1/16
Learn about using the nested level model for validation.
Name | Mastery | Learn | Test | Matching | Spaced |
---|
No study sessions yet.
how do you know your visualisation works 1
does it amplify cognition?
how do you know your visualisation works 2
does it create insight?
how do you know your visualisation works 3
how can we measure amplified cognition or insight?
how can we argue that one design is better than another 1
what does better mean? faster? more effective? more engaging?
how can we argue that one design is better than another 2
what does effective mean? how do we measure engagement? what sort of task does one tool do better than the other?
domain situation threat to validity
you misunderstood their needs
data/task abstraction threat to validity
you’re showing them the wrong thing
visual encoding/ interaction idiom threat to validity
the way you show the data does not work
algorithm threat to validity
your algorithm is too slow
immediate validation
done while or just after work on a certain level has finished
downstream validation
can be done after work on subsequent levels is finished
immediate validation approach for domain situation 1
observe or interview the target audience and verify characterisation of practices
immediate validation approach for domain situation 2
field study in the real world setting, contextual inquiry
downstream validation approach for domain situation
measure adoption rates and observe the target audience
downstream validation approach for domain situation question
what do users do with the visualisation on their own account, without being prompted?
immediate validation approach for visual encoding/ interaction approach
make sure no known guidelines are being violated, justify your design with respect to known principles
downstream validation approach for visual encoding/ interaction approach
controlled experiment with qualitative discussion and comparison of resulting visualisations