Evaluation of datavis Flashcards
What are the possible purposes of a visualization? (5)
- explore
- explain
- inform
- plan
- monitoring
What makes a good visualization? (6)
- Serve a purpose
- Tailored to an audience
- Tailored to the medium
- Do a task/insight
- Is memorable and makes an impact
- illustrates a point
What makes a bad visualization? (5)
- Difficult to interpret
- Misleading
- Redundant
- Much useless information
- Cluttered
What do you need to know before designing a visualization? (2)
- Caracteristics of the data
2. Goals and tasks
Define usability:
the extent to which a product can be used by specified users to achieve specified goals with
effectiveness, efficiency, and satisfaction in a specified context of
use
Define effectiveness:
to what extent users achieve their objective
Define efficiency:
The amount of effort users put into the task
Define satisfaction:
Subjective opinions, attitudes, preferences
Which are the categories empirical evaluations methods can belong to? (2)
- Quantitative
2. Qualitative
Which metrics do we measure in controlled experiments? (2)
- Accuracy for effectiveness
2. Completion time for efficiency
Which are the issues we may have in quantitative evaluations? (3)
- Challenge to involve users
- Reliability of the measure
- Reproducibility of the measure
Which are the issues we may have in qualitative evaluations? (3)
- plenty of labor required
- unclear sample size
- Investigator’s dependent data quality
Which is the goal of qualitative evaluation?
Gather information about subjective experience of users
Does qualitative evaluation only collect subjective data?
False, it collects both objective and subjective data
List some qualitative approaches: (7)
- Think-Aloud protocol
- Semi-structured interviews
- Video and audio recordings
- Computer logs
- Artifacts (diagrams, drawings)
- Focus groups
- Questionnaires