The doi above is for this specific version of this dataset, which is currently the latest. Newer versions may be published in the future. For a link that will always point to the latest version, please use
doi: 10.4121/9c4d9c35-3330-4536-ab8d-d5bb237c277d

Albers, Nele; Neerincx, Mark; Brinkman, Willem-Paul (2024): Reinforcement learning for proposing smoking cessation activities that build competencies: Combining two worldviews in a virtual coach - Data, analysis code, and appendix for the PhD thesis chapter. Version 1. 4TU.ResearchData. dataset. https://doi.org/10.4121/9c4d9c35-3330-4536-ab8d-d5bb237c277d.v1

Dataset

• Artificial Intelligence and Image Processing
• Psychology
• Psychology and Cognitive Sciences
• Information and Computing Sciences

Behavior change, Conversational agent, Digital health, eHealth, Mental model, Persuasion, Persuasive technology, Reinforcement learning, Smoking, Virtual coach

CC BY 4.0

RefWorks, BibTeX, Reference Manager, Endnote, DataCite, NLM, DC, CFF

by Nele Albers , Mark Neerincx , Willem-Paul Brinkman

This repository contains the data, analysis code, and appendix for the chapter "Reinforcement learning for proposing smoking cessation activities that build competencies: Combining two worldviews in a virtual coach" from the PhD thesis by Nele Albers.

Study

The chapter is based on data collected in three studies.

Study 1

We conducted this study on the online crowdsourcing platform Prolific between 6 September and 16 November 2022. The Human Research Ethics Committee of Delft University of Technology granted ethical approval for the research (Letter of Approval number: 2338). 

In this study, daily smokers who were contemplating or preparing to quit smoking first filled in a prescreening questionnaire and were then invited to a repertory grid study if they passed the prescreening. In the repertory grid study, participants were asked to divide sets of 3 preparatory activities for quitting smoking into two subgroups. Afterward, they rated all preparatory activities on the labels given to the subgroups.

Participants also rated all preparatory activities on the perceived ease of doing them and the perceived required time to do them. This data can be found in this repository: https://doi.org/10.4121/5198f299-9c7a-40f8-8206-c18df93ee2a0.

The study was pre-registered in the Open Science Framework (OSF): https://osf.io/cax6f.

Study 2

We performed a second repertory grid study with smoking cessation experts. These smoking cessation experts were also asked to divide sets of 3 preparatory activities for quitting smoking into two subgroups based on the question “When it comes to competencies for quitting smoking that smokers build by doing the activities, how are two activities alike in some way but different from the third activity?”

The study was pre-registered in OSF together with the repertory grid study with smokers: https://osf.io/cax6f. The same ethical approval also applies.

Study 3

We conducted a third study on the online crowdsourcing platform Prolific. In this study, daily smokers interacted with the conversational agent Mel in up to five

conversational sessions between 21 July and 27 August 2023. The Human Research Ethics Committee of Delft University of Technology granted ethical approval for the research (Letter of Approval number: 2939) on 31 March 2023.

In each session, participants were assigned a new activity for quitting smoking: one of 44 preparatory activities or one of 9 persuasive activities. 682 people started the first session and 349 people completed session 5.

The study was pre-registered in OSF: https://doi.org/10.17605/OSF.IO/NUY4W.

The implementation of the conversational agent Mel is available online: https://doi.org/10.5281/zenodo.8302492.

Data

We provide data on the 3 studies:

-Data on study 1 (e.g., the subgroup labels and activity ratings provided by smokers). Additional data from study 1 not used in this chapter can be found here: https://doi.org/10.4121/5198f299-9c7a-40f8-8206-c18df93ee2a0.

-Data on study 2 (e.g., the subgroup labels and activity ratings provided by experts, as well as the self-reported expertise of the experts)

-Data on study 3:

1. Data from participants' Prolific profiles (e.g., age, gender)
2. Data from the prescreening questionnaire (e.g., smoking frequency, quitter self-identity)
3. Data from the conversational sessions with Mel (e.g., effort spent on activities)
4. Data from the post-questionnaire (e.g., smoking frequency, quitter self-identity)
5. Data from the follow-up questionnaire (e.g., smoking frequency, quitter self-identity, weekly exercise amount)
6. The variable "rand_id" is a random participant identifier and can be used to link data from different data files.

Analysis code

All our analyses are based on either R or Python. We provide code to allow them to be reproduced.

Appendix

We also provide the chapter's appendix, which includes, for example, the formulations of the 44 preparatory and 9 persuasive activities.

In the case of questions, please contact Nele Albers (n.albers@tudelft.nl) or Willem-Paul Brinkman (w.p.brinkman@tudelft.nl).

• 2024-12-10 first online, published, posted

4TU.ResearchData

.py, .txt, .csv, .xlsx, .md, .pdf, .rmd, .docx

• https://osf.io/cax6f
• https://doi.org/10.5281/zenodo.8302492
• https://doi.org/10.17605/OSF.IO/NUY4W

https://doi.org/10.4121/5198f299-9c7a-40f8-8206-c18df93ee2a0

• This work is part of the multidisciplinary research project Perfect Fit, which is supported by several funders organized by the Netherlands Organization for Scientific Research (NWO), program Commit2Data - Big Data & Health (project number 628.011.211). Besides NWO, the funders include the Netherlands Organisation for Health Research and Development (ZonMw), Hartstichting, the Ministry of Health, Welfare and Sport (VWS), Health Holland, and the Netherlands eScience Center.

TU Delft, Faculty of Electrical Engineering, Mathematics and Computer Science, Department of Intelligent Systems, Interactive Intelligence