Page 40 -

REQUIREMENTS ELICITATION TECHNIQUES AS COMMUNICATION CHANNELS 25
tion. The range of responses that can be provided by the user is constrained so as to provide
auditable “proof” of the dialogue. While they are less expensive and easier for distribution, they
are limited in the richness of the responses that could be received from the user, and provide
little control over the potential (mis)interpretation of questions. The primary assumptions of
these types of elicitation techniques are that the analyst knows the right questions to ask, that
a relatively finite set of questions can develop an understanding of the business scenario, and
that the user is able to understand and appropriately respond to the questions. These elicitation
techniques exploit the analysts’ high level of experience with an organizational context or ap-
plication domain, and focus on the provision of effective methods to verify these experiences.
Techniques such as questionnaires and structured interviews exemplify this class of elicitation
techniques.
Collaboration Focus

Other elicitation techniques are less structured, controlled, and driven by the analyst, and place
greater emphasis on collaborative—rather than confirmatory—communication between user and
analyst. We categorize these techniques as collaborative since they tend to be less formal and
allow users some latitude to negotiate the focus or agenda of the analysis within the structural
constraints of the technique. Unlike verification techniques, these techniques allow the user to
provide requirements in various formats and even allow the user some degree of control over the
elicitation process. As a result, the format of the requirements elicited often requires additional
work on the part of the analyst to condense and understand. They also tend to require a more
iterative process between user and analyst to make sure that understanding of the requirements
has occurred. These techniques generally assume some level of knowledge by the analyst of the
organizational context and/or application domain, but more information is necessary to enhance
understanding. Therefore, collaborative techniques allow the analyst to retain some control over
the elicitation process to develop understanding in those areas that are less understood, while al-
lowing for some verification where prior knowledge does exist. Techniques such as semistructured
interviews, scenario analysis, and use cases are exemplars.

Generation Focus

Our final category encompasses those techniques that are much more free-form in nature and al-
low for maximum variation in user responses with potentially significant loss of control over the
interaction from the perspective of the analyst. These types of techniques we identify as exploratory,
as they tend to be free of bias and overall control from the part of the analyst, and there tends to
be less structure involved in the interaction. The focus is on learning as much as possible about
potential requirements, and the direction of the interaction is not driven by the analyst. Given the
lack of experience with the organizational context and application domain, these techniques are
designed to provide a rich set of information to allow the analyst to more effectively develop these
knowledge bases and understand the requirements specific to the business scenario of interest.
Techniques such as observation, RepGrid analysis, and contextual inquiry exemplify these types
of techniques.
The knowledge bases and communication foci discussed above raise questions about the ef-
ficacy of requirements elicitation techniques as communication channels in the wide range of
organizational contexts in which systems analysis and design is undertaken.