Master of Science in Engineering, University of Toledo, 2021, Engineering (Computer Science)

The advancement in data collection has resulted in unprecedented quantity and variety of network data with diverse applications. This thesis analyzes two such network settings. The first is a computed transplant compatibility network which has a bipartite representation. We conduct local network analysis on this network and attempt link prediction on missing compatibilities. We conclude that while the techniques we develop result in modest prediction accuracy, we are able to provide an overview of the network and demonstrate the absence of some network properties we may otherwise expect. The second analysis performed is conducted on social network data, one of the most common targets for network analysis. Data compiled from these networks are perfect for analyzing social trends. One such trend that this thesis aims to address is political homophily. Evidence of political homophily is well researched and indicates that people have a strong tendency to interact with others with similar political ideologies. Additionally, as links naturally form in a social network either through recommendations or indirect interaction, new links are very likely to reinforce communities. This serves to make social media more insulated and ultimately more polarizing. We aim to address this problem by providing link recommendations that will reduce network homophily. We propose several variants of common neighbor-based link prediction algorithms that aim to recommend links to users who are similar but also would decrease homophily. We demonstrate that acceptance of these recommendations can indeed reduce the homophily of the network, whereas acceptance of link recommendations from a standard Common Neighbors algorithm does not.

Committee: Kevin Xu (Advisor); Qin Shao (Committee Member); Gursel Serpen (Committee Member) Subjects: Computer Science; Mathematics

Doctor of Philosophy, The Ohio State University, 2021, Communication

Political homophily, or the tendency for relationships or discussion ties to form more frequently between like-partisans than between people with differing political identities, is a well-studied phenomenon in the political communication and social networks literatures. Such political similarity has been observed in a variety of contexts, including in work environments, church congregations, universities, romantic relationships, neighborhoods, and online social media. Homophilous network structures have profound effects on normative democratic outcomes, such as participation and exposure to diverse sources of information. However, comparatively little attention has been paid to the antecedent processes which give rise to political homophily. This dissertation advances the concept of political selectivity, or the degree to which one's decisions about the status of ties in a political discussion network favor discussion partners who one perceives to be similar (i.e., shared party identity) to themselves. The culmination of networked theories of homophily, along with interpersonal theories of relational uncertainty and topic avoidance, together provide a holistic view of how the dyadic and network structures co-evolve over time. The goals of this dissertation are threefold: to isolate selection from other generative mechanisms, to explain individual variances in selectivity, and to provide a framework with which interpersonal processes, like topic avoidance, affect selection in dynamic discussion networks. To these ends, a five-phase, two-condition quasi-experiment was conducted in which participants shared political opinions with one another and made decisions regarding who they would like to continue discussing political matters with. Subjects (n = 366) were recruited from Amazon's Mechanical Turk participant pool into 24 cohorts. In condition 1, participants shared their opinions with each of their alters; in condition 2, participants were permitted to decide from (open full item for complete abstract)

Committee: Robert Bond (Advisor); Skyler Cranmer (Committee Member); William Eveland Jr. (Committee Member); Hillary Shulman (Committee Member) Subjects: Communication; Political Science

MS, Kent State University, 2021, College of Education, Health and Human Services / School of Foundations, Leadership and Administration

The purpose of this study was to investigate the audience's resonance experience when watching travel vlogs and its influence on their behavioral intentions. The premise of this study was when the audience's inherent needs are being fulfilled, it would reform their behavioral intention. Three theoretical concepts including the theory of resonance, U&G theory, and the involvement theory were employed in this study. Data were collected from students at Kent State University and general consumers through Amazon MTurk (MTurk). Participants were asked to complete an online survey. Multiple regression analysis was conducted, and the results revealed that audiences' resonance experiences were significantly related with their continuance intention to watch travel vlogs and travel intention of presented destinations. Also, audiences' resonance experiences were positively related to their perceived involvement, followed by continuance intention to watch and travel intention.

Committee: Kiwon Lee (Committee Chair); Seon Jeong Lee (Committee Member); Wang Philip (Committee Member) Subjects: Behavioral Psychology; Marketing

Doctor of Philosophy, University of Toledo, 2019, Higher Education

African American faculty at predominantly White institutions face a conundrum. African American students experience an achievement gap and Student Involvement Theory suggests that faculty interaction has greater impact on student achievement than any other type of involvement. These faculty may feel an obligation to serve such students yet simultaneously feel disincentivized to do so in order to satisfy tenure requirements, which typically do not prioritize service. This study sought to discover how these faculty perceive this challenge within the context of campus racial climate. Campus Racial Climate was the conceptual framework and Critical Race Theory was the theoretical framework employed in this study. Information was collected by analyzing institutional documents, interviewing faculty, and conducting member-checking to verify accuracy. Three themes regarding faculty perceptions about student-faculty interaction emerged: Faculty Experiences, Student Rapport, and Institutional Milieu. This study also arrived at four key conclusions. First, homophily (gravitation toward those who share important social characteristics) influenced perceptions of student-faculty interaction more than campus racial climate. Negative campus racial climates have been noted to motivate African American student-faculty interaction and were believed to be a primary impetus for it. However, this study revealed that the motivation for such interaction exists whether the climate is positive or negative. Homophily, was perceived as having a constant and direct influence on views about student-faculty interaction while campus racial climate was regarded as having a contextual and indirect influence on the same. Second, campus racial climate did not directly influence views about African American student-faculty interaction. However, campus environment (without respect to race) and race (without respect to campus environment) did influence perceptions. Third, the ques (open full item for complete abstract)

Committee: Penny Poplin Gosetti Ph.D. (Advisor); Shanda Gore Ed.D. (Committee Member); Kevin Jordan Ed.D. (Committee Member); Ronald Opp Ph.D. (Committee Member) Subjects: African Americans; Higher Education

Undergraduate Honors Program, Malone University, 2016, Honors Thesis

Social network analysis is becoming increasingly popular in our world of ever-expanding user data sets. Surprising relationships have been found between seemingly unrelated attributes of human behavior. At Malone University all prospective students fill out admissions applications, including a section identifying future activities in which they are interested. Potential university students can be defined as belonging to network communities with other trait-sharing students. Perhaps these students do not actually become involved in the activities, but do their choices reveal something about their future retention status? This study looks at student interests data from admissions applications of the 477 incoming freshman at Malone University in 2009. The involvement interests marked are studied along with the retention rates of those students. The goal of this study is to use the analysis of social networks to determine if there is a relationship between retention rates and the interests chosen by incoming freshman.

Committee: Kyle Calderhead Ph.D. (Advisor); David Hahn Ph.D. (Committee Member); Scott Waalkes Ph.D. (Committee Member) Subjects: Applied Mathematics; Mathematics; Personality; School Administration; Sociology; Statistics

Doctor of Philosophy, The Ohio State University, 2013, Human Ecology: Hospitality Management

As a form of electronic word-of-mouth (eWOM), online consumer reviews have attracted tremendous attention from researchers and practitioners. This research investigates how eWOM plays an important role in consumer review websites within hospitality and tourism services and how it affects the purchase decision-making process of consumers. The author examined the influence of eWOM communications in various online review websites based on the online social network framework by Brown, Boderick, and Lee (2007). Their framework postulates that individuals consider the websites to be primary "actors" and develop social relationships with the websites themselves rather than the other online participants. Drawing on the perspective of the consumer-website relationship, the current research proposes and empirically tests an integrative model of three social network constructs associated with the website (tie strength, homophily, and source credibility) and their relationship to consumers' evaluations associated with attitudes and perceived influence of eWOM effectiveness. An online survey with a sample of 793 respondents drawn from a large Midwestern university was conducted to examine predictors and consequences of eWOM communications in consumer review platforms. The results suggested that the social relationship constructs (homophily and tie strength) between a website and a consumer are important drivers of source credibility, which in turn influence attitude toward the reviews as well as the website. Moreover, the attitudes formed through the perceptions of tie-strength, homophily and source credibility ultimately determines the influence of eWOM consumers' purchase decisions. When consumers perceive higher tie strength, homophily, and source credibility associated with the reviews, they are more likely to have a favorable attitude toward the reviews as well as the review websites. An interesting finding was that attitude toward reviews does not directly influence eW (open full item for complete abstract)

Committee: Jay Kandampully (Advisor); Bilgihan Anil (Committee Member); Binkley Margaret (Committee Member); Stoel Leslie (Committee Member) Subjects: Business Administration; Communication; Marketing

Master of Applied Communication Theory and Methodology, Cleveland State University, 2008, College of Liberal Arts and Social Sciences

In the process of computer-mediated exchange, some online videos travel from one person to another resulting in the process of diffusion of the video. However, there are very few empirical investigations of the audience involved in the process. This exploratory research employs Rogers' diffusion of innovations as a theoretical framework to study online video users. Theories from social networks on tie strength and homophily are applied to create an integrated diffusion model. Based on survey data from college students, online video audience was profiled in two ways: one based on individual characteristics and another on activities with video content. Participants in the viral transmission process were found to be novelty-seekers, highly connected to others and appreciative of entertaining videos. An integrated model exploring the antecedents of viral transmission of online videos identified age, sex, Internet usage, and network connectedness as significant predictors. Contrary to previous findings, strong and homophilous ties were found to significantly contribute toward the viral spread. The findings of this study will add to the body of knowledge on diffusion research by enhancing understanding of individuals involved in an evolving medium. A profile of online video users will help marketers identify and reach the right audience.

Committee: Gary Pettey PhD (Advisor); Cheryl Bracken PhD (Committee Member); Robert Whitbred PhD (Committee Member) Subjects: Communication