methodology, but only completing one
iteration of the process. Subsequent iterations that will further refine the proposed guidelines
is left for future research.