Lecture 2: Discovering what people can't tell you: Contextual Inquiry and Analysis Methodology Brad Myers 05-863 / 08-763 / 46-863: Introduction to Human Computer Interaction for Technology.
Download
Report
Transcript Lecture 2: Discovering what people can't tell you: Contextual Inquiry and Analysis Methodology Brad Myers 05-863 / 08-763 / 46-863: Introduction to Human Computer Interaction for Technology.
Lecture 2:
Discovering what people can't tell you:
Contextual Inquiry and Analysis
Methodology
Brad Myers
05-863 / 08-763 / 46-863: Introduction to
Human Computer Interaction for
Technology Executives
Fall, 2014, Mini 2
© 2014 - Brad Myers
1
Happy Halloween!
Take 2 candies!
© 2014 - Brad Myers
2
Resolve Devices for Assignments
On the GoogleDoc
© 2014 - Brad Myers
3
Some Usability Methods
Contextual Inquiry
Contextual Analysis (Design)
Paper prototypes
Think-aloud protocols
Heuristic Evaluation
Affinity diagrams (WAAD)
Personas
Wizard of Oz
Task analysis
Cognitive Walkthrough
KLM and GOMS (CogTool)
Video prototyping
Body storming
Expert interviews
A vs. B studies
Questionnaires
Surveys
Interaction Relabeling
Log analysis
Focus groups
Card sorting
Diary studies
Improvisation
Use cases
Scenarios
Cognitive Dimensions
“Speed Dating”
…
© 2014 - Brad Myers
4
Contextual Inquiry and Analysis/Design
One method for organizing the development process
We teach it to our MS and BS students
Proven to be very successful
Hartson-Pyla text: Chapters 3-6
(doing things in a different order than text)
Also described in this classic book:
H. Beyer and K. Holtzblatt. 1998. Contextual Design: Defining
Customer-Centered Systems. San Francisco, CA:Morgan
Kaufmann Publishers, Inc. ISBN: 1558604111.
http://www.incent.com/
© 2014 - Brad Myers
5
Contextual Inquiry & Analysis/Design
Contextual Inquiry
A kind of “ethnographic” or “participatory design”
method
Combines aspects of other methods:
Interviewing, think-aloud protocols,
participant/observer in the context of the work
Afterwards: Contextual Analysis (HartsonPyla term)
Beyer-Holtzblatt call it “Contextual Design”
Also includes diagrams (“models”) to describe
results
© 2014 - Brad Myers
6
“Contextual Inquiry”
Interpretive field research method
Depends on conversations with users in the
context of their work
Used to define requirements, plans and
designs.
Discover the real requirements of the work
Drives the creative process:
In original design
In considering new features or functionality
© 2014 - Brad Myers
7
Context
Definition:
“The interrelated conditions within which
something occurs or exists”
Understand work in its natural environment
Go to the user
Observe real work
Use real examples and artifacts
“Artifact”: An object created by human workmanship
Interview while she/he is working
More reliable than asking them
Context exists even when not a “work” activity
Use “work” here just to mean “doing something”
Can be home, entertainment, etc.
© 2014 - Brad Myers
8
Elements of User's Context: Pay
Attention to all of these
User's work space
User's work
User’s workarounds
User's work intentions
User's words (language used)
Tools used
How people work together
Business goals
Organizational and cultural structure
© 2014 - Brad Myers
9
Why Context?
Design complete work process
Integration!
Fits into “fabric” of entire operations
Not just “point solutions” to specific problems
Consistency, effectiveness, efficiency, coherent
Design from data
Not just opinions, negotiation
Not just a list of features
© 2014 - Brad Myers
10
Key distinctions about CIs
Interviews, Surveys, Focus
Groups
Contextual Inquiry
Summary data & abstractions
Ongoing experience &
concrete data
What customers say
What users do
Subjective
Objective
Limited by reliability of human
memory
Spontaneous, as it happens
What customers think they want
What users actually need
© 2014 - Brad Myers
11
Who?
Users
Between 6 – 20
Representative of different roles
Note: may not be people who will be doing the
purchasing of the system
E.g., if for an enterprise; public kiosk
Interviewers: “Cross-functional” team
Designers
UI specialists
Product managers
Marketing
Technical people
© 2014 - Brad Myers
12
Partnership
Definition:
A relationship characterized by close cooperation
Build an equitable relationship with the user
Suspend your assumptions and beliefs
Invite the user into the inquiry process
© 2014 - Brad Myers
13
Why is Partnership Important?
Information is obtained through a dialog
The user is the expert.
Not a conventional interview or consultant relationship
Alternative way to view the relationship:
Master/Apprentice
The user is the “master craftsman” at his/her work
You are the apprentice trying to learn
© 2014 - Brad Myers
14
Establishing Partnership
Share control
Use open-ended questions that invite users to talk:
"What are you doing?"
"Is that what you expect?"
"Why are you doing...?"
Let the user lead the conversation
Listen!
Pay attention to communication that is non-verbal
© 2014 - Brad Myers
15
Some Alternative Contextual
Inquiry Interview Methods
For intermittent tasks
For uninterruptible tasks
Artifact walkthrough
New technology within current work
Post-observation inquiry
For extremely long or multi-person tasks
In-context cued recall
Activity logs
Future Scenario
Prototype or prior version exists
Prototype/Test drive
© 2014 - Brad Myers
16
Interview Recording and Note-Taking
Do record interview
Video recordings
Screen capture software with laptop microphone for user
When to take notes?
Note taking can help you pay closer attention
Notes lead to faster turn-around
Do not let it interfere with interviewing
Usually would use a second person
How to record?
What the user says – in quotes
What the user does – plain text
Your interpretation – in parentheses
© 2014 - Brad Myers
Write fast!
17
Analysis
In the moment:
Simultaneous data collection and analysis during
interview
Post interview:
Using notes, tapes, and transcripts
Analysis by a group:
Integrates multiple perspectives
Creates shared vision
Creates shared focus
Builds teams
Saves time
© 2014 - Brad Myers
18
Defining the Tasks
In a real Contextual Inquiry, user decides the
tasks
But you still must decide the focus
Investigate real-world tasks, needs, context
What tasks you want to observe
That are relevant to your product plan
But for Assignment 1, you will have to invent
some tasks
© 2014 - Brad Myers
19
Test Tasks
Task design is difficult part of usability testing
Representative of “real” tasks
Appropriate difficulty and coverage
Sufficiently realistic and compelling so users are motivated to
finish
Can let users create their own tasks if relevant
Should last about 2 min. for expert, less than 30 min. for novice
Short enough to be finished, but not trivial
Tasks not humorous, frivolous, or offensive
Easy task first, progressively harder
But better if independent
© 2014 - Brad Myers
20
Initial Questions for the Users
Find out the context through initial questions
When would you normally do this kind of task?
Who would be involved in making the decisions?
What would influence any decisions?
How would you know what to do?
What information would you use to help decide?
Getting their feelings about the tasks and the
context
© 2014 - Brad Myers
21
Test Script
Useful to have a script
Should read instructions out loud
Make sure say everything you want
Make sure all users get same instructions
Ask if users have any questions
Make sure instructions provide goals only in a general
way, and doesn’t give away information
Describe the result and not the steps
Avoid product names and technical terms that appear on the
web site
Don’t give away the vocabulary
Example:
“The clock should have the right time”;
not: “Use the hours and minutes buttons to set the time”
22
© 2014 - Brad Myers
Example of CI
Video of sample session with a eCommerce site:
http://www.cs.cmu.edu/~bam/uicourse/EHCIcontexualinquiry.mpg
Issues to observe
Interview of work in progress, in “context”
Actual session of doing a task
Not an interview asking about possible tasks, etc.
Note that focusing on expert behavior & breakdowns
Questions to clarify about routine, motivations
Why do certain actions: need intent for actions
Notice problems (“breakdowns”)
Notice what happens that causes users to do something
(“triggers”)
E.g. appearance of error messages, other feedback, external
events (phone ringing), etc.
© 2014 - Brad Myers
23