SoMeIL: A social media infodemic listening for public health behaviours conceptual framework
| dc.contributor.author | Tsao, Shu-Feng | |
| dc.date.accessioned | 2023-10-10T12:42:49Z | |
| dc.date.available | 2023-10-10T12:42:49Z | |
| dc.date.issued | 2023-10-10 | |
| dc.date.submitted | 2023-09-29 | |
| dc.description.abstract | Introduction The coronavirus disease 2019 (COVID-19) pandemic has escalated health infodemics given substantially digitalized daily life since the pandemic began. The number of social media users has skyrocketed. However, this has brought issues given misleading health information circulating on social media platforms that can lead to undesirable behaviours compromising individual or public health in real life. One long-lasting health issue is vaccine hesitancy, which has been further compounded by health infodemics on social media. According to the World Health Organization, health infodemics occur when too much information that makes true information competes with misinformation for people’s attention, understanding, and adherence to recommended health interventions. Existing theories and theoretical constructs have been applied to study public behaviours influenced by health infodemics on social media. However, these theories have limited to individual behaviours and ignored other critical factors. Furthermore, the current theories have rarely reflected the nature of social media as information can be disseminated instantly and massively without geographical restrictions regardless of information quality. Therefore, this dissertation aimed to address these limitations by proposing a solution that can listen to public discourse on social media and infer their behavioural intentions in real life. Methods The scoping review (Study I) was conducted by following the methods of Arksey and O'Malley as well as Levac et al. to identify and synthesize literature related to the research question. The theory construction methodology was used in the conceptual paper (Study II) to review existing theories and propose a new conceptual framework. Next, the Latent Dirichlet allocation topic modelling and qualitative thematic analysis were applied in the preliminary and partial qualitative validation study (Study III). The last study (Study IV) applied structural equation modeling (SEM) to infer people’s intentions toward COVID-19 vaccination in real life from Twitter amid the pandemic as a preliminary and partial validation for the proposed conceptual framework. Results A total of 2,405 articles published between November 1, 2019, and November 4, 2020, were retrieved from PubMed, Scopus, and PsycINFO. After removing duplicates, non-empirical literature, and irrelevant studies, a total of 81 articles written in English published in peer-reviewed journals were included in the scoping review (Study I). Six themes were found and reported: (1) surveying public attitudes, (2) identifying infodemics, (3) assessing mental health, (4) detecting or predicting COVID-19 cases, (5) analyzing government responses to the pandemic, and (6) evaluating quality of health information in prevention education videos. The findings also suggested knowledge gaps in real-time COVID-19 surveillance using social media data and limited machine learning or artificial intelligence techniques used in overall COVID-19 research using social media data except the first theme. In the conceptual paper (Study II), a new conceptual framework—social media infodemic listening for public health behaviors (SoMeIL) —was proposed to address limitations in existing theories given lacking systematic and theoretical foundation for such research. After the SoMeIL was proposed, validations were needed. A preliminary qualitative validation and demonstration using Twitter data about the Canadian Freedom Convoy were conductedto partially validate and illustrate how the SoMeIL conceptual framework could be applied (Study III). Finally, the findings from SEM in the last study (Study IV) showed statistically significant associations between the latent variable and the observed variables derived from Twitter. This study provided preliminary evidence to validate partial components in the proposed SoMeIL conceptual framework that could be used as a proxy to infer people’s vaccination intentions in real life. It also demonstrated the feasibility of using Twitter data in SEM research besides typical surveys. Conclusion The scoping review (Study I) was important since it identified various roles that social media data have played in research related to the COVID-19 pandemic. It also informed us of knowledge gaps to be bridged. This led us to the conceptual paper (Study II) since we identified limitations in existing theories when the current theories or theoretical constructs were applied in health research that analyzed social media data. A new conceptual framework—SoMeIL—was proposed accordingly. A preliminary qualitative study was followed to validate and demonstrate partial components of the SoMeIL conceptual framework. The last study (Study IV) showed preliminary evidence to show that parts of the SoMeIL conceptual framework was workable given statistically significant relationships found among certain constructs. As a result, Twitter data in this dissertation could be used as a proxy to infer people’s vaccination behavior in real life as suggested by the proposed conceptual framework. Yet more research is needed to further validate and improve the proposed SoMeIL conceptual framework. If social media listening can be integrated into future pandemic preparedness as the proposed conceptual framework suggests, it can help health authorities and governmental agencies promptly shape public perception, disseminate more scientific information, and influence behaviors during a health crisis in a timely fashion. | en |
| dc.identifier.uri | http://hdl.handle.net/10012/20029 | |
| dc.language.iso | en | en |
| dc.pending | false | |
| dc.publisher | University of Waterloo | en |
| dc.relation.uri | https://github.com/shutsao/SEM_Twitter | en |
| dc.subject | conceptual framework | en |
| dc.subject | social media | en |
| dc.subject | machine learning | en |
| dc.subject | social listening | en |
| dc.subject | health infodemics | en |
| dc.subject | en | |
| dc.subject | COVID-19 | en |
| dc.title | SoMeIL: A social media infodemic listening for public health behaviours conceptual framework | en |
| dc.type | Doctoral Thesis | en |
| uws-etd.degree | Doctor of Philosophy | en |
| uws-etd.degree.department | School of Public Health Sciences | en |
| uws-etd.degree.discipline | Public Health and Health Systems | en |
| uws-etd.degree.grantor | University of Waterloo | en |
| uws-etd.embargo.terms | 0 | en |
| uws.contributor.advisor | Chen, Helen | |
| uws.contributor.advisor | Butt, Zahid | |
| uws.contributor.affiliation1 | Faculty of Health | en |
| uws.peerReviewStatus | Unreviewed | en |
| uws.published.city | Waterloo | en |
| uws.published.country | Canada | en |
| uws.published.province | Ontario | en |
| uws.scholarLevel | Graduate | en |
| uws.typeOfResource | Text | en |