Wieger-Chapter 7

119-142

Software
Requirements
Engineering
Requirement elicitation technique
SE 2102 (3-0)
BS (Software Engineering)

Lecture-8 By: Dr. Arfan Arshad

Workshops
• Workshops encourage stakeholder collaboration in defining
requirements.
• “A structured meeting in which a carefully selected group of
stakeholders and content experts work together to define, create,
refine, and reach closure on deliverables (such as models and
documents) that represent user requirements.”
• Workshops are facilitated sessions with multiple stakeholders
and formal roles, such as a facilitator and a scribe.
• Workshops often include several types of stakeholders, from
users to developers to testers.
• They are used to elicit requirements from multiple stakeholders
concurrently.
• Working in a group is more effective for resolving disagreements
than is talking to people individually.
By: Dr. Arfan Arshad
Some Useful Tips
Following are a few tips for conducting effective elicitation workshops,
many of which also apply to interviews.

 Establish and enforce ground rules

 Fill all of the team roles and required information

 Plan an agenda

 Stay in scope

 Use parking lots to capture items for later consideration

 Timebox discussions

 Keep the team small but include the right stakeholders

 Keep everyone engaged

By: Dr. Arfan Arshad
Continued….
 Requirements elicitation workshops that involve too many
participants can slow to a contentious crawl.

 Too many participants can hold extended discussions of

unnecessary details and couldn’t agree on how each use case
ought to work.

 The team’s progress accelerates nicely when reduced number of

participants who represent the key roles are included.

 The workshop may loose some input by using the smaller

team, but the rate of progress is more than compensated for that
loss.

 The workshop participants should exchange information off-

line with colleagues who don’t attend and then bring the collected
input to the workshops. By: Dr. Arfan Arshad
Focus groups
• A focus group is a representative group of users who
convene in a facilitated elicitation activity to generate input and
ideas on a product’s functional and quality requirements.

• Focus group sessions must be interactive, allowing all users

a chance to voice their thoughts.

• Focus groups are useful for exploring users’ attitudes,

impressions, preferences, and needs.

• They are particularly valuable if you are developing

commercial products and don’t have ready access to end
users within your company.

By: Dr. Arfan Arshad

Some Useful tips - Focus groups
• Select the focus group members carefully.
• Include users who have used previous versions or products similar to the
one you’re implementing.
• Either select a pool of users who are of the same type (and hold multiple
focus groups for the different user classes)
• Or select a pool representing the full spectrum of user classes so
everyone is equally represented.
• Focus groups must be facilitated.
• Keep them on topic, but without influencing the opinions being expressed.
• Record the session so you can go back and listen carefully to comments.
• Do not expect quantitative analysis from focus groups, but rather a lot of
subjective feedback that can be further evaluated and prioritized as
requirements are developed.
• Elicitation sessions with focus groups benefit from many of the same tips
described previously for workshops.
By: Dr. Arfan Arshad
Observations
• Sometimes you can learn a lot by observing exactly how
users perform their tasks.

• Observations are time consuming, so they aren’t suitable for

every user or every task.

• To avoid disrupting the users’ regularly assigned work activities,

limit each observation time to two hours or less.

• Select important or high-risk tasks and multiple user classes

for observations.

• If you use observations in agile projects, have the user

demonstrate only the specific tasks related to the forthcoming
iteration.
By: Dr. Arfan Arshad
Continued….
• Observing a user’s workflow in the task environment allows the BA:

 to Validate information collected from other sources,

 to identify new topics for interviews,

 to see problems with the current system, and

 to identify ways that the new system can better support the
workflow.

• The BA must abstract and generalize beyond the observed

user’s activities to ensure that the requirements captured apply to
the user class as a whole, not just to that individual.

• A skillful BA can also often suggest ideas for improving the

user’s current business processes.
By: Dr. Arfan Arshad
Continued….
• Observations can be silent or interactive.

• Silent observations are appropriate when busy users

cannot be interrupted.

• Interactive observations allow the BA to interrupt the

user mid-task and ask a question.

• This is useful to understand immediately why a user made

a choice or to ask him what he was thinking about when he
took some action.

• Document what you observe for further analysis after the

session.

• You might also consider video recording the session, if

By: Dr. Arfan Arshad
Questionnaires
• Questionnaires are a way to survey large groups of users to
understand their needs.

• They are inexpensive, making them a logical choice for

eliciting information from large user populations, and they can
be administered easily across geographical boundaries.

• The analyzed results of questionnaires can be used as an

input to other elicitation techniques.

• Use a questionnaire to identify users’ biggest pain points

with an existing system, then use the results to discuss
prioritization with decision makers in a workshop.

• You can also use questionnaires to survey commercial

product users for feedback.
By: Dr. Arfan Arshad
Tips to prepare questionnaires
• Provide answer options that cover the full set of possible responses.
• Make answer choices both mutually exclusive (no overlaps in numerical
ranges) and exhaustive (list all possible choices and/or have a write-in spot for
a choice you didn’t think of).
• Don’t phrase a question in a way that implies a “correct” answer.
• If you use scales, use them consistently throughout the questionnaire.
• Use closed questions with two or more specific choices if you want to use the
questionnaire results for statistical analysis.
• Open-ended questions allows users to respond any way they want, so it’s
hard to look for commonalities in the results.
• Consider consulting with an expert in questionnaire design and
administration to ensure that you ask the right questions of the right people.
• Always test a questionnaire before distributing it.
• It’s frustrating to discover too late that a question was phrased ambiguously
or to realize that an important question was omitted.
By: Dr. Arfan Arshad
• Don’t ask too many questions or people won’t respond.
 Thank You
Q&A

By: Dr. Arfan

Arshad