How do we maintain a track record of success in design?
How can we also retain our creativity?
Subsidiary Questions
What are interactive systems?
The user interface
Technocentric versus anthropocentric approaches
Why do we design them?
To resolve a situation of concern
Success?
How do we know if we’ve succeeded?
By testing whether the situation is resolved
But we can’t do this during design
By measuring or predicting usability
The Need for Usability Analysis
The need arises when we’re faced with the following kinds of questions during design:
Will the operator be able to handle emergency telephone calls faster than before?
Have we simplified the design of this ticket machine to a point where people will use it successfully on their first attempt?
Is the small size of this screen target
going to result in a significant number of
errors in selecting it?
If the user invokes this command by mistake,
will he or she find the escape route?
The Need for Usability Analysis
Will the word-processor user remember that there are three different ways of changing the properties of a formatting style?
Is it so difficult to change the layouts of menus that hardly any users will bother?
Once the system is set up to support work-groups of a particular size and structure, how much effort is involved in changing the system to support changes in the group?
How many of the people who try the system
will actually continue to use it?
To answer these questions, we analyze the
design in terms of its usability.
Usability Factors
The speed of performance of the activity, which affects how many people are needed to perform it
The incidence of errors while performing the activity
The user’s ability to recovery from errors that occur
The magnitude of the user’s task in learning to use the system
The user’s retention of learned skills
The user’s ability to customize the system to suit their way of working or the situation of use
The ease with which people can reorganize
activities supported by the system— their own
activities and other people’s
Users’ satisfaction with the system.
What happens if we fail?
Breakout #1
Divide into your groups
Do the following:
Watch the demonstration video.
List all the usability issues you can identify in the use of this product.
Fundamentals of Interactive System Design
Identifying the human activity that the proposed interactive system will support
Identifying the people, or users, who will perform the activity
Setting the levels of support that the system will provide, otherwise known as the system’s usability
Selecting the basic form of solution to the design problem.
Product Development Phases
Usability touches each phase of Product Development
Using Human Factors Tools & Techniques
Two Ways of Assessing Usability
Analytically - by simulating how the user’s activity will be performed
Empirically - by building and testing a prototype
Analytical Methods Advantages
Analytical methods have advantages:
We can test designs that we can’t build
We can save time by not building a prototype
We don’t need to plan and conduct an experiment
In other words, they can be used more quickly, earlier in the design
Empirical Methods Advantages
Empirical methods have advantages:
We receive more precise information about how a user will interact with the product
We can see major flaws easier and therefore reducing modifications to the product once it is fielded
In other words, it will provide richer information; however, there is are also increased costs associated with the experimentation
Two Type of Information
Qualitative
Quantitative
Tools & Techniques
Interviews
Purpose: Useful for reaching target populations and gathering rich information
Types: structured, unstructured
Making sure the following are covered:
The interview’s purpose, explained at the outset.
Enumerating activities by asking a general question, e.g., “What are your tasks?” and following this up with more specific questions.
Work methods: finding out how tasks are performed.
Performance issues. These provide a measure
of the “usability” of the current support system
and the need for improvement.
Questionnaires
Purpose: Useful for reaching large populations and thus gathering large amounts of data
Issues to consider in design:
The need to make things easy for the subject.
The need for unambiguous questions.
The need to gather precise data.
The need to support the intended analysis.
Focus Groups
Purpose: Useful for reaching target groups of users to get consensus information on product information
Method:Facilitated workshops of groups of 5-10 people - current or likely users
Type of Information:subjective (tasks, requirements, product ideas, etc.)
Function Allocation
Function Allocation Strategy
Mandatory - system requirements, safety, legal or labor constraints, etc.
Balance of value - assignment based on relative performance
Utilitarian - human is available and is capable
Cost-based - relative cost of performance
Affective and cognitive support – meaningful work and maintaining adequate knowledge of the system
Guidelines
Purpose: Guidelines provide us with advice on the solution of design problems.
They suggest possible solution strategies.
Each guideline has a context or domain within which it applies.
Guidelines act as heuristics, drawing on assumptions derived from past experience.
In many cases, the experience we draw on includes empirical research.
Roles of Guidelines
Raising awareness of concepts
Assisting in design choices
Offering strategies for solving design problems
Supporting evaluation
Problems in selecting guidelines:
Problems in selecting guidelines:
the tendency to apply the first guideline that seems relevant
Problems in applying guidelines, e.g., from Tullis (1988): Example: Reduce search times by minimizing the number of groups of items while designing each group to subtend a visual angle as close as possible to 5 degrees
Does this apply to our problem?
Will it have the desired effect?
Will the resulting design really be more
usable?
Check the guideline against the problem statement:
Check the guideline against the problem statement:
Is the guideline appropriate to the activity that the design is to support?
Is it applicable to the type of user who will perform the activity?
Does it address the particular levels of support or usability factors that determine the success of the design?
Is it appropriate to the form of solution chosen?
Guideline Categories
Five contexts that cover the spectrum of guideline use:
General principles that apply to any user interface
Guidelines that apply to forms of solution for interactive display layouts, including those that use color
Guidelines for use with specific interaction styles
Sets of guidelines offered in style guides associated with proprietary systems and standards
Guidelines for the design of individual user interface components supporting particular user tasks.
General Design Principles
Two universal principles:
Design with a view to supporting the user’s task or process.
Know the user (Hansen, 1971 in Newman and Lamming,1995).
Suggested general principles:
Shneiderman (1992) Nielsen and Molich (1989)
Strive for consistency. Be consistent.
Enable frequent users to use Provide short cuts.
shortcuts.
Offer informative feedback. Provide feedback.
Design dialogues to yield closure. Good error messages.
Offer simple error handling. Provide clearly marked exits.
Permit easy reversal of actions. Support internal locus of control.
Reduce short-term memory load. Minimize user memory load.
Simple and natural dialogue.
Speak the user’s language.
Prevent errors.
Examples of Guidelines
Make all facets of design consistent with user expectations considering both the user’s prior experience and well established conventions, such as symbology
Design workstations, controls, and displays around the basic capabilities of users regarding such characteristics as strength, dexterity, memory, reach, visual acuity, and hearing
Be sure that auditory signals are well within users’ threshold values for amplitude and frequency considering the effects of ambient noise
Be sure the brightness of visual signals must be sufficient to be perceived by users working under various conditions of ambient illumination and that the brightness and contrast are adequate to optimize legibility
Be careful that labels and displayed information
are easy to read from the typical viewing angles
and distances. Symbol size, contrast, color, and
display depth must be considered.
Ensure abbreviations, symbols, text, and acronyms placed on, or displayed by, the device are also used in the instructional manual
Ensure abbreviations, symbols, text, and acronyms placed on, or displayed by, the device are also used in the instructional manual
Design control knobs and switches to correspond with both general conventions and any that are unique to the user population
Arrange and design knobs, switches, and data-entry key in a way that reduces the likelihood of accidental activation
Use color and shape coding to facilitate the rapid identification and discrimination of controls and displays. Color and codes should correspond to universal industry convention.
Space keys, valves, and control knobs sufficiently apart for easy manipulation. This will also reduce likelihood of accidental
activation.
Anthropometrics
Performance Task Analysis
Time
Errors
Quality
Quantity
Workload, etc.
Methods of Performance Analysis
Three widely-used methods:
GOMS analysis methods, including keystroke-level analysis
Heuristic evaluation: introducing walkthrough and performance analyses as needed
Cognitive walkthrough, in which performance analysis is folded into the sequence analysis
Two Stages of Analysis
We are making predictions about how a human activity, performed as a sequence of steps, will be supported.
So:
1. We must establish the sequence of steps
2. We must analyze the performance of each step
GOMS Analysis
Analysis in terms of four components of the activity:
Goals that users are trying to achieve
Operators, i.e., basic actions that users perform
Methods employed by users to attain goals, made up of sequences of operators
Selection rules for choosing between methods
Heuristic Evaluation
Can be applied to problems where GOMS and Cognitive Walkthrough are unsuitable, i.e.:
(a) method of operation is not fully predictable, and
(b) user is not a complete novice
The Heuristic Evaluation method:
employ a team of evaluators to identify problems in the design
provide a list of heuristics (general guidelines) to guide their evaluation, e.g.:
Simple and natural dialogue Provide clearly
marked exits
Speak the user’s language Provide short cuts
Minimize user memory load Good error
messages
Be consistent Prevent errors
Provide feedback
Advantages of Heuristic Evaluation
Low cost
Intuitive to perform
Requires little training
No advance planning required
Can be used early in the design process
Provides high-level evaluation, but inherently less repeatable than other analysis methods
Alternate Set of Heuristics
Learning:
Help and Documentation:
design for use without documentation
provide easy-to-use task-oriented documentation
Adopt the User’s Viewpoint:
speak the user’s language (avoid jargon)
make use of existing knowledge
Simple and Natural Dialogue:
avoid extraneous information, steps, actions
information should be in a logical, natural order
Design for Advancement:
provide shortcuts (quick keys, customization)
Alternate Set of Heuristics
Adapting to the User:
Provide Maps and a Trail:
give the user a way to preview where to go and what will happen
give the user a way to review and return to previous contexts
Show the User What is (Not) Possible:
provide affordances to indicate what can be done
Intuitive Mappings:
design good response compatibility between controls and actions
Minimize Memory Load:
remove the need to remember across dialogues
provide multiple views for easy comparisons
Consistency in the System and to Standards:
make sure the same term / action has one meaning
when there is no better way, conform to a standard
Alternate Set of Heuristics
Feedback and Errors:
Feedback:
provide timely feedback about all processes and system status
Prevent Errors:
make it difficult to make errors
Error Messages:
diagnose the problem and suggest a solution
Clearly Marked Exits and Error Recovery:
make sure the user can get out of an undesirable state easily
design assuming that people will make errors and need to recover previous states
Heuristic Evaluation
Find evaluators:
independent usability AND application experts.
Apply heuristics:
apply each heuristic to many parts of the system.
apply heuristics while completing benchmark tasks.
Enter violations into a database or form:
evaluator records problems or observer takes notes on evaluator.
Heuristic Evaluation
Breakout #2
Divide into your groups
Using the provided devices:
PART 1: perform a heuristic evaluation using the calculator to perform the following math function to solve for y.
y = (3)2 * (40)-6 + 1000 – 7*100 + (-100)
PART 2: perform a GOMS analysis of the setting the time on your watch to central time.
Cognitive Walkthrough Analysis
Determining the Sequence
Earlier, we discussed two methods of analysis:
Analytically - by analyzing the options available to the user at each step—a form of Walkthrough
Empirically - by studying how the user performs the activity, and choosing a representative (‘benchmark’) sequence.
Walkthroughs play a double role:
in determining the sequence analytically
in performing step-by-step analysis of performance
Steps in Conducting a Cognitive Walkthrough
Identify what the user is trying to do, and then ask the following questions repeatedly:
Q1: Will the correct action be made sufficiently evident to the user?
Q2: Will the user connect the correct action’s description with what they are trying to do?
Q3: Will the user interpret the system’s response to the chosen action correctly, i.e., will they know if they have made a right or wrong choice?
The result is to expose design flaws that may interfere with exploratory learning.
The method is best applied by small teams walking through the design together.
Observation
Field study versus laboratory study
Video recording
Concurrent verbal protocols
Passive observation
Prototypes
Prototype Stages
Identifying key properties. We focus on properties identified in the problem statement or in early requirements documents.
Developing the prototype. All we need is a prototype that
has the functions to support the tasks of interest
has the performance to allow a realistic test and
has enough robustness to survive each test without serious failure.
Experimental design. We need a small number of users, to whom we set a suitable range of benchmark tasks, chosen to exercise the prototype’s functionality as fully as possible.
Collecting data. Direct observation and recording of video and concurrent protocols are especially effective.
Data analysis. The good and bad features of the design will probably be obvious right away; we may also take simple performance measurements.
Drawing conclusions. The primary outcome of informal testing is a list of design changes.
