Evaluations Flashcards
Qualitative driven evaluation
Observational findings that identify design features as easy or hard to use
Quantitative driven evaluation
One or more metrics that are measured that reflect whether the tasks were easy to perform
Quick and Dirty Evaluation
Comes from a consultant or small number of users - a quick, cheap evaluation that can be done at any stage of design process
Analytical evaluation (2 types)
Inspection - expert assesses the system with structured techniques
Performance modeling - theoretical models predict user performance
Usability testing
Users are observed using a system in a controlled environment and asked to perform a set of prescribed tasks
Interview/questionnaires after assigned to users
Field studies/in the wild
Users use the system in its natural setting, information is gathered through observation and interviews or systems can be instrumented to collect user data
Issue with evaluation - Validity
Surroundings, bias and hawthorne effect can alter the validity of a test
Hawthorne Effect
People modify their behavior when observed
How should tasks be made for usability tests
Minimal bias, realistic, representative tasks, no confusing or guiding wording
How many participants are recommended for a standard qualitative usability test
5
How many participants are recommended for a quantitative usability test
At least 20
How many users are recommended for a usability test that uses eye-tracking
At least 39
Likert scale
A questionnaire question where numbers represent agree/disagree
Think Aloud protocol
Encourage users to constantly verbalize their thoughts as they move throw the design while facilitator stays quiet
