A pinboard by
Yann Gor'dan

I'm me, an individual, and one of many. Fascinated by the science behind the rise of social media.


A taste of the science behind social media. Research into the social, data and health implications.


Analysing social media data for disaster preparedness: understanding the opportunities and barriers faced by humanitarian actors

Abstract: The use of social media applications by citizens, public authorities, and humanitarian organisations generates vast quantities of data. Research predominantly focuses on the use of social media and associated analysis tools during the short-term response phase of a disaster. As such, the use of social media analysis tools to harness social media data for preparedness purposes is currently unclear. This research uses a combination of semi-structured interviews with 20 Red Cross Red Crescent and humanitarian actors, an online survey, two workshops and desk-based research to examine the opportunities and barriers faced by humanitarian actors in using social media analysis tools to analyse social media data for disaster preparedness. Whilst social media analysis tools provide humanitarian actors with an opportunity to understand the effectiveness of their preparedness communication on social media, to monitor risks and disasters, and to build community preparedness networks, this study identified a limited use of social media analysis tools by research participants. This (non) use of social media analysis tools was influenced by the interaction between seven categories of barriers relating to the user of the tool or the tool itself: language, culture, value, financial, human resources, technology, and data. In discussing these barriers, the authors highlight the key role that context plays in determining the significance of each barrier on the selection and use of social media analysis tools for preparedness.

Pub.: 24 Nov '16, Pinned: 09 Feb '17

Toward Automating HIV Identification: Machine Learning for Rapid Identification of HIV-Related Social Media Data.

Abstract: "Social big data" from technologies such as social media, wearable devices, and online searches continue to grow and can be used as tools for HIV research. Although researchers can uncover patterns and insights associated with HIV trends and transmission, the review process is time consuming and resource intensive. Machine learning methods derived from computer science might be used to assist HIV domain experts by learning how to rapidly and accurately identify patterns associated with HIV from a large set of social data.Using an existing social media data set that was associated with HIV and coded by an HIV domain expert, we tested whether 4 commonly used machine learning methods could learn the patterns associated with HIV risk behavior. We used the 10-fold cross-validation method to examine the speed and accuracy of these models in applying that knowledge to detect HIV content in social media data.Logistic regression and random forest resulted in the highest accuracy in detecting HIV-related social data (85.3%), whereas the Ridge Regression Classifier resulted in the lowest accuracy. Logistic regression yielded the fastest processing time (16.98 seconds).Machine learning can enable social big data to become a new and important tool in HIV research, helping to create a new field of "digital HIV epidemiology." If a domain expert can identify patterns in social data associated with HIV risk or HIV transmission, machine learning models could quickly and accurately learn those associations and identify potential HIV patterns in large social data sets.

Pub.: 13 Jan '17, Pinned: 09 Feb '17

Utility of social media and crowd-sourced data for pharmacovigilance: a scoping review protocol.

Abstract: Adverse events associated with medications are under-reported in postmarketing surveillance systems. A systematic review of published data from 37 studies worldwide (including Canada) found the median under-reporting rate of adverse events to be 94% in spontaneous reporting systems. This scoping review aims to assess the utility of social media and crowd-sourced data to detect and monitor adverse events related to health products including pharmaceuticals, medical devices, biologics and natural health products.Our review conduct will follow the Joanna Briggs Institute scoping review methods manual. Literature searches were conducted in MEDLINE, EMBASE and the Cochrane Library from inception to 13 May 2016. Additional sources included searches of study registries, conference abstracts, dissertations, as well as websites of international regulatory authorities (eg, Food and Drug Administration (FDA), the WHO, European Medicines Agency). Search results will be supplemented by scanning the references of relevant reviews. We will include all publication types including published articles, editorials, websites and book sections that describe use of social media and crowd-sourced data for surveillance of adverse events associated with health products. Two reviewers will perform study selection and data abstraction independently, and discrepancies will be resolved through discussion. Data analysis will involve quantitative (eg, frequencies) and qualitative (eg, content analysis) methods.The summary of results will be sent to Health Canada, who commissioned the review, and other relevant policymakers involved with the Drug Safety and Effectiveness Network. We will compile and circulate a 1-page policy brief and host a 1-day stakeholder meeting to discuss the implications, key messages and finalise the knowledge translation strategy. Findings from this review will ultimately inform the design and development of a data analytics platform for social media and crowd-sourced data for pharmacovigilance in Canada and internationally.Our protocol was registered prospectively with the Open Science Framework (https://osf.io/kv9hu/).

Pub.: 21 Jan '17, Pinned: 09 Feb '17

Unveiling self-harm behaviour: what can social media site twitter tell us about self-harm? A qualitative exploration.

Abstract: To report the findings from a unique analysis of naturally occurring data regarding self-harm behaviour generated through the global social media site, Twitter.Self-harm behaviours are of global concern for health and social care practice. However, little is known about the experiences of those who harm and the attitudes of the general public towards such behaviours. A deeper, richer and more organic understanding of this is vital to informing global approaches to supporting individuals through treatment and recovery.Exploratory, qualitative design.Three hundred and sixty two Twitter messages were subject to inductive thematic analysis.Five themes were identified: 1) celebrity influence, 2) self-harm is not a joke (with sub-themes of you wouldn't laugh if you loved me and you think it's funny, I think it's cruel), 3) support for and from others, 4) eating disorders and self-harm and 5) videos and personal stories.The findings indicate that self-harm behaviour continues to be largely misunderstood by the general public and is often the source of ridicule which may contribute to delays in accessing treatment. Whilst Twitter may also provide a source of valuable support for those who self-harm, the sense of community, relatedness and understanding generated by such support may contribute to normalising self-harm and perpetuating the behaviours.Our understanding of the complexity of, aetiology and most effective treatment options for self-harm behaviours are still unclear. The findings demonstrate that there is a critical opportunity to conduct further qualitative research to better understand self-harm and to utilise this valuable and internationally relevant data to support the development of effective public education campaigns and personally tailored treatment options. This article is protected by copyright. All rights reserved.

Pub.: 08 Sep '16, Pinned: 11 Nov '16

Facebook dethroned: Revealing the more likely social media destinations for college students' depictions of underage drinking.

Abstract: Studies examining representations of college drinking on social media have almost exclusively focused on Facebook. However, recent research suggests college students may be more influenced by peers' alcohol-related posts on Instagram and Snapchat, two image-based platforms popular among this demographic. One potential explanation for this differential influence is that qualitative distinctions in the types of alcohol-related content posted by students on these three platforms may exist. Informed by undergraduate focus groups, this study examined the hypothesis that, of the three platforms, students tend to use Instagram most often for photos glamourizing drinking and Snapchat for incriminating photos of alcohol misuse and negative consequences. Undergraduate research assistants aided investigators in developing hypothetical vignettes and photographic examples of posts both glamorizing and depicting negative consequences associated with college drinking. In an online survey, vignette and photo stimuli were followed by counterbalanced paired comparisons that presented each possible pair of social media platforms. Undergraduates (N=196) selected the platform from each pair on which they would be more likely to see each post. Generalized Bradley-Terry models examined the probabilities of platform selections. As predicted, Instagram was seen as the most probable destination (and Facebook least probable) for photos depicting alcohol use as attractive and glamorous. Conversely, Snapchat was selected as the most probable destination (and Facebook least probable) for items depicting negative consequences associated with heavy drinking. Results suggest researchers aiming to mitigate the potential influences associated with college students' glamorous and consequential alcohol-related photos posted social media posts should shift their focus from Facebook to Instagram and Snapchat.

Pub.: 25 Oct '16, Pinned: 11 Nov '16

Negative Campaigning in the Social Media Age: Attack Advertising on Facebook

Abstract: Abstract Recent studies examine politicians’ decisions to use social media, as well as the content of the messages that these political actors disseminate on social media platforms. We contribute to this literature by examining how race competitiveness and a candidate’s position in the race relative to her opponent affect their decisions to issue attacks. Through content analysis of nearly 15,000 Facebook posts for tone (positive or negative), we find that while competitive races encourage both candidates to issue more negative posts, candidates in less competitive races embrace attack messages with more or less frequency depending on whether they trail or lead their opponent. We find that social media negativity is much more likely to be a desperation strategy employed by underdog candidates in less competitive races. We also run separate models examining the factors that drive policy and personal attacks. While underdog candidates are more likely to engage in issue attacks, candidates in competitive races are significantly more likely to use Facebook to make personal attacks.AbstractRecent studies examine politicians’ decisions to use social media, as well as the content of the messages that these political actors disseminate on social media platforms. We contribute to this literature by examining how race competitiveness and a candidate’s position in the race relative to her opponent affect their decisions to issue attacks. Through content analysis of nearly 15,000 Facebook posts for tone (positive or negative), we find that while competitive races encourage both candidates to issue more negative posts, candidates in less competitive races embrace attack messages with more or less frequency depending on whether they trail or lead their opponent. We find that social media negativity is much more likely to be a desperation strategy employed by underdog candidates in less competitive races. We also run separate models examining the factors that drive policy and personal attacks. While underdog candidates are more likely to engage in issue attacks, candidates in competitive races are significantly more likely to use Facebook to make personal attacks.

Pub.: 01 Dec '16, Pinned: 11 Nov '16

Health information sharing on Facebook: An exploratory study on diabetes mellitus.

Abstract: Increasingly, people are using Facebook (FB) to share health information. However, little is known about the type of information sharing and its potential health consequences in the Arabic speaking world. This study attempts to fill this knowledge gap for diabetes mellitus (DM).We conducted a retrospective qualitative FB content analysis using predefined eligibility criteria. The analysis was restricted to diabetes related groups in the Arabic speaking world. The data were collected between June 2010 and December 2015. A total of 55 groups were screened of which seven met the eligibility criteria.We found 6107 posts in Arabic related to DM of which 1551 posts were included for further analysis. There were 458 (30%) FB posts from Egypt with no posts from Somalia, Yemen, Comoros, and Djibouti. The majority of the posts, 863 (56%), were from females. The focus of the posts was on sharing personal experiences (n=423, 27%), raising awareness (n=210, 3.5%), providing spiritual support (n=162, 10.4%), sharing latest research (n=147, 9.5%), and providing education (n=110, 7.1%) on DM. A large number of the posts by people in 40-60 year age group were around finding out diagnosis related information due to limited access to care in their home countries.Patients with DM are increasingly sharing their health information with other FB users. This study will help inform future research with regard to health information sharing and designing appropriate interventions to harness the power of social media in improving public health.

Pub.: 13 Sep '16, Pinned: 21 Sep '16

Use of Web 2.0 Social Media Platforms to Promote Community-Engaged Research Dialogs: A Preliminary Program Evaluation.

Abstract: Community-engaged research is defined by the Institute of Medicine as the process of working collaboratively with groups of people affiliated by geographic proximity, special interests, or similar situations with respect to issues affecting their well-being. Traditional face-to-face community-engaged research is limited by geographic location, limited in resources, and/or uses one-way communications. Web 2.0 technologies including social media are novel communication channels for community-engaged research because these tools can reach a broader audience while promoting bidirectional dialogs.This paper reports on a preliminary program evaluation of the use of social media platforms for promoting engagement of researchers and community representatives in dialogs about community-engaged research.For this pilot program evaluation, the Clinical and Translational Science Office for Community Engagement in Research partnered with the Social Media Network at our institution to create a WordPress blog and Twitter account. Both social media platforms were facilitated by a social media manager. We used descriptive analytics for measuring engagement with WordPress and Twitter over an 18-month implementation period during 2014-2016. For the blog, we examined type of user (researcher, community representative, other) and used content analysis to generate the major themes from blog postings. For use of Twitter, we examined selected demographics and impressions among followers.There were 76 blog postings observed from researchers (48/76, 64%), community representatives (23/76, 32%) and funders (5/76, 8%). The predominant themes of the blog content were research awareness and dissemination of community-engaged research (35/76, 46%) and best practices (23/76, 30%). For Twitter, we obtained 411 followers at the end of the 18-month evaluation period, with an increase of 42% (from 280 to 411) over the final 6 months. Followers reported varied geographic location (321/411, 78%, resided in the United States); 99% (407/411) spoke English; and about half (218/411, 53%) were female. Followers produced 132,000 Twitter impressions.Researchers and community stakeholders use social medial platforms for dialogs related to community-engaged research. This preliminary work is novel because we used Web 2.0 social media platforms to engage these stakeholders whereas prior work used face-to-face formats. Future research is needed to explore additional social media platforms; expanded reach to other diverse stakeholders including patients, providers, and payers; and additional outcomes related to engagement.

Pub.: 11 Sep '16, Pinned: 21 Sep '16

Please Like Me: Facebook and Public Health Communication.

Abstract: Facebook, the most widely used social media platform, has been adopted by public health organisations for health promotion and behaviour change campaigns and activities. However, limited information is available on the most effective and efficient use of Facebook for this purpose. This study sought to identify the features of Facebook posts that are associated with higher user engagement on Australian public health organisations' Facebook pages. We selected 20 eligible pages through a systematic search and coded 360-days of posts for each page. Posts were coded by: post type (e.g., photo, text only etc.), communication technique employed (e.g. testimonial, informative etc.) and use of marketing elements (e.g., branding, use of mascots). A series of negative binomial regressions were used to assess associations between post characteristics and user engagement as measured by the number of likes, shares and comments. Our results showed that video posts attracted the greatest amount of user engagement, although an analysis of a subset of the data suggested this may be a reflection of the Facebook algorithm, which governs what is and is not shown in user newsfeeds and appear to preference videos over other post types. Posts that featured a positive emotional appeal or provided factual information attracted higher levels of user engagement, while conventional marketing elements, such as sponsorships and the use of persons of authority, generally discouraged user engagement, with the exception of posts that included a celebrity or sportsperson. Our results give insight into post content that maximises user engagement and begins to fill the knowledge gap on effective use of Facebook by public health organisations.

Pub.: 16 Sep '16, Pinned: 21 Sep '16

Digital technology for health sector governance in low and middle income countries: a scoping review.

Abstract: Poor governance impedes the provision of equitable and cost-effective health care in many low- and middle-income countries (LMICs). Although systemic problems such as corruption and inefficiency have been characterized as intractable, "good governance" interventions that promote transparency, accountability and public participation have yielded encouraging results. Mobile phones and other Information and Communication Technologies (ICTs) are beginning to play a role in these interventions, but little is known about their use and effects in the context of LMIC health care.Multi-stage scoping review: Research questions and scope were refined through a landscape scan of relevant implementation activities and by analyzing related concepts in the literature. Relevant studies were identified through iterative Internet searches (Google, Google Scholar), a systematic search of academic databases (PubMed, Web of Science), social media crowdsourcing (targeted LinkedIn and Twitter appeals) and reading reference lists and websites of relevant organizations. Parallel expert interviews helped to verify concepts and emerging findings and identified additional studies for inclusion. Results were charted, analyzed thematically and summarized.We identified 34 articles from a wide range of disciplines and sectors, including 17 published research articles and 17 grey literature reports. Analysis of these articles revealed 15 distinct ways of using ICTs for good governance activities in LMIC health care. These use cases clustered into four conceptual categories: 1) gathering and verifying information on services to improve transparency and auditability 2) aggregating and visualizing data to aid communication and decision making 3) mobilizing citizens in reporting poor practices to improve accountability and quality and 4) automating and auditing processes to prevent fraud. Despite a considerable amount of implementation activity, we identified little formal evaluative research.Innovative digital approaches are increasingly being used to facilitate good governance in the health sectors of LMICs but evidence of their effectiveness is still limited. More empirical studies are needed to measure concrete impacts, document mechanisms of action, and elucidate the political and sociotechnical dynamics that make designing and implementing ICTs for good governance so complex. Many digital good governance interventions are driven by an assumption that transparency alone will effect change; however responsive feedback mechanisms are also likely to be necessary.

Pub.: 21 Sep '16, Pinned: 21 Sep '16