ABSTRACT. Social media has become an emergent phenomenon in education. Yet, there is a void in our understanding and research on teachers’ professional behavior within virtual spaces. This paper examines a sample of early career teachers’ curation of curriculum materials within Pinterest during a time of national education policy reform—the adoption of the Common Core State Standards in the U.S. By aligning the content of the resources that teachers actively seek out online to the Common Core State Standards, we provide an account of the instructional resources accessed and shared as teachers make sense and respond to the policy implementation. Furthermore, by leveraging a novel network approach—Epistemic Network Analysis, we provide a visualization of teachers’ conceptualization of varying content and how they relate instruction across standard within their curriculum organization. Finally, by comparing networks of teachers from different states, we find evidence of differences in teachers’ behavior patterns as they respond to policy change.

ABSTRACT. This paper presents findings from a study of high school student and teacher use of social media within the school context, focusing on the role that drama plays in their descriptions of personal and peer social media use. Findings suggest that most individuals feel social media drama has a constant presence in the high school context, but they also carefully position drama as an act that others engage in, rejecting the notion that they might be perpetrators of drama themselves.

ABSTRACT. In this paper we showcase how to utilize latent growth modeling (LGM) to investigate Twitter growth trajectories over a series of three years. While we analyze Twitter follower growth, LGM has a variety of uses in the social media/social movement context and can be applied to research questions that involve exploring growth over time. The purpose of this paper is to illustrate how LGM can be used in order to help social movement scholars interested in implementing a quantitative or mixed method approach to their research. To demonstrate the application of this technique we applied our model to an exemplar of 200 randomly selected higher educational institutions’ Twitter accounts. Our preliminary results reveal significant follower growth rate, however after controlling for the effect of school enrollment the growth rate disappeared. We conclude this paper with a discussion of our plans to resample our population, include only institutions at or above a considerable size, identify additional covariates/predictors to determine if we receive significant results.

ABSTRACT. The UnBias Youth Juries aim to engage young people in discussion of issues that affect their online lives, especially in relation to algorithms, through the presentation of relevant issues as prompts. Results from the first wave of juries, held in February 2017, produced valuable data about the concerns of young people and recommendations for improving their digital environments. Feedback regarding the structure of the juries suggested a series of interactive tasks to elicit more debate amongst the participants. This paper presents the results from a pilot study using these interactive tasks and discusses the plans for a second wave of juries. This includes creation of an advisory group to ensure the materials are relevant to the target audience, and an open educational resource that enables others to run youth juries with their own groups.

ABSTRACT. Due to the rising importance of social media platforms for news diffusion, newspapers are relying on social media editors to promote the distribution of their news items on these platforms. In this study, we investigate how much of an impact these social media editors really have, focusing on the impact of newspapers’ public pages on Facebook. Since the actions of individual users are not visible on many platforms due to privacy consideration, we propose a method that leverages time series of aggregated scores for total user engagement, which are available for various platforms. We use this method to study and compare the influence of Facebook pages for six newspapers from the UK, the Netherlands and Flanders, for all news items published over two weeks in 2017.

ABSTRACT. The problem associated with the propagation of fake news continues to grow at an alarming scale. This trend has generated much interest from politics to academia and industry alike. We propose a framework and method that detects and classifies fake news messages from Twitter posts using a hybrid of convolutional neural networks and long-short term recurrent neural network models. We find in our work that using this deep learning approach achieves an 82% accuracy. Intuitively identifying relevant features associated with fake news stories without previous knowledge of topic domain.

ABSTRACT. This paper presents preliminary findings of a content analysis of tweets posted by false accounts operated by the Internet Research Agency (IRA) in St Petersburg. We relied on a historical database of tweets to retrieve 4539 tweets posted by IRA-linked accounts in 2012-2017 and coded 2501 tweets manually. The messages cover US newsworthy events, the Charlie Hebdo terrorist attack in 2015, and the Brexit referendum in 2016. Tweets were annotated using 19 control variables to investigate whether IRA operations on social media are consistent with classic propaganda models. The results show that the IRA operates a composite of user accounts tailored to perform specific tasks, with the lion’s share of their work focusing on US daily news activity and the diffusion of polarized news across different national contexts.

ABSTRACT. Social media such as Twitter are now widely used by professional journalists in their day-to-day practice. Active use of such platforms for reporting the news and engaging with audiences requires a range of choices about the kind of public persona journalists wish to present; one key choice is what links to their own and other sources’ news stories they are prepared to share. This work-in-progress paper presents a comparative analysis of the news sharing choices of members of the Australian and German national press corps, showing considerably different selection patterns.

ABSTRACT. In this work-in-progress paper, we draw upon previous work and our own ongoing studies of the practices surrounding social media to discuss the conceptualization of social media use. We ask ‘what is social media use?’, and discuss how very different levels of engagement, ranging from active involvement with producing or consuming social media to more passive ways of monitoring or planning social media activities, tend to be summarized under the general notion of ‘social media use’. Our informants orient to social media activities even when they are not actively engaged with their phones. As a consequence, we argue that there is a problem of using time to define or measure social media use. In a ‘permanently online, permanently connected’ world, we need to move beyond such ways of conceptualizing how people live with technology.

ABSTRACT. Smartphone addiction and teenagers preoccupation with smartphones is a hot topic. The heavy use of smartphones may be associated with a form of psychological dependency on this technology. A one hour separation from smartphones was found to lead to continuously increasing anxiety by heavy users [Cheever et al., 2014]. Our research evaluated adolescents’ level of anxiety, technological dependency, and fear of missing out following separation from smartphones on a much longer time span. Each year, 34 fourteen- and fifteen-year-old adolescents sail around the globe for half a year. There is no smartphones allowed while sailing, although the adolescents receive their devices back during some stops. We gathered data before the trip (October 2017) and three weeks after being separated from smartphones (November 2017). Future data gatherings are planned. Research instruments included standardized questionnaires, in-depth interviews and short essays that teenagers wrote about their experiences being offline. First results show that teenagers with low smartphone usage and low level of technological anxiety/dependence are minimally affected by the separation. However, among teenagers with higher usage, some found the separation very difficult initially, and then easier as time passed; meanwhile others have shown the reverse tendency. Three weeks of separation were enough to influence teenagers’ level of technological anxiety/dependence, which has dropped significantly along with their positive attitudes towards media and technology. The teenagers’ fear of missing out was found to be correlated with technological anxiety/dependence and neuroticism.

ABSTRACT. Payment infrastructures are going through rapid change with the rise of next generation mobile networks and smartphone ownership. From mobile wallets to rideshare apps, social payments allow users to split receipts with friends, charge exes for breakup expenses, or troll celebrities. New layers of data, sociality, and markets are being created and influenced by expanding economic imaginaries, regulations, and business models leveraging these new infrastructures. In this paper we discuss how mobile payment systems have become social media. After discussing the recent history of mobile payments innovation—SMS, mobile wallets, delivery and ridesharing apps, we examine Venmo, a social payments platform that allows users to broadcast transactions to a social activity stream or public transaction feed. Our findings detail how transaction feeds of mobile payments support social practices, communication, and commerce with mobile devices and wireless networks. We present findings from a case study on Venmo to develop some implications for the design, study, and impact of mobile payment infrastructures as social media.

ABSTRACT. GPS-based dating apps have changed the way users find potential partners. Their minimalistic design, aimed at encouraging fast transitions between online match and offline encounter, leaves little space for users’ verbal self-presentation. As such, biographies become important tools of impression management. Based on an anonymized sample of 50,406 Tinder biographies, we analyze how users employ words to attract the right partner, and explore gendered differences in word choice and use.

ABSTRACT. This study explores privacy from the perspective of the user, leveraging a “framing in thought” approach to capture how users make sense of privacy in their social media use. Definitions of privacy collected from 608 social media users are analyzed through topic modelling and the clustering of word pairs to surface themes present in the data. Results indicate the dominance of frames related to horizontal privacy, or privacy vis-à-vis peers, over vertical privacy, i.e., that from institutions and governments. Themes relating to economic and legal frameworks had a reduced level of prominence. These findings suggest that user conceptualization of privacy reflects a cognate-based approach that emphasizes control and limits to information access.

ABSTRACT. In an effort to help users address privacy concerns, Facebook has created a complex system of controls to manage disclosure. Do these controls work in practice and what about controls for disclosure to Facebook itself? In this paper, we explore user relationships with Facebook and its privacy mechanisms. We confronted participants with privacy intruding situations using scenario building and explored their reactions. We then compared their answers to their actual practices by using Facebook’s own privacy mirrors. While the majority of respondents felt responsible for their data disclosure, they failed to live up to their own expectations. We argue that complexity of privacy controls places an unrealistic amount of epistemic responsibility on the users, while masking the way Facebook itself collects user data. We note the urgent need to establish clear basic privacy norms in SNSs and for our relationship with these companies.

ABSTRACT. Researchers have only just begun grappling with the ethical implications of social media research, since more research is conducted online in virtual communities. Ethical review boards may not have the understanding or training to advise on projects with elements of social media research. This paper is a reflexive account that explores the author’s decision to undertake research in a private Facebook community, of which she was already a member. This paper details the negative response that was received from community gatekeepers, and explores the research decisions which elicited this response. This paper uses Lee-Treweek & Linkogle’s four-part framework for understanding researcher risk, and presents in this instance that the risk of social isolation faced by the researcher was too significant to proceed with the study. Insider research, and netnography are two complicated areas of research. This paper contributes to ongoing learning in this growing field.

ABSTRACT. Older adults (65+) use a wide range of digital media, yet have been slow in adopting social media specifically. In this paper, we explore to what extent privacy concerns could be a barrier to social media adoption in this demographic. By analyzing in-depth interviews with 40 older adult users and non-users of social media living in East York, Toronto, Canada, we explore the types of social media privacy concerns that older adults have as well as the strategies they employ to mitigate these concerns. We found that older adult social media users and non-users shared similar privacy concerns; the most popular being a concern for unauthorized access to personal information, and information misuse. While older adult non-users of social media protected themselves by avoiding social media, older adults who were social media users protected themselves by limiting the information they shared. This study has policy implications for training programs geared toward older adults and informs understandings of privacy literacy across the lifespan.

ABSTRACT. The media environment is becoming more difficult to navigate as individuals become responsible for seeking out their own understanding of the world rather than relying on the nightly news. This panel pulls together a set of scholars who have been working on questions related to online information diffusion. This panel will explore questions related to the cognitive processes, attitude formation, interpersonal networks, media ecology, and platform design that inform how information is created, consumed, and believed via social media.

The Social Media + Society conference cuts across disciplinary lines to bring together scholars interested in all types of questions related to social media use. This panel attempts to capitalize on that benefit by allowing research teams who have been working on questions related to processes that inform our understanding of how people create, share, evaluate, and believe information online to consider angles of the question that may previously been obscured by disciplinary boundaries. We hope this panel will provide a space for a) a deep level of engagement from scholars working on questions related to online information diffusion, b) the sharing of insights across research teams working on similar questions, and c) inspire future collaborations on these questions between both panelists and audience members.

ABSTRACT. This panel brings together researchers with different educational and political backgrounds, from information and media studies to political science and journalism who all study political communication on Facebook, albeit with different methodological frameworks. Three out of four presenters take their point of departure in data from Denmark but as the possibilities, problems and limitations discussed are inherent in the study of social media in general, the points will be relevant for everybody studying social media politics. Further, although the presenters focus mainly on Facebook, most topics addressed apply to the study of social media in general. The order of presentations move from the overall macro study of really big data sets, through longitudinal survey studies to more narrow data sets on politicians’ and voters’ interactions in a particular election campaign.

ABSTRACT. Anonymous social media provide opportunities and challenges for the organizations where they are used—but relatively little is known about how these tools are used and perceived by various types of organizational members. This paper reports preliminary findings from a survey of Microsoft employees using the anonymous smartphone application called Blind. Results suggest that this technology is used frequently for a variety of purposes related to seeking information, seeking support, and (less so) expressing opinions. The tool is seen as relatively useful and anonymous, but only moderately safe and trustworthy. Significant differences in perceived anonymity, perceived trust/safety, and use of the tool for seeking support are found based on frequency of use, as well as tenure, age, and hierarchical level of users. The paper offers theoretical and practical implications of findings and directions for further work in this emerging area.

ABSTRACT. Some recent studies have revealed that cyber criminals tend to exchange cyber attack knowledge in online social networks (OSNs). Cyber security experts are another information providers on OSNs who frequently share information about cyber security incidents and their personal opinions and analyses. Therefore, in order to improve our knowledge about evolving cyber attacks and the underlying human behavior for different purposes (e.g., prediction and prevention of impeding cyber attacks, crime investigation, understanding career development and business models of cyber criminals and cyber security professionals), it is very useful to automatic detect cyber security related accounts on OSNs and monitor their activities. This paper reports our preliminary work on automatic detection of cyber security related accounts on OSNs using Twitter as an example, which hopefully will allow us to discover unknown cyber security experts and cyber criminals for monitoring purposes. Three machine learning based classification algorithms were applied and compared: decision trees, random forests, and SVM (support vector machines). Experimental results showed that both decision trees and random forests had performed well with an overall accuracy over 95%, and when random forests was used with behavioral features the accuracy had reached as high as 97.877%.

ABSTRACT. The social media ecosystem is constituted by platforms and applications with varying affordances for and rules regarding anonymity. How and why users use and view online anonymity remains a focus of social media and Internet research, including how beliefs about anonymity shape - and are shaped by - the behaviors of others on and beyond social media platforms. Our study contributes to knowledge about anonymity in online social platforms by exploring the shared beliefs of a sample of young adults about the recently-defunct platform Yik Yak. Findings from our mixed-methods study of students at a large university in the Southeastern United States suggest that young adults have shared beliefs about social media and anonymity, but the degree to which those beliefs are shared is tenuous. The specific of those beliefs are discussed in context of prior research about social media, anonymity and cyberbullying.

ABSTRACT. A large number of studies have examined the role of personal networks, social capital, and social support in ones’ subjective well-being. However, there is still a death of study to understand how personal networks were constructed by multimodal media use, and in turn, influence individuals’ well-being simultaneously with other impact factors. This study exceptionally addressed the context of Chinese internal migrant workers who have dynamic network constructs due to their mobility nature. Based on survey data collected from 504 Chinese internal migrant workers, the study found that WeChat and mobile phone calling intensity predicted bonding social capital, bridging social capital was predicted by only by QQ intensity. Two dimensions of personal networks – density and tie multiplexity-explained migrant workers’ well-being. Notably, perceived social support consistently predicted well-being outcomes and played a mediating role between tie proximity and well- being. Importantly, bridging social capital and weak-tie in the personal network were found to be strongly and significant predictors of well-being, indicating migrant workers’ weak-tie support preference. The results offer implications for a more nuanced understanding of the role of diverse media use in ones’ personal network construct, and subsequently influence their social capital, social support, and well-being outcomes.

ABSTRACT. This study explored how Twitter users engaged with and responded to a popular Netflix series on the topic of teen suicide. This research is set within an emerging research stream focused on how people use social media to discuss issues, events, or in this case, an entertainment product. Because this study focused on a Netflix program that released all 13 episodes on the day it premiered, it differed from previous studies that explored a second-screen experience on Twitter. Topics discussed by users in original tweets and most popular retweets, the communities formed around the content, and the relationship between follower count and retweets are examined in this study.

ABSTRACT. Social media provides users with public and persistent transcripts of conversations between their friends. Although often unintentional, these records can be interpreted as social exclusion signals. This study randomly assigned 163 participants to one of two conditions where they were exposed to hypothetical written scenarios describing conversations between their friends in which they were excluded or included. Results suggest that individuals in the excluded condition experienced more negative and less positive affect than those in the included condition. Further, we found that network monitoring-a multidimensional construct assessing individual differences in cognitive resources dedicated towards understanding one's social network-affected individuals' emotional responses to the stimuli. Specifically, network social awareness and advantage were significant predictors of negative affect in the excluded condition. Results are discussed in terms of theoretical and practical implications.

ABSTRACT. This preliminary research study examines the emotions expressed on Instagram by women farmers through the hashtag #womenwhofarm. Previous research has found a positivity bias on Instagram, wherein users predominantly express themselves with a positive emotional tone. Using open-ended coding of 651 Instagram posts, this study finds that women farmers use a mostly neutral emotional tonality in their images and text content.

ABSTRACT. How do teens deal with the complexities of the digital world? Although many scholars have asked this question, few have considered not only the interpersonal concerns which loom large in the lives of teens and young adults but also their relationship with the social media platforms which have become so central to their lives. In this paper we present a study of Danish teen attitudes towards social media platforms they use most – Facebook and Snapchat. We find that gender differences in attitudes towards these platforms and differences in the way teens perceive each platform. Finally, we illustrate success of an digital education effort in changing teen attitudes.

ABSTRACT. Social media have been widely credited for facilitating young people’s political engagement, most notably by providing a conducive platform for political expression. There has been little attention, however, to the possible pitfalls for young people in engaging with politics on social media. Through in-depth interviews, this study presents the paradoxical case of a group of politically active youths who have no qualms with participating in an offline large-scale protest but are wary of publicly engaging with politics on social media. The findings indicate that perceptions of hostility, social risk, and futility impede online political expressions while the promises of embodied experience draw participants toward offline participation. Rather than disengage with politics completely on social media, the young people in this study adopt certain “disconnective practices”—selected functions, audiences, and apps—to manage their relationships with others in their social networks who may or may not share their political views.

ABSTRACT. This study addresses how young Danish adults (18-26 years old) describe and critically reflect on the role of Face-book, the algorithm and the news feed as a way to receive information and relate to the world. The analysis is based on a critical framework by Feenberg’s and Ihde’s theoretical work on the relationship between subject, technology and life world. The empirical material is based on ten individual interviews, which were thematically coded. Preliminary findings point towards an omnipresent role of Facebook, which users seem to be critically aware of. While all informants are clearly critical towards the role Facebook has in society as well as in their daily life, this does not always translate into concrete actions to mitigate, improve or avoid negative consequences of their individual Facebook use. In conclusion, we suggest that further research focuses on conceptualizing the emancipatory potential of a more direct relation with Facebook as a conceptual other, rather than understanding Facebook as a simple technology through which information about and experience with the world is simply facilitated.

ABSTRACT. This study is a segment of a larger investigation of high school students’ use of social media platforms and networks. In the spring of 2017, the researchers conducted detailed, individual interviews with high school students (10th and 12th grade) about their social media platforms and networks. During interviews inquiring about their social media practices more generally, many students shared examples of their informal learning practices on social media. Across these informal learning practices, three themes emerged: career/future planning, entrepreneurship, and hobbies.

ABSTRACT. Air quality is recognized to be major risk factor for human health. Air quality information is now becoming increasingly available globally, allowing scientists to estimate potential health impacts and economic damage attributable to poor ambient air. These estimates are difficult to verify and often the impacts are only known after a period of time. In most countries, public health records are either poorly maintained or very difficult to access and thus any association of health impacts to air pollution events is difficult to make. Also, the ability of people to understand air quality information and take actions to protect their health is not clear. Here, we analyze Twitter data in three major cities around the world: Paris, London, and New Delhi, and determine the correlation of tweet frequencies to their local air pollution episodes. Our analysis determined the top 3 hashtags for each city that best correlated tweets to air quality events. Using tweets with just these 3 hashtags, we determined that people's response to air quality in the three cities were nearly identical when considering relative changes in air pollution. Using machine learning, we determined that health concerns dominated the public response when air quality degraded, with the strongest increase in concern being in New Delhi, where air pollution is the highest. Our analysis also suggests that the public are looking for local political solutions to the problem of poor air quality. Our analysis procedure can be extended to understanding changes in air quality and the related public response at a global level.

ABSTRACT. This paper presents an in-depth qualitative of analysis of n=13,373 tweets which relate to the peak of the Swine Flu outbreak of 2009, and the Ebola outbreak of 2014. Tweets were analysed using thematic analysis and a number of themes and sub-themes were identified. The results were brought together in an abstraction phase and the commonalities between the cases were studied. An interesting similarity which emerged was the rate at which Twitter users expressed intense fear and panic akin to that of the sociological concept of a moral panic. Moreover, a number of discussions were found to emerge which were not reported in previous literature. Our study is the first and largest in-depth analysis of tweets on infectious diseases. Our results will inform public health strategies for future infectious disease outbreaks. Future work will seek to conduct further comparisons and explore relevant health theory.

ABSTRACT. Preppers are individuals who gather and stock supplies, food, and tools in anticipation of disasters or emergencies. In the process of making themselves ready for disaster, many preppers have built community on social networking sites and online platforms such as Reddit, Tumblr, Pinterest, YouTube and subject specific bulletin boards. This paper presents findings from a study of preppers’ #EDC (“everyday carry”) pocket dumps photographs posted to Instagram, a mobile social media platform that allows users to post photographs, comment, tag, and follow users in the social network. Amongst other subcultures of consumption and display, preppers’ pocket dumps are a form of self-representation and an emerging personal media genre on social media. We argue that preppers’ perceptions of future disaster and performing preparedness on visual social media extends current social media research on self-representation practices, consumption of objects, and personal photography in surprising and meaningful ways.

ABSTRACT. Wildfires have significant effects on human populations worldwide. Smoke pollution, in particular, from either prescribed burns or uncontrolled wildfires, can have profound health impacts, such as reducing birth weight in children and aggravating respiratory and cardiovascular conditions. Scarcity in the measurements of particulate matter responsible for these public health issues makes addressing the problem of smoke dispersion challenging, especially when fires occur in remote regions. Previous research has shown that in the case of the 2014 King fire in California, crowdsourced data can be useful in estimating particulate pollution from wildfire smoke. In this paper, we show that the previous model continues to provide good estimates when extended statewide to cover several wildfires over an entire season in California. Moreover, adding the semantic information contained in the social media data to the predictive model significantly increases model accuracy, indicating a confluence of social and spatio-temporal data.