CS352

• Presentations of user studies Mon 7/12 (15-
20 min each group)
• Project – Concepts and Prototypes due next Thu 7/15 at 11:59pm (presentation next day in class)
• Reading:
– Heuristic Eval: Read 15.2
– Cognitive Walkthrough (first-time users): Read
15.3
2

• P redispositions: Did this in Project Proposal.
• RI : R esearch was studying users. Hopefully led to I nsights.
• CP : C oncept and initial (very low-fi)
P rototypes due Monday of Week #6.
 E valuate throughout, repeat iteratively!!
3

• Analytical vs. empirical evaluations
• Examples of analytical evaluations?
– GOMS
– KLM

• Analytical – based on your head
• Empirical – based on data
– Formative
• inFORMs design
• what is (still) needed?
– Summative
• did it work?

• You follow established guidelines/procedures/models to decide
(in your head) how good your design is.
• Examples:
– GOMS/KLM – for skilled users.
• evaluating efficiency of regular use.
– Heuristic Evaluation
– Cognitive Walkthrough – for first-time users.
• evaluating ease of learning.

 One of the “Discount Usability” methods
• Apply heuristic guidelines (e.g., Nielsen’s) to a UI to find problems.
• Who does it:
– Multiple usability experts.
• Default: use Nielsen’s:
– http://www.useit.com/papers/heuristic/heuristic_ list.html

http://www.useit.com/papers/guerrilla_hci.html

• One of the “Discount Usability” methods
• Apply heuristic guidelines (e.g., Nielsen’s) to a UI to find problems.
• Who does it:
– Multiple usability experts.
• Default: use Nielsen’s:
– http://www.useit.com/papers/heuristic/heuristi c_list.html

Nielsen’s 10 Usability Heuristics
1. Visibility of system status
2. Match between system and the real world
3. User control and freedom
4. Consistency and standards
5. Error prevention (e.g., default choice being “Cancel”)
6. Recognition rather than recall

Nielsen’s 10 Usability Heuristics
(cont’d)
7. Flexibility and efficiency of use (e.g., short cut keys vs. menus)
8. Aesthetic and minimalist design

Nielsen’s 10 Usability Heuristics
(cont’d)
9. Help users recognize, diagnose, and recover from errors
10. Help and documentation (but keep in mind that users do not like to read manuals)

• Heuristics for ambient displays:
– pp. 699-700.
Nabaztag ambient display http://infosthetics.com/archive s/2005/10/nabaztag_ambient
_email_display.html

• How:
– Pick some usability guidelines, experts, task.
– Brief the experts.
– Experts indep’ly evaluate UI’s conformance with those guidelines for that task (1-2 hrs.)
• Pass 1: flow.
• Pass 2: UI details.
• Outcome: UI problems.
– Experts meet to discuss problems, assign priorities, suggest solutions.

• Heuristics for the web:
1. Avoid orphan pages, i.e., pages not connected to homepage
2. Avoid long pages with excessive white space that force scrolling
3. Provide navigation support, such as a strong site map that is always present
4. Avoid narrow, deep, hierarchical menus that force users to burrow deep into the menu structure
5. Avoid non-standard link colors
6. Provide consistent look and feel for navigation and information design
• Pick any three and do a heuristic evaluation – write down problems you find

• “Walk through” the UI, asking yourself questions along the way.
• Who does it:
– 1 or more usability experts. (3 is ideal)
• How:
– Identify characteristics of user and task.
– Make up multiple copies of the Questions
(next slide...)

• Walk through the task while answering these Questions:
– Will the user know what to do? (Gulf of exec)
– Will the user see how to do it? (Gulf of exec)
– Will the user understand from feedback whether their action was correct? (Gulf of eval)

• Record the answers but also:
– Assumptions about what would cause the problems and why .
– Notes about side issues .
– Notes about possible solutions .
• Example for Amazon.com
– pp. 703-704.

• Go to rei.com
• Task: buy a pair of skis
• For each step, write down:
1. What is the user trying to do?
2. Answer to each of the questions:
– Will the user know what to do? (Gulf of exec)
– Will the user see how to do it? (Gulf of exec)
– Will the user understand from feedback whether their action was correct? (Gulf of eval)

• GOMS/KLM vs. HE vs. CogWalk
– how long it takes you to do
– how much detail it gives you
– about what kind of user