02 - Methods Flashcards
What are the most commonly used methods?
Usability testing, ethnography, questionnaire studies, controlled experiments
Exploratory vs Confirmatory
Exploratory: Finding out the general “lay of the land” about phenomena that have not been examined into detail yet. There is not enough prior knowledge to formulate testable expectations.
Confirmatory: Enough prior knowledge to formulate testable expectations.
Correlation vs Causation
Correlation: clear pattern by which variables change value in relation to each other, but not necessarily caused by each other
Causation: indicates that one event is the result of the occurence of the other event
What are user studies used for? (Usability testing)
Often needed in HRI research at some point to report on peoples attitude and interaction with the robot.
Measure how users respond to variations of the
robot, the interaction itself, or the context of the interaction.
Focus area examples: Ease of use, first impressions, intuitive use, trust developement etc
Qualitative
What are system studies used for?
Evaluation of the technical capabilities of the robot. Users might be involved, but are not always needed.
What are observational studies used for?
Can be exploratory. Optimally, the researchers are not intervening.
Explain Ethnographic studies.
Uses both observation, interviews and focus groups, so beyond just observing, the context and knowledge about the users is also important
Explain the think-aloud technique.
Making the user think aloud while conducting the experiment.
It can be helpful if the experimenter asks questions, but do not disturb or guide too much. Good examples: “what are you thinking/looking at/doing now/trying to do? “ (from “Things a good therapist would say”)
What is programming by demonstration?
The user “teaches” the robot how to move by moving it. The robot records the trajectory/states
