CJC at the 2021 International Communication Association Conference
May 27-31, 2021
Virtual Conference
Jordan Alpert, assistant professor, Department of Advertising, and Carma Bylund, professor, Department of Public Relations
Transactional eHealth literacy in cancer: Comparing skills of patients and caregivers
Authors: Taylor S. Vasquez, Carma L. Bylund, Jordan Alpert, Julia Close, Tien Le, Merry Jennifer Markham, Greenberry T. Taylor and Samantha R. Paige
Abstract: Cancer patients and caregivers of patients with cancer have reported different experiences using the Internet to learn about cancer. This study examines the eHealth literacy of patients and caregivers, including an investigation about how each group evaluates online cancer information. Between October 2019 and January 2020, we conducted an online survey with 282 cancer patients and caregivers. We used hierarchical linear regression analyses to explore differences in functional, communicative, critical, and translational eHealth literacies between patients and caregivers. We also conducted a thematic content analysis to examine processes for how each group evaluates online cancer information. eHealth literacy scores did not differ between cancer patients and caregivers, even after adjusting for socio-demographic variables. Patients and caregivers considered the credibility of online cancer information based on its channel (e.g., National Institutes of Health). However, in evaluating online information, caregivers were more likely than patients to consider the presence and quality of scientific references supporting the information. Patients were more likely than caregivers to cross-reference other websites and online sources to establish consensus. In conclusion, online cancer information accessibility and evaluation procedures are different among cancer patients and caregivers of patients with cancer and should be considered in future efforts to design online cancer education.
Raegan Burden, lecturer of cultural engagement, Department of Advertising
“I Am Speaking.” 2020 VP Nominee Kamala Harris’ Impact of Black Feminism as Social Influencers on Twitter
Authors: Rachel Grant, Raegan Burden, Spenser Cheek
Award: Top Paper Award in Ethnicity and Race in Communication Division
Abstract: By using social media as a cultural, political mobilizing tool, Black politicians of both genders have achieved elevated engagement and influence among voters. Online movements and social justice campaigns such as #BlackLivesMatters, #OscarsSoWhite, and #SayHerName have created resistive counternarratives that challenge the dominant, mainstream media. With the Black community making up the largest population of social media users, Twitter, specifically “Black Twitter,” has turned into an online space of interconnection within the Black community to discuss trending topics or issues.
By examining how Vice President Kamala Harris (who was the vice-presidential nominee at the time of the study) used Twitter as a social media platform to achieve political prominence the study created a lens through which future researchers and the public may better understand the redefinition of Black women in both social media and politics, as well as understand the broader issue of oppression among all marginalized, less mainstream communities.
Sylvia Chan-Olmsted, professor, Department of Media Production, Management and Technology
Smart speakers require smart management: How user gratifications affect privacy settings via social presence and privacy concerns.
Authors: Kun Xu, Sylvia Chan-Olmsted, Fanjue Liu
Abstract: Smart speakers’ voice recognition technology has not only advanced the communication efficiency between users and machines, but also raised users’ privacy concerns. As smart speakers listen to users’ voice commands and collect audio data to improve algorithms, it is crucial to understand how users manage their privacy settings to protect their personal information. Combining the uses and gratifications approach, communication privacy management theory, and the media equation, this study recruits 1052 participants to ask about their daily smart speaker use experiences. The study explores the major gratifications that users seek, identifies two strategies that users adopt to manage their privacy, and suggests that users apply similar interpersonal privacy management rules to interactions with smart speakers.
Trust building and trust transfer: Consumer decisions in influencer commerce.
Authors: Rang Wang and Sylvia Chan-Olmsted
Abstract: Based on the argument that influencer-follower relationship is at the core of influencer commerce, this study draws on the trust concept and the trust transfer theory to explain influencers’ impact on followers’ brand-related attitudes and behaviors. It examined factors affecting followers’ trust in influencers and how this influencer trust affected brand-related outcomes. The proposed model explained most of the variance in influencer trust and sufficient variance in the three brand-related outcomes. As such, this study developed a trust building model devoted to the influencer commerce context and offered support for the trust transfer theory. The findings are significant in our understanding of a contemporary, integrated media and commerce environment and the central role that a new kind of “agent” plays in the process.
Spenser Cheek, doctoral student
“I Am Speaking.” 2020 VP Nominee Kamala Harris’ Impact of Black Feminism as Social Influencers on Twitter
Authors: Rachel Grant, Raegan Burden, Spenser Cheek
Award: Top Paper Award in Ethnicity and Race in Communication Division
Abstract: By using social media as a cultural, political mobilizing tool, Black politicians of both genders have achieved elevated engagement and influence among voters. Online movements and social justice campaigns such as #BlackLivesMatters, #OscarsSoWhite, and #SayHerName have created resistive counternarratives that challenge the dominant, mainstream media. With the Black community making up the largest population of social media users, Twitter, specifically “Black Twitter,” has turned into an online space of interconnection within the Black community to discuss trending topics or issues.
By examining how Vice President Kamala Harris (who was the vice-presidential nominee at the time of the study) used Twitter as a social media platform to achieve political prominence the study created a lens through which future researchers and the public may better understand the redefinition of Black women in both social media and politics, as well as understand the broader issue of oppression among all marginalized, less mainstream communities.
Huan Chen, associate professor, Department of Advertising
A Revised Social Identity Model of Deindividuation Effects: An Ecological and Algorithmic Perspective
Authors: Yang Feng, San Diego State University, Huan Chen
Abstract: Since the social identity model of deindividuation (SIDE) was proposed, it has become one of the most dominant theories of computer-mediated communication (CMC). However, it has several limitations. To address the SIDE model’s limitations, we propose the ESIDE model that is derived from the media ecology and social constructionism perspectives and built from the focus theory of normative conduct, social identity theory, and the encoding/decoding model, the proposed ESIDE model, to offer an updated theoretical framework to explain CMC in today’s algorithmic media environment. The model has four premises: (a) group norms emerge dynamically; (b) group norms shall govern behavior even if they differ from societal norms; (c) there are two ways of norm conformity – exact versus variant; and (d) visual anonymity and group immersion lead to intergroup conflict. The model was tested and validated with social media data collected from two gender-related campaigns on YouTube.
Mo Chen, doctoral student
How to Build a Credible and Socially Present Robot? A Meta-Analysis of the Power of Social Cues in Human-Robot Interaction
Authors: Mo Chen, Leping You, Miami University, Kun Xu
Abstract: Social cues have been construed as an important concept in human-robot interaction (HRI), as they can be manipulated to reflect robots’ perceived gender, personalities, emotions, identities, and so on. This study seeks to understand the distinct effects of individual social cues and uses meta-analysis to explore a hierarchy of social cues that elicit different degrees of social responses. A total of 25 and 44 effect sizes were calculated respectively to represent the magnitude of the effects of social cues on users’ social presence and trust in HRI. Results suggest that although the overall effects of social cues are small, manipulating robots’ facial and kinetic cues can induce medium-to-large sized effects on users’ social presence and trust. The results of the meta-analysis can contribute to both the theoretical refinement of the Computers are Social Actors paradigm and the practical and methodological design of human-robot interaction.
Do we heuristically trust machine generated information: the perceived credibility of information sources?
Authors: Mo Chen, Fanjue Liu
Eliana DuBosar, doctoral student
Examining the Relationship Between Media Use, Ideology, Political Trust, and Political Participation
Abstract: This paper examines the conditional indirect relationship between media use and political participation. Specifically, it looked at whether the relationship between media use and political participation was mediated by political trust and whether the relationship between media use and political trust was moderated by ideology. This was done by analyzing data from the American National Election Studies (ANES) 2016 Time Series Study. Consistent with prior research, results showed a relationship between media use and participation. More specifically, as media use increased, so too did political participation. However, the conditional indirect effects of ideology as a moderator and trust as a mediator as well as the relationship between media use and political participation were not found to be statistically significant. This paper concludes with implications and suggestions for future research based on the findings.
Women on the Trail: Political Entertainment and Trust
Authors: Eliana DuBosar, Myiah Hutchens
Abstract: This study examined the effects of three different television genres on perceptions of a political candidate during an election period. Specifically, students at a large Southeastern university (N = 172) were assigned to one of three conditions: late-night comedy, traditional news, and sketch comedy to assess differential impacts on perceived trust, presidentiality, and electability of Elizabeth Warren during the 2020 Democratic presidential primary period. Results showed that participants in the late-night condition rated Warren as more trustworthy than those in the traditional news condition. Similarly, Warren was rated as both more electable and having more leadership potential by those assigned to the late-night condition as compared to those in the traditional news condition. Finally, this study found that political knowledge, including both civics and current events knowledge, mediated the relationship between condition and trust. More specifically, results showed that as knowledge increased, participants in the sketch comedy condition reported higher levels of trust in Warren whereas the opposite occurred for those in the traditional news condition. Results demonstrate the importance of understanding how politicians (or their likeness) are perceived in different television settings and paints more of a picture of how appearing in certain media might differentially affect viewer perceptions of politicians.
Examining the Relationship Between Media Use, Ideology, Political Trust, and Political Participation
Abstract: This paper examines the conditional indirect relationship between media use and political participation. Specifically, it looked at whether the relationship between media use and political participation was mediated by political trust and whether the relationship between media use and political trust was moderated by ideology. This was done by analyzing data from the American National Election Studies (ANES) 2016 Time Series Study. Consistent with prior research, results showed a relationship between media use and participation. More specifically, as media use increased, so too did political participation. However, the conditional indirect effects of ideology as a moderator and trust as a mediator as well as the relationship between media use and political participation were not found to be statistically significant. This paper concludes with implications and suggestions for future research based on the findings.
Carla Fisher, associate professor, Department of Advertising
Work-Life-Imbalance During the COVID-19 Pandemic: Exploring Social Support Behavior and Health Outcomes
Authors: Kevin Wright, George Mason University, Wendy Riemann, George Mason University, Carla Fisher
Abstract: The current study explored work-life imbalance and social support during the Spring 2020 shelter-in-place restrictions in the United States and their impact on perceived stress and depression. The study findings indicate that perceptions of work-life imbalance during the COVID-19 restrictions was a significant predictor of both perceived stress and depression. Findings also indicate that higher work-life imbalance predicts higher weak-tie support preference, and lower perceived social support predicted higher weak-tie support network preference. These findings suggest that closer ties may not necessarily meet all of the social support needs of an individual during a crisis like COVID-19.
Sarah Fisher, doctoral student
The Mentoring Model: Eliminating Redundancies for Improved Online Higher Education
Abstract: Educational institutions suffered during the COVID-19 shutdowns worldwide as many professors struggled with the online platform and how to effectively reach students. The workload was increased as professors had to create lesson plans and grade assignments as usual, but now had the added challenge of interacting with students from a distance. For many professors, this created an unreasonable time burden through extra hours spent in Zoom meetings and emails with students who were searching for human interaction as well as understanding of the course materials. The expectation of individual response resulted in many exhausted professors trying to juggle the demands of the constantly connected classroom with their own personal situations during the COVID-19 pandemic. Is this the new normal or is there a light at the end of the online tunnel?
Rachel Grant, assistant professor, Department of Journalism
“I Am Speaking.” 2020 VP Nominee Kamala Harris’ Impact of Black Feminism as Social Influencers on Twitter
Authors: Rachel Grant, Raegan Burden, Spenser Cheek
Award: Top Paper Award in Ethnicity and Race in Communication Division
Abstract: By using social media as a cultural, political mobilizing tool, Black politicians of both genders have achieved elevated engagement and influence among voters. Online movements and social justice campaigns such as #BlackLivesMatters, #OscarsSoWhite, and #SayHerName have created resistive counternarratives that challenge the dominant, mainstream media. With the Black community making up the largest population of social media users, Twitter, specifically “Black Twitter,” has turned into an online space of interconnection within the Black community to discuss trending topics or issues.
By examining how Vice President Kamala Harris (who was the vice-presidential nominee at the time of the study) used Twitter as a social media platform to achieve political prominence the study created a lens through which future researchers and the public may better understand the redefinition of Black women in both social media and politics, as well as understand the broader issue of oppression among all marginalized, less mainstream communities.
Myiah Hutchens, assistant professor, Department of Public Relations
The good, the bad, and the evil media: Influence of online comments on media trust.
Authors: Myiah Hutchens, Ekaterina Romanova
Abstract: Trust in institutions is declining, and in the United States, trust in news media hit a low in 2016. Given the importance of media in a democracy, understanding what impacts trust is of critical importance. The manuscript uses two experiments to examine the impact of online comments in assessments of media trust, outlet trust, and intentions to use a source in the future. The first experiment varied the tone of the comments and determined that negative comments reduced media trust and outlet trust in comparison to positive comments. The second study examined the target of the comments and determined that negative comments targeting the author of the story decreased media trust, and negative comments targeting the outlet reduced trust in the specific media outlet. Neither tone of comments nor comment targets were related to future intentions to use a source in the future.
Women on the Trail: Political Entertainment and Trust
Authors: Eliana DuBosar, Myiah Hutchens
Abstract: This study examined the effects of three different television genres on perceptions of a political candidate during an election period. Specifically, students at a large Southeastern university (N = 172) were assigned to one of three conditions: late-night comedy, traditional news, and sketch comedy to assess differential impacts on perceived trust, presidentiality, and electability of Elizabeth Warren during the 2020 Democratic presidential primary period. Results showed that participants in the late-night condition rated Warren as more trustworthy than those in the traditional news condition. Similarly, Warren was rated as both more electable and having more leadership potential by those assigned to the late-night condition as compared to those in the traditional news condition. Finally, this study found that political knowledge, including both civics and current events knowledge, mediated the relationship between condition and trust. More specifically, results showed that as knowledge increased, participants in the sketch comedy condition reported higher levels of trust in Warren whereas the opposite occurred for those in the traditional news condition. Results demonstrate the importance of understanding how politicians (or their likeness) are perceived in different television settings and paints more of a picture of how appearing in certain media might differentially affect viewer perceptions of politicians.
Benjamin Johnson, assistant professor, Department of Advertising
The dark at the end of the tunnel: Doomscrolling on social media newsfeeds.
Authors: Bhakti Sharma, Susanna Lee, & Benjamin Johnson
Abstract: Doomscrolling refers to a unique media habit where social media users persistently attend to negative information in their newsfeeds about crises, disasters, and tragedies. Given the potential prevalence of this practice and its timely but chronic relevance for user experience, political perceptions, well-being, and other outcomes, this project aims to develop and validate a self-report measure of doomscrolling. Participant interviews and an expert panel contribute to refinement of an item bank, assessed in two survey studies for item structure and reliability, and for construct validity.
Are they being authentic? The effects of self-disclosure and message sidedness on sponsored post effectiveness.
Authors: Susanna Lee and Benjamin Johnson
Abstract: With increasing fragmentation of the social media marketing landscape, influencer marketing with self-promoted individuals, also known as “influencers,” has arisen as a relatively novel form of celebrity endorsement. However, the mechanisms behind creating an authentic and credible sponsored influencer post remain undeciphered. Thus, this study examines the effect of self-disclosures (present/not present) and message sidedness (one-sided/two-sided) on attitude toward the brand, influencer authenticity, influencer credibility, purchase intention, and eWOM intention. The findings revealed that Instagram product reviews with two-sided messages yielded higher influencer authenticity, influencer credibility, attitudes toward the ad, and eWOM intentions than Instagram product reviews with one-sided messages. Theoretical, practical, and managerial implications are discussed.
Plurality in the measurement of social media use and psychological well-being among adolescents and young adults
Authors: Diamantis Petropoulos Petalas, Elly Konijn, Benjamin Johnson, Jolanda Veldhuis, & others
Abstract: On a daily basis, individuals between 12 and 25 years of age engage with their mobile devices for many hours. Social Media Use (SMU) has important implications for the social life of younger individuals in particular. However, measuring SMU and its effects often poses challenges to researchers. In this exploratory study we focus on some of these challenges, by addressing how plurality in the measurement and age-specific characteristics of SMU can influence its relationship with measures of subjective mental health (MH). We conducted a survey among a nationally representative sample of Dutch adolescents and young adults (N = 3,669). Using this data we show that measures of SMU show little similarity with each other, and that age-group differences underlie SMU. Similar to the small associations previously shown in social media-effects research, we also find some evidence that greater SMU associates to drops and to increases in MH. Albeit nuanced, associations between SMU and MH were found to be characterized by both linear and quadratic functions. These findings bear implications for the level of association between different measures of SMU and its theorized relationship with other dependent variables of interest in media-effects research.
Susanna Lee, doctoral student
Are they being authentic? The effects of self-disclosure and message sidedness on sponsored post effectiveness.
Authors: Susanna Lee and Benjamin Johnson
Abstract: With increasing fragmentation of the social media marketing landscape, influencer marketing with self-promoted individuals, also known as “influencers,” has arisen as a relatively novel form of celebrity endorsement. However, the mechanisms behind creating an authentic and credible sponsored influencer post remain undeciphered. Thus, this study examines the effect of self-disclosures (present/not present) and message sidedness (one-sided/two-sided) on attitude toward the brand, influencer authenticity, influencer credibility, purchase intention, and eWOM intention. The findings revealed that Instagram product reviews with two-sided messages yielded higher influencer authenticity, influencer credibility, attitudes toward the ad, and eWOM intentions than Instagram product reviews with one-sided messages. Theoretical, practical, and managerial implications are discussed.
Transparency management of social media influencers: Motivation, tenure, and status
Authors: Lee, S., Won, J., & Shin, J.
Abstract: As the power of YouTubers has considerably grown as influencers in recent years, they have been called to greater accountability and transparency. Although there exist external guidelines and regulations surrounding disclosure, content creators’ practice still varies widely. This study examines this phenomenon focusing on beauty YouTubers, a distinct community connected by shared styles, routines, and language. We collected information about 1,067 beauty YouTube channels and examined the use of disclosure-related keywords by the creators, which we term “transparency management”. Using automated text analysis, we observed that over 60% of channels mentioned at least one keyword, while showing a wide range of frequency of the use. We also found that those who were professionally motivated YouTubers tended to engage in transparency management more than those who lacked such motivation. We also found that YouTubers with a large number of subscribers were more likely to engage in transparency management than those with a small number of subscribers. This article contributes to a better understanding of content creators’ communication patterns that manifest transparency.
Fanjue Liu, doctoral student
Smart speakers require smart management: How user gratifications affect privacy settings via social presence and privacy concerns.
Authors: Kun Xu, Sylvia Chan-Olmsted, Fanjue Liu
Abstract: Smart speakers’ voice recognition technology has not only advanced the communication efficiency between users and machines, but also raised users’ privacy concerns. As smart speakers listen to users’ voice commands and collect audio data to improve algorithms, it is crucial to understand how users manage their privacy settings to protect their personal information. Combining the uses and gratifications approach, communication privacy management theory, and the media equation, this study recruits 1052 participants to ask about their daily smart speaker use experiences. The study explores the major gratifications that users seek, identifies two strategies that users adopt to manage their privacy, and suggests that users apply similar interpersonal privacy management rules to interactions with smart speakers.
Do we heuristically trust machine generated information: the perceived credibility of information sources?
Authors: Mo Chen, Fanjue Liu
Ekaterina Romanova, doctoral student
The good, the bad, and the evil media: Influence of online comments on media trust.
Authors: Myiah Hutchens, Ekaterina Romanova
Abstract: Trust in institutions is declining, and in the United States, trust in news media hit a low in 2016. Given the importance of media in a democracy, understanding what impacts trust is of critical importance. The manuscript uses two experiments to examine the impact of online comments in assessments of media trust, outlet trust, and intentions to use a source in the future. The first experiment varied the tone of the comments and determined that negative comments reduced media trust and outlet trust in comparison to positive comments. The second study examined the target of the comments and determined that negative comments targeting the author of the story decreased media trust, and negative comments targeting the outlet reduced trust in the specific media outlet. Neither tone of comments nor comment targets were related to future intentions to use a source in the future.
Bhakti Sharma, doctoral student
The dark at the end of the tunnel: Doomscrolling on social media newsfeeds.
Authors: Bhakti Sharma, Susanna Lee and Benjamin Johnson
Abstract: Doomscrolling refers to a unique media habit where social media users persistently attend to negative information in their newsfeeds about crises, disasters, and tragedies. Given the potential prevalence of this practice and its timely but chronic relevance for user experience, political perceptions, well-being, and other outcomes, this project aims to develop and validate a self-report measure of doomscrolling. Participant interviews and an expert panel contribute to refinement of an item bank, assessed in two survey studies for item structure and reliability, and for construct validity.
Jieun Shin, assistant professor, Department of Media Production, Management and Technology
Coming Together in Tough Times : NGOs’ Covid-19 issue discourse on social media
Authors: Li, Y., Shin, J., Yang, A., Sun, J., Kim, H.
Abstract: NGOs are important civil actors in societies’ emergency and disaster responses, and they come together on social media to identify prominent issues and coordinate issue responses. This research explores how U.S. NGO form topic-driven communities on social media to discuss and build representational strategic networks around issues related to the COVID-19 pandemic. Drawing from Issue Niche Theory, we examined how NGOs’ networks and discourse evolved before and after the general public paid great attention to the COVID-19 issue and how such patterns changed across the whole issue niche and sub-issue niches. We analyzed the evolution of Twitter-based networks and discourse of 2,588 NGOs in the first five months of the COVID-19 outbreak in the United States. Our analysis revealed important factors that shape tie formation patterns in the NGOs’ communities in this novel issue niche.
Communicating stakeholders’ national crisis responses to the COVID-19 on Twitter.
Authors: Sun, J., Shin, J., Yang, A, & Liu, W.
Abstract: As societies face mounting challenges posed by the COVID-19 crisis, multi-stakeholder issue networks are imperative to the mobilization of resources from government, business, and nonprofit sectors. Combining organizational community ecology theory and issue niche theory, this study examines the evolution of cross-sector communication networks among 1,980 organizations on Twitter as the COVID-19 issue evolves from the imminent stage to the prominent stage. The analysis reveals that the increase of public attention to the COVID-19 issue is accompanied by a greater number of cross-sectoral ties in all networks examined. Furthermore, we find that issue niche width, businesses’ industry type, and nonprofits’ issue identity all influence tie formation patterns in government–business, government–nonprofit, and business–nonprofit networks. Meanwhile, fatal COVID-19 cases and unemployment rates mainly affect government agencies’ relationship building, but not the other two types of organizations.
Transparency management of social media influencers: Motivation, tenure, and status
Authors: Lee, S., Won, J., & Shin, J.
Abstract: As the power of YouTubers has considerably grown as influencers in recent years, they have been called to greater accountability and transparency. Although there exist external guidelines and regulations surrounding disclosure, content creators’ practice still varies widely. This study examines this phenomenon focusing on beauty YouTubers, a distinct community connected by shared styles, routines, and language. We collected information about 1,067 beauty YouTube channels and examined the use of disclosure-related keywords by the creators, which we term “transparency management”. Using automated text analysis, we observed that over 60% of channels mentioned at least one keyword, while showing a wide range of frequency of the use. We also found that those who were professionally motivated YouTubers tended to engage in transparency management more than those who lacked such motivation. We also found that YouTubers with a large number of subscribers were more likely to engage in transparency management than those with a small number of subscribers. This article contributes to a better understanding of content creators’ communication patterns that manifest transparency.
Taylor Vasquez, doctoral student
Transactional eHealth literacy in cancer: Comparing skills of patients and caregivers
Authors: Taylor S. Vasquez, Carma L. Bylund, Jordan Alpert, Julia Close, Tien Le, Merry Jennifer Markham, Greenberry T. Taylor and Samantha R. Paige
Abstract: Cancer patients and caregivers of patients with cancer have reported different experiences using the Internet to learn about cancer. This study examines the eHealth literacy of patients and caregivers, including an investigation about how each group evaluates online cancer information. Between October 2019 and January 2020, we conducted an online survey with 282 cancer patients and caregivers. We used hierarchical linear regression analyses to explore differences in functional, communicative, critical, and translational eHealth literacies between patients and caregivers. We also conducted a thematic content analysis to examine processes for how each group evaluates online cancer information. eHealth literacy scores did not differ between cancer patients and caregivers, even after adjusting for socio-demographic variables. Patients and caregivers considered the credibility of online cancer information based on its channel (e.g., National Institutes of Health). However, in evaluating online information, caregivers were more likely than patients to consider the presence and quality of scientific references supporting the information. Patients were more likely than caregivers to cross-reference other websites and online sources to establish consensus. In conclusion, online cancer information accessibility and evaluation procedures are different among cancer patients and caregivers of patients with cancer and should be considered in future efforts to design online cancer education.
Rang Wang, doctoral student
Trust building and trust transfer: Consumer decisions in influencer commerce.
Authors: Rang Wang and Sylvia Chan-Olmsted
Abstract: Based on the argument that influencer-follower relationship is at the core of influencer commerce, this study draws on the trust concept and the trust transfer theory to explain influencers’ impact on followers’ brand-related attitudes and behaviors. It examined factors affecting followers’ trust in influencers and how this influencer trust affected brand-related outcomes. The proposed model explained most of the variance in influencer trust and sufficient variance in the three brand-related outcomes. As such, this study developed a trust building model devoted to the influencer commerce context and offered support for the trust transfer theory. The findings are significant in our understanding of a contemporary, integrated media and commerce environment and the central role that a new kind of “agent” plays in the process.
Kun Xu, assistant professor of emerging media, Department of Media Production, Management and Technology
Smart speakers require smart management: How user gratifications affect privacy settings via social presence and privacy concerns.
Authors: Kun Xu and Sylvia Chan-Olmsted
Abstract: Smart speakers’ voice recognition technology has not only advanced the communication efficiency between users and machines, but also raised users’ privacy concerns. As smart speakers listen to users’ voice commands and collect audio data to improve algorithms, it is crucial to understand how users manage their privacy settings to protect their personal information. Combining the uses and gratifications approach, communication privacy management theory, and the media equation, this study recruits 1052 participants to ask about their daily smart speaker use experiences. The study explores the major gratifications that users seek, identifies two strategies that users adopt to manage their privacy, and suggests that users apply similar interpersonal privacy management rules to interactions with smart speakers.
I Like My Relational Machine Teacher: An AI Teacher’s Communication Styles and Social Presence in Online Education
Authors: Jihyun Kim, University of Central Florida, Kelly Merrill, The Ohio State University, Kun Xu, Deanna Sellnow, University of Central Florida
Abstract: New advancements in technology have made machines teachers, or technology-powered robots or AI that assist in the overall learning experience, a possibility. Though adoption rates are low at this point, colleges and universities will likely incorporate some aspects of machine teachers (e.g., AI, robots) in their curriculums in the foreseeable future. However, little is known about how to create an effective machine teacher-based education. As an initial step, the present study examines whether an AI teacher’s communication style would have any impact on students’ perceptions about an AI teacher-based education. To test this inquiry, the study conducted an online experiment using a 2 (communication styles: functional vs. relational) x 2 (course topic: natural science vs. social science) between-subjects design. Primary results indicated that students develop more favorable perceptions about an AI teacher-based education when the AI teacher is relational rather than functional. This tendency is particularly strong in a social science class. Further, social presence of an AI teacher functions as a mediator, which explains the reason why a relational AI teacher leads to more favorable perceptions about an AI teacher-based education is because of one’s social presence of an AI teacher. Collectively, the study’s findings indicate the importance of communication styles and social presence of an AI teacher.
How to Build a Credible and Socially Present Robot? A Meta-Analysis of the Power of Social Cues in Human-Robot Interaction
Authors: Mo Chen, Leping You, Miami University, Kun Xu
Abstract: Social cues have been construed as an important concept in human-robot interaction (HRI), as they can be manipulated to reflect robots’ perceived gender, personalities, emotions, identities, and so on. This study seeks to understand the distinct effects of individual social cues and uses meta-analysis to explore a hierarchy of social cues that elicit different degrees of social responses. A total of 25 and 44 effect sizes were calculated respectively to represent the magnitude of the effects of social cues on users’ social presence and trust in HRI. Results suggest that although the overall effects of social cues are small, manipulating robots’ facial and kinetic cues can induce medium-to-large sized effects on users’ social presence and trust. The results of the meta-analysis can contribute to both the theoretical refinement of the Computers are Social Actors paradigm and the practical and methodological design of human-robot interaction.