What Did You Learn Last Week

1
What Did You Learn Last Week?

• Are there rules people follow when conversing?
  Elaborate.
• What is communication breakdown? Is it normal?
  Describe two types.
• What is the bandwidth assumption? Is it valid?
• Defend or refute People will always prefer to
  use software to help them coordinate their work
• Describe computer software youre familiar with
  that supports peripheral awareness.

2
Lecture 6Establishing RequirementsData
Gathering Techniques(Preece 7 10.1 10.4)
3
Lecture Overview

• Part I Establishing Requirements
• Part II Data Gathering Techniques
• Part III Discuss Next Group Project Deliverable
  (Design Doc)

4
Part I Establishing Requirements
5
User-Centered Design Process
Identify needs/ establish requirements
(Re)Design
Empirically Evaluate
Build an interactive version
Final product
6
Establishing Requirements isnt Easy!
7
Establishing Requirements

• What?
• Understand as much as possible about users, task,
  context
• Produce a stable set of requirements
• How?
• Data gathering and analysis activities
• Express data as requirements
• Iterate
• Why?
• Requirements definition the stage where failure
  occurs most commonly
• Getting requirements right is crucial

8
Why Establish Requirements?

• Users dont necessarily know what their
  requirements are
• Requirements arise from understanding users
  needs
• Requirements can be justified by empirical data

9
Functional Requirements

• What the system should be able to do (traditional
  focus of requirements gathering activities)
• Examples
• "User should be able to create a new calendar
  entry."
• "Users should be able to delete an existing
  calendar entry."
• "Users should be able to edit an existing
  calendar entry."

10
User Requirements

• Characterize the user population that is targeted
  by the system
• Use of "personas" (archetypal users) has gained
  popularity as a means of defining distinct user
  groups who will use a product
• User requirements encompass such things as
• previous experience
• attitude toward computers
• Level of use novice, expert, casual, frequent
• E.g., "Users of the calendar system must be
  familiar with the Windows operating system and
  wall calendars."

11
Usability Requirements

• Establish user performance levels for the system,
  in measurable, observable terms
• Task efficiency
• Error rates
• Learnability
• Memorability
• Safety
• E.g., "Users must be able to add a new calendar
  entry in 20 seconds"

12
User Experience Requirements

• Establish levels for the user's subjective
  experience with the system, in observable,
  measurable terms (Likert scales)
• Ease of Use
• Satisfaction
• Confidence
• Ease of Learning
• E.g., "On a scale of 1 to 10, users should rate
  the system a 9 in terms of ease of use."

13
For Discussion

• With your project group, take a couple of minutes
  to come up with the following requirements for
  your groups target technology
• One functional requirement
• One usability requirement
• One user experience requirement

14
Part IIData Gathering Techniques
15
Questionnaires

• A series of questions designed to elicit
  specific information
• Questions may require different kinds of
  answers
• YES/NO
• Choice of pre-supplied answers (e.g., Likert
  Scale)
• Comment
• Can give quantitative or qualitative data
• Good for answering specific questions from a
  large, dispersed group of people

16
For Reflection Design a Questionnaire

• Suppose that you are a member of a Microsoft
  design team that is charged with designing a new
  calendar application. Create a questionnaire to
  gather early data.
• Whom would you administer it to?
• What kinds of questions would your questionnaire
  have?
• What would you ask?

17
Interviews

• Verbal questionnaire
• Three basic types
• Structured Predetermined questions
• Semi-structured Predetermined questions with
  open-ended follow-up
• Unstructured No predetermined questions
• Props can be used to stimulate responses
• Sample scenarios of use, prototypes
• Can prove helpful to audiotape and transcribe
• Good for getting personal perspectives and
  exploring issues
• Can be time-consuming
• May be difficult to interview all key stakeholders

18
For ReflectionDesign an Interview Protocol

• Suppose you are asked to conduct interviews as
  part of the data gathering process for the new
  calendar system
• Whom would you recruit as informants?
• What type of interviews would you conduct?
• What questions would you ask?

19
Focus Groups

• Group interviews
• Good for consensus-building
• Good for highlighting areas of contention
• Require a skilled facilitator for best results
• See this primer for guidelines on running a focus
  group

20
Naturalistic Observation(Ethnographic Field
Techniques)

• Spend time with stakeholders in their day-to-day
  environments, observing work as it happens
  ("hangin' out with the natives")
• Gain insights into stakeholders real life tasks
  and problems, firmly grounded in context
• Several ethnographic field techniques can help
• Participant observation
• Audio and videotaping
• Artifact collection
• Good for understanding the nature and context of
  the tasks
• But requires potentially significant time
  commitment to conduct study and analyze data

21
Tips for Applying Ethnographic Field Techniques

• Be aware of and open about your biases
• Dont take anything for granted
• Naiveté is an important attribute for an
  ethnographer
• Be curious!
• Render the ordinary extraordinary ask lots of
  questions
• Be reasonable, courteous, and unthreatening
  (Hughes et al., 1993)
• Be systematic
• Ethnography isnt arbitrary!
• It involves using established research techniques
  to systematically explore a set of research
  questions

22
An Offshoot Contextual Design

• http//www.incent.com/cd/cdhow.html
• Method for collecting and interpreting fieldwork
  data, with end-goal of developing software
• Is widely used commercially (e.g., Microsoft)
• Has seven parts
• Contextual inquiry
• Work modeling
• Consolidation
• Work redesign
• User environment design
• Mockup and test with customers
• Putting it into practice
• Well focus on first step, a data gathering
  method that aims to understand users work

23
Step 1 Contextual Inquiry

• An approach to ethnographic study where user is
  expert, designer is apprentice
• A form of interview, but
• at users workplace (workstation)
• only 2 to 3 hours long
• Four main principles
• Context see workplace what happens
• Partnership user and developer collaborate
  theres no dominant partner
• Interpretation observations interpreted by user
  and developer together
• Focus Inquiry must remain relevant to the design
  being developed a project focus is established

24
Contextual Inquiry is Different From Ethnography

• Much shorter (2-3 hours vs. months)
• More intense and focused
• Inquiry, not participant observation
• Clear focus on system design
• A useful technique for early data gathering
• Can you suggest a way to use contextual inquiry
  as a means of gathering data for the calendar
  tool project?

25
Choosing an Appropriate Data Gathering Technique

• Techniques differ in two key ways
• Amount of time, level of detail and risk
  associated with the findings
• Knowledge the analyst requires
• Techniques are complementary
• E.g., questionnaire early, interviews later

26
Problems with Data Gathering

• Identifying and involving broad range of
  stakeholders users, managers, developers,
  customer reps, union reps?
• Getting real users, not managers
• Traditionally a problem in software engineering,
  but better now
• Req'ts management version control, ownership
• Communication between parties
• within development team, between users
• different parts of organization use different
  terminology
• Domain knowledge distributed and implicit
• difficult to dig up and understand
• knowledge articulation how do you walk?
• Availability of key people

27
Some Tips for Data Gathering

• Focus on identifying the stakeholders needs
• Involve all the stakeholder groups
• Involve more than one representative from each
  stakeholder group
• Triangulate using a combination of data gathering
  techniques
• Support the process with props such as prototypes
  and task descriptions
• Run a pilot session
• Know your key research questions
• Consider carefully how to record the data

28
Interpreting and Analyzing Data

• Aims
• Identify requirements
• Firmly ground in empirical data
• Start soon after data gathering phase
• Strive for initial interpretation before deeper
  analysis
• Different approaches emphasize different elements
• class diagrams for object-oriented systems,
• entity-relationship diagrams for data intensive
  systems
• Following table can help structure requirements
  descriptions (will use in your projects)

29
Volere Shell
30
For Discussion

• What kinds of data gathering techniques would be
  appropriate in the following situations, and how
  would you use them? Assume that you are at the
  beginning of the development process and have
  sufficient time and resources to use any of the
  techniques.

31
Situation A

• You are developing a new software system to
  support a small accountants office. There is a
  system running already with which the users are
  reasonably happy, but it is looking dated and
  needs upgrading

32
Situation B

• You are looking to develop an innovative device
  for diabetes patients to help them record and
  monitor their blood sugar levels. There are some
  products already on the market, but they tend to
  be large and unwieldy. Many diabetes patients
  rely on manual recording and monitoring methods
  involving a needle, some chemicals, and a written
  scale.

33
Situation C

• You are developing a website for a young persons
  fasion e-commerce site.

34
Summary

• Getting requirements right is crucial
• There are different kinds of requirements each
  is significant for interaction design
• The most commonly-used techniques for data
  gathering are questionnaires, interviews, focus
  groups, and naturalistic observation

35
Part IIIDiscuss Group Project Design Document