Investigating the influence of a matching personality between user and chatbot on the user satisfaction of chatbot interactions
Keywords
Loading...
Authors
Issue Date
2022-06-19
Language
en
Document type
Journal Title
Journal ISSN
Volume Title
Publisher
Title
ISSN
Volume
Issue
Startpage
Endpage
DOI
Abstract
In the last couple of years, there has been an increase in the usage of chatbots
because of the low costs and convenience. Despite its increasing popularity,
there sometimes is a gap in chatbots’ understanding of users. Due to a lack
of understanding of user intentions and queries, vague answers are given
by chatbots, which in turn results in frustrated users. In these situations,
users would rather interact with humans. Previous research suggested that
personalities could help overcome this user experience with chatbots. This
study investigates how a match in personality between users and chatbots
influences user satisfaction. The focus lies on the Extraversion personality
dimension. A 2×2 between-subject experiment is performed to test the hypothesis
that a match in personality between a user and the chatbot would
increase user satisfaction. Two chatbots are designed which differ in personality
through linguistics, one with an introvert personality and one with an
extravert personality. The personality of users is determined using the Big
Five Inventory. The users interact with one chatbot at a time, and the User
Satisfaction for Information Chatbots (USIC) scale is used to assess their
satisfaction with these interactions. Based on these results, conclusions are
drawn about the influence of personality match between user and chatbot
on the user satisfaction of the interaction with the chatbots. In addition,
it is investigated how a match in personality influences the user satisfaction
of the four factors perceived speed, perceived privacy, communication
quality, and response quality of the interaction with the chatbots. These factors
are defined in the USIC scale. The responses were analyzed using the
Wilcoxon signed-rank test. One should also implement extravert chatbots
to increase the satisfaction of the perceived privacy and the communication
and response quality of the interaction. To increase the satisfaction of
the perceived speed of the interaction, an introvert chatbot would be more
satisfying. For introvert users, a significant difference was obtained for the
communication quality, which resulted in the suggestion to implement extravert
chatbots. However, no valid conclusion could be drawn about the
satisfaction of introvert users because there was no significant result, and
there were only 6 introvert users. According to this study, one should implement
extravert chatbots to increase user satisfaction of chatbot interactions
when an extravert user is involved, so a matching personality.
Description
Citation
Supervisor
Faculty
Faculteit der Sociale Wetenschappen