Skip to Content

Changes for page Test

Last modified by Mathieu Jung-Muller on 2022/04/04 13:52
From version Icon 38.1 Icon
edited by Pietro Piccini
on 2022/03/20 16:46
Change comment: There is no comment for this version
To version Icon 37.2 Icon
edited by Mathieu Jung-Muller
on 2022/03/19 13:40
Change comment: typo HPC/HCP fixed

Summary

Details

Icon Page properties
Author
... ... @@ -1,1 +1,1 @@
1 -XWiki.PietroPiccini
1 +XWiki.Mathieu
Content
... ... @@ -94,14 +94,6 @@
94 94  
95 95  In our situation, we recruit 20 students in our class to simulate the research. Since they are not real PwD, we ask them to act as if they were in home settings and we observe their behavior and expression. These data will also be reviewed after the evaluation to obtain the data and feedback.
96 96  
97 -In order to collect qualitative data, we have prepared two questionnaires to give to the participants: the affect assessment questionnaire and the system assessment questionnaire.
98 -
99 -* affect assessment
100 -the affect assessment questionnaire has the purpose of tracking changes in the participant's affect state before and after interacting with Pepper. The questionnaire consists of a list of 10 statements each representing one emotion and participants will be asked to give a score of 1 through 5 to each statement depending on the intensity of the emotion as experienced during the questionnaire. The affect questionnaire will be given twice, once before the interaction with Pepper and once after.
101 -
102 -* system assessment
103 -for assessing the quality of Pepper's interaction we used the Subjective Assessment of Speech System Interfaces (SASSI) questionnaire. The SASSI questionnaire is wildly used in research to evaluate speech and conversational systems. The questionnaire contains 36 questions divided into six main factors: System Response Accuracy, Likeability, Cognitive Demand, Annoyance, Habitability, and Speed. The questions are short and unambiguous and the length of the questionnaires is not too long which could cause the participants to give fast and unreliable answers but it is also long enough so that a good amount of information is collected.
104 -
105 105  == Participants ==
106 106  
107 107  For this study, we simulate the real research by including 20 students in our class. To avoid too much gender imbalance, there will be at least five men and five women. All participants will be asked to pretend to be a demented person and to make the simulation as real as possible.