Transcript Research methods for HCI, HCC

Research methods for HCI, HCC
Judy Kay
CHAI: Computer Human Adapted Interaction
School of Information Technologies
Overview
• What is HCI?
• What is HCC?
• What is the hallmark of excellence in HCI/HCC
research?
• Research methods
• Case studies
2
What is HCI? HCC?
definition – research methods
3
HCI is....
1. Creating and studying new hardware and
software architectures for building humancomputer interfaces
2. Creating and studying new interactive
techniques, metaphors and evaluation
3. Studying processes and techniques for
designing human-computer interfaces
4. Studying users and groups of users to
understand their needs
http://tochi.acm.org/charter.shtml
4
HCI, HCC venues
• Journals
– TOCHI, IJHCS, HCI, ...
• Conferences
– CHI, INTERACT
– others, such as Pervasive, Ubicomp
– Regional eg OzCHI, Australian UIC
• Workshops
5
Hardware and software architectures
• User Interface Management Systems - Models
for specifying interactive dialogs. Algorithms for
generating or interpreting user interface control
from such descriptions.
http://tochi.acm.org/charter.shtml
6
Hardware and software architectures
• Windowing Systems - Software architectures for
managing the interactive workspace and the
allocation of interactive resources such as screen
space, interactive devices. This also includes
architectures for integrated usage of multiple
workstations by multiple users.
http://tochi.acm.org/charter.shtml
7
Hardware and software architectures
• Interface Software Tools - for interactively
designing user-based systems: including screen
layout algorithms, design assistance tools and
interface testing models and tools.
8
Hardware and software architectures
• Interactive Devices - New hardware for accepting
user input and displaying information. Examples of
such are 3D input devices, head mounted displays,
eye tracking devices, new keyboard designs etc.
9
Hardware and software architectures
• Interface Aspects of Multimedia - Integration of
sound, video, animation and other interactive media
into the user environment.
10
Interactive Techniques, Metaphors and
Evaluation
• Interactive Techniques - New ways to express
inputs. eg new ways to express searches, new
menu models and techniques which exploit new
input devices.
11
Interactive Techniques, Metaphors and
Evaluation
• Workspace Models - New ways to organize work.
New models for end users to customize their
interfaces. New metaphors which make interactive
processes concrete for their users.
12
Interactive Techniques, Metaphors and
Evaluation
• Data Presentation - Models for mapping data into
pictures. Models for editing data by interactively
manipulating the visual presentation. Algorithms for
visual layout including graph layout algorithms and
picture prettification algorithms.
13
Interactive Techniques, Metaphors and
Evaluation
• Tutorial and Help Systems - Studies of how help
and instruction should be delivered. Techniques for
integrating help and instruction into user interface
support software.
14
Interactive Techniques, Metaphors and
Evaluation
• Experimental and Empirical Studies - This
includes laboratory experiments and field studies.
Also included are case studies evaluating user
interfaces, interaction techniques, tools, and
methods.
• Empirical Studies of Programmers
15
User Interface Design Processes
• Analysis and Evaluation Techniques - Methods
for analyzing and evaluating the effectiveness of
designs and implemented systems. New software
tools for interface evaluations. Predictive models of
user performance with an interface. Tools and
methods for evaluating alternative designs.
16
User Interface Design Processes
• Design Processes - Explorations of the design
process, techniques for capturing designs, and
methodologies for producing good interface
designs. Models for integrating design tools with
implementation and evaluation tools.
17
Users and Groups of Users
• Universal Access - Techniques, devices, models
which facilitate access to and use of computers by
people with special needs.
18
Users and Groups of Users
• Group Work - Explorations of people using
computers to work together, and systems for
enhancing group work. Software and hardware
architectures which support simultaneous group
work.
19
Users and Groups of Users
• Organizational Context - Understanding how user
interface design and implementation fits into the
organizations that use and develop interfaces.
20
Users and Groups of Users
• Application-Specific Designs - Interfaces for
specific application areas in which the domain
places significant constraints on the design or
implementation of interfaces. This would exclude
application areas explicitly covered by other
transactions.
21
Halmarks of excellence
Hallmark of excellence in HCI/HCC
• Addresses important problems for people
–
–
–
–
everyday activities for typical people
special situations, people
safety critical contexts
when using technology
• Solid validation
– building on existing theories
– studies
– comparisons
23
…. appropriate research methods
HCI is....
1. Creating and studying new hardware and
software architectures for building humancomputer interfaces
2. Creating and studying new interactive
techniques, metaphors and evaluation
3. Studying processes and techniques for
designing human-computer interfaces
4. Studying users and groups of users to
understand their needs
http://tochi.acm.org/charter.shtml
25
Research methods - familiar
• Classic computer science
– Design
– Methodologies
– Theories
– evaluations
• Scalability
• Accuracy
• Effectiveness…
26
Research methods – new for HCI
• Theories
– eg Fitt's Law, Norman's interaction model....
• Design principles
• User studies to determine needs
– eg. interviews, grounded analysis
• Evaluations
– with users
– techniques that do not require users
– statistical methods
27
Case studies
Case study 1 – tabletop interaction
Novel Interfaces: tabletop
30
31
Case study: tabletop interaction
•
•
•
•
•
Design of new interaction
New technology
Comparing two classes of interaction
Theoretical
Analytic
32
Case Study 1:
• T. Apted, J. Kay, and A. Quigley. Tabletop sharing of digital
photographs for the elderly. In CHI '06: Conference on
Human Factors in Computing Systems, pages 781-790,
New York, NY, USA, 2006. ACM Press.
• Goals:
– Design of new way to interact
– Challenging new software creation
• Validation:
– Theoretical analysis
– User study, two populations
34
Tabletop interaction – design issues
•
•
•
•
Tables as social places
Interaction without keyboard or mouse
Usable by all
Interaction design to support co-located
collaboration and social interaction
35
Evaluation of design: analytic
• Identify design drivers
• Identify suitable guidelines for design
• Analyse design according to these
• Value for future researchers, designers
• Demonstrates validity in terms of established
guidelines which, in turn, are based on research
36
Evaluation of design: user studies
• Design of user study
• Hypotheses and goals
• Participants
• Tasks
• Ethics clearance * aside for Hons
• Trial the study
• Actual study
• Analysis of results
– Qualitative v quantitative
37
Case study 2 – tabletop file system
Case study – file system interaction
Design issues
• tabletop file system interaction
• beyond hierarchies
• collaboration across different people's file systems
• very limited interaction
50
Evaluation: comparative user study
• Comparing two ways to interact
• Double cross over user study
• New file-system v hierarchical
52
Case study 3 – new interface to support
long term collaboration on trac
Case study – Narcissus, WikiNavMap, EDM
Design issues
• help people collaborate more effectively
in a group by gaining a better sense of
the activity on a wiki
• long term collaboration is complex
Kay, J, K Yacef and P Reimann, Visualisations for team learning: small teams
working on long-term projects. In C. Chinn, G. Erkens & S. Puntambekar (Eds.),
Minds, mind, and society. Proceedings of the 6th International Conference on
Computer-supported Collaborative Learning (CSCL 2007) 351-353, New
Brunswick, NJ: International Society of the Learning Sciences
.
54
Narcissus
55
Design of Narcissus
• user study
• select suitable users
• design questionnaire
• administer it
• analyse results
• common alternatives: focus groups, ethnographic
studies, draw on literature
56
Evakuation of Narcissus
• user study
• select suitable users
• design user tasks and questionnaire
• administer
• analyse results
• common alternatives: focus groups, ethnographic
studies, draw on literature
57
Case study 4 – educational data mining –
modelling users from trace data
EDM
• data mining to learn about users
Perera, D, J Kay, K Yacef and I Koprinska, (2007) Mining learners' traces from an
online collaboration tool, Educational Data Mining Workshop at Artificial
Intelligence and Education, 60 – 69.
http://aied.inf.ed.ac.uk/AIED2007/AIED-EDM_proceeding_full2.pdf
59
Case study 5 – pervasive interaction
Case study: Memento
• Design issues:
• future interfaces, current technology limitations
• Evaluation issues:
• proof of concept, usability evaluation
West, D, A Quigley and J Kay, (2006) MEMENTO: A
Digital Physical Scrapbook for Memory Sharing,
Personal and Ubiquitous Computing, 11(4), 313328
71
Summary
• HCI/HCC is broad and multi-disciplinary
• Research methods
– For driving design and analysis (models)
– for creation of new things (architectures)
– for assessing whether you have succeeded
(evaluation)
72