Eases and Difficulties of Talking to a Virtual Coach for Quitting Smoking and Becoming More Physically Active: A Mixed-Methods Analysis

Eases and Difficulties of Talking to a Virtual Coach for Quitting Smoking and Becoming More Physically Active: A Mixed-Methods Analysis

Ekinci, Arsen (TU Delft Electrical Engineering, Mathematics and Computer Science)

Albers, N. (mentor)
Brinkman, W.P. (mentor)
Kellnhofer, P. (graduation committee)

Delft University of Technology

Computer Science and Engineering

CSE3000 Research Project

2022-06-23

As the awareness of the risks of smoking tobacco increased, usage declined as people started to attempt to quit smoking more. Currently, a wide variety of mobile health (mHealth) applications focused on smoking cessation exist, however users commonly quit these applications over time. In the case of Sam, a conversational agent that plays the role of a virtual coach that helps people to quit smoking and become more physically active, it is still unclear what eases and difficulties users experience when talking to Sam and what others factors might play a role in this. Using a mixed-methods analysis of data, gathered from participants after using Sam, six themes were identified that could provide an insight into these eases and difficulties. The identified themes were conversations feel unnatural, conversations feel natural, clarity towards the user, comfortable to talk, ease of replying and conversations felt impersonal. Recommendations were given for each team on both what to possibly improve and what to leave unchanged. Though users were positive about talking to Sam in general, there still was room for improvement.

mHealth
Conversational Agents
Perfect Fit
AI Acceptance
Thematic Analysis

http://resolver.tudelft.nl/uuid:74b565fb-899f-44ef-ab73-4de286093df6

Student theses

bachelor thesis

© 2022 Arsen Ekinci