Cmsc 434 Flashcards
(11 cards)
What does the term “direct manipulation” mean in the context of HCI and how does it relate to information visualizations? Give a good example of an interaction that demonstrates this.
Direct Manipulation: An interface that behaves as though the interaction was with a real-world object rather than with an abstract system.
For example, some maps you can click and some maps you can use the arrow. Clicking and dragging on the map is direct but if there are arrows to move it then it is indirect.
Direct is if you highlight a region and use the mouse to move it over.
Another example is on a mobile phone you can pinch out to zoom into an image and pinch in to zoom out.
Are there places where a “direct manipulation” metaphor might not make sense to utilize? If so, give an example of one in a context and why it would not be a good idea.
Yes, if it restricts what a user can do.
For example,
A strict file/folder hierarchy instead of the system allowing links between directories.
Also Yes, if it will it sets unrealistic expectations
For example,
If it’s a novelty like Microsoft Office’s “Clipit” (assistant) in its old versions, which quickly wore off.
Describe a realistic example of one way to use design and/or visualization techniques to make it easy for someone to spot all of the places where there are intersections with curb cutouts for wheelchair access on an aerial map of a city.
Highlight intersections
Circle intersections
Icon at intersections
How would you describe the difference between the Action-Object model and the Object-Action model to a 300-level student and what example would you give to demonstrate each?
Object Action o select object, then action o Closer to real world model o Modeless interaction o Actions within the context of the object (inappropriate actions hidden) o Generic commands
Example: Click then drag to recycle item
Action Object
o select action, then object
o Pick a setting then reuse it easily.
o Higher efficiency
Example: command line, clicking bold button then typing
Give an example of a type of thing that is best tested via quantitative evaluation
Research questions
Testable hypothesis -
Example: “There is no difference in the number of cavities in children and teenagers using Brand A and Brand B toothpaste”
Give an example of one of the bigger challenges during the controlled study experimental design process.
There can be challenges with research when you get results that seem counterintuitive.
-When I have had classes do a team project on a controlled study, the most common outcome reported by teams was that they learned how challenging it can be to design and run a controlled study. ○ By shifting away from those and focusing on things like heuristic evaluations and qualitative user studies, I have aimed to provide experience with techniques that are more common in “day to day” applied human-computer interaction.
Do any of the psychological effects that we discussed apply to the design of controlled studies? Why or why not?
Yes,
Specifically when we talk about Clever Hans and Bias in controlled studies.
- We must design unbiased instructions and experimental protocols that are prepared, reviewed, and then practiced ahead of time.
- Additionally, we must consider approaches such as double-blind experiments where the person running the study doesn’t know what’s been studied either.
Is the statement “Product A is better than Product B” a good experimental hypothesis? Explain your answer.
No because saying “better” is too vague and open to interpretation.
- Picking two existing solutions that exemplify each can introduce many other differences.
- To do such a study well, we should build out two versions of the same exact product other than where the minute details are. More time and work, but more reliable results, We let the user do something that is quantitative.
Explain why a controlled study can be better than an open-ended study.
Because our results are less open to criticism of confounding factors.
What is the difference when comparing within-subjects design and between-subjects design, and how do the number of independent and dependent variables come into play with this?
-For within-subject testing, you have each participant try all treatments/variations.
-For between-subject testing, each participant only tries a single treatment/variation.
o When it comes to the number of independent and dependent variables. The more you have, the trickier within-subjects and the potential for learning effects can get.
Why might the selection of appropriate colors and images as well as attention to information scent be more of an issue in web page design than in desktop software such as a word processor or spreadsheet program?
Because what made the web different from desktop software was that the data was at a remote location (and even some of the processing might be).
