Critical phenomena in complex contagions

Why do some contagions “go viral” and others do not? Research on “small world” networks (Watts and Strogatz, 1998) shows how a very small number of long-range ties that bridge between clusters can allow contagions to spread almost as rapidly as on a random network of equal density. Recent research shows how long-range ties that accelerate the spread of information and disease can impede the spread of complex contagions—behaviors, beliefs and preferences that diffuse via contact with multiple adopters ( Centola and Macy, 2007). In confirming this result analytically and extending the analysis from small world to power law networks, we discovered that complex contagions require a critical mass of infected nodes that corresponds to a phase transition in the ability of the contagion to take advantage of the “shortcuts” created by long-range ties. We demonstrate how this critical mass is related to the dynamics of the contagion process and identify implications for modeling behaviors that spread via social influence, such as viral marketing and social movements.     

A Critical Examination of the Social Contagion Image of Collective Behavior: the Case of the Enfield Monster*

This paper makes an initial statement regarding the conceptual and empirical utility of the social contagion image as posited by Blumer and Klapp. Their position is then criticized on the basis of its assumption that unverified and unusual sensory experiences, mobilization processes, and mass preoccupations are equivalent and undifferentiated products of social contagion. Further, the social contagion approach is unable to adequately account for differential participation in these collective behavior events. Most of these problems stem from the assumed discontinuity between collective and institutional behavior embodied in the social contagion perspective. An alternative approach is posited which suggests that communication processes, availability of the population for participation, and institutional demands provide a more adequate explanation of differential participation with respect to unusual sensory experiences, mobilization, and mass preoccupation. A case study of a monster sighting in a rural community is presented and examined from this alternate perspective.     

Influence of the gregarious instinct and individuals’ behavior patterns on macro migrations: Simulation experiments

The objective of the model is to simulate the migrations of social groups over certain geographical regions, influenced by the gregarious effect and individual behavior patterns. The people move in search of better work opportunities and welfare. The model components are regions and individuals belonging to certain social groups. A region is characterized by more than 30 parameters, such as geographical data, actual population statistics, education facilities, and work opportunities. For the social groups, the parameters include the demographic and educational statistics. The simulated individuals attempt to optimize their welfare by moving over the model map. Spontaneous migration, without apparent goal, is also considered. The region parameters change owing to the number and the parameters of incoming individuals, which provides a dynamic and nonlinear feedback. The qualitative results reveal that the model never reaches a steady state and that the social groups remain in periodical movement. The simulations are agent-based. No differential equations (like in the System Dynamics approach) or any extant theory or mathematical model is used.     

How to search a social network

We address the question of how participants in a small world experiment are able to find short paths in a social network using only local information about their immediate contacts. We simulate such experiments on a network of actual email contacts within an organization as well as on a student social networking website. On the email network we find that small world search strategies using a contact’s position in physical space or in an organizational hierarchy relative to the target can effectively be used to locate most individuals. However, we find that in the online student network, where the data is incomplete and hierarchical structures are not well defined, local search strategies are less effective. We compare our findings to recent theoretical hypotheses about underlying social structure that would enable these simple search strategies to succeed and discuss the implications to social software design.     

Social networks and occupational choice: The endogenous formation of attitudes and beliefs about tax compliance

The paper analyses the emergence of group-specific attitudes and beliefs about tax compliance when individuals interact in a social network. It develops a model in which taxpayers possess a range of individual characteristics – including attitude to risk, potential for success in self-employment, and the weight attached to the social custom for honesty – and make an occupational choice based on these characteristics. Occupations differ in the possibility for evading tax. The social network determines which taxpayers are linked, and information about auditing and compliance is transmitted at meetings between linked taxpayers. Using agent-based simulations, the analysis demonstrates how attitudes and beliefs endogenously emerge that differ across sub-groups of the population. Compliance behaviour is different across occupational groups, and this is reinforced by the development of group-specific attitudes and beliefs. Taxpayers self-select into occupations according to the degree of risk aversion, the subjective probability of audit is sustained above the objective probability, and the weight attached to the social custom differs across occupations. These factors combine to lead to compliance levels that differ across occupations.     

The Game of Contacts: Estimating the Social Visibility of Groups

Estimating the sizes of hard-to-count populations is a challenging and important problem that occurs frequently in social science, public health, and public policy. This problem is particularly pressing in HIV/AIDS research because estimates of the sizes of the most at-risk populations-illicit drug users, men who have sex with men, and sex workers-are needed for designing, evaluating, and funding programs to curb the spread of the disease. A promising new approach in this area is the network scale-up method, which uses information about the personal networks of respondents to make population size estimates. However, if the target population has low social visibility, as is likely to be the case in HIV/AIDS research, scale-up estimates will be too low. In this paper we develop a game-like activity that we call the game of contacts in order to estimate the social visibility of groups, and report results from a study of heavy drug users in Curitiba, Brazil (n = 294). The game produced estimates of social visibility that were consistent with qualitative expectations but of surprising magnitude. Further, a number of checks suggest that the data are high-quality. While motivated by the specific problem of population size estimation, our method could be used by researchers more broadly and adds to long-standing efforts to combine the richness of social network analysis with the power and scale of sample surveys.     

Investing in people: The role of social networks in the diffusion of a large-scale fraud

This paper draws from social network analysis and diffusion theory to study the case of a mortgage fraud that spread undetected for five years in British Columbia, Canada. The fraud is studied from the point of view of 559 victims who unknowingly invested in the Ponzi scheme which defrauded 2285 investors for a total of $ 240 million dollars. Results show diffusion played a role in the success of the Eron fraud even though the fraud ended before it reached the final stages of a classic diffusion process. A closer look at the social structure of the Eron network revealed the elements that made the fraud successful: (1) change agents, particularly Eron principals and Eron employees invested their personal time and effort recruiting investors; (2) independent brokers actively spread the fraud to their clients; and (3) opinion leaders, investors themselves, unknowingly spread the fraud through their social networks by recruiting their friends and family to invest in Eron.     

Zur Erklärung von sozialer Ansteckung beim Übergang zur Elternschaft

Using panel data (N = 1679 married and cohabiting couples), this paper investigates the impact of social contagion processes in first births. The empirical results confirm an association between the transition rate into parenthood and the share of network members (friends, acquaintances, siblings) with young children. Several bridge assumptions concerning the explanation of social contagion are tested. It can be shown that social networks have an impact on reproductive planning in that network members provide Ego with information on the joys and challenges of parenthood. Furthermore, younger couples whose network partners mostly still live with children, place a high priority on starting a family. Beyond that, the transition to parenthood is also found to be functional for maximizing one’s social approval when the share of network members with young children rises. Older childless couples are increasingly exposed to social pressure from friends and siblings with children. This type of social pressure decreases a couple’s propensity to start a family, though.     

Influentials,novelty, and social contagion : The viral power of average friends,close communities,and old news

What is the effect of (1) popular individuals, and (2) community structures on the retransmission of socially contagious behavior? We examine a community of Twitter users over a five month period, operationalizing social contagion as ‘retweeting’, and social structure as the count of subgraphs (small patterns of ties and nodes) between users in the follower/following network.     

Social Contagion and General Diffusion Models of Adolescent Religious Transitions: A Tutorial,and EMOSA Applications

Epidemic Models of the Onset of Social Activities (EMOSA) describe behaviors that spread through social networks. Two social influence methods are represented, social contagion (one-to-one spread) and general diffusion (spread through cultural channels). Past models explain problem behaviors—smoking, drinking, sexuality, and delinquency. We provide review, and a tutorial (including examples). Following, we present new EMOSA models explaining changes in adolescent and young adult religious participation. We fit the model to 10 years of data from the 1997 U.S. National Longitudinal Survey of Youth. Innovations include a three-stage bi-directional model, Bayesian Markov Chain Monte Carlo (MCMC) estimation, graphical innovations, and empirical validation. General diffusion dominated rapid reduction in church attendance during adolescence; both diffusion and social contagion explained church attendance stability in early adulthood.     

Gender identity and lexical variation in social media

We present a study of the relationship between gender, linguistic style, and social networks, using a novel corpus of 14,000 Twitter users. Prior quantitative work on gender often treats this social variable as a female/male binary; we argue for a more nuanced approach. By clustering Twitter users, we find a natural decomposition of the dataset into various styles and topical interests. Many clusters have strong gender orientations, but their use of linguistic resources sometimes directly conflicts with the population‐level language statistics. We view these clusters as a more accurate reflection of the multifaceted nature of gendered language styles. Previous corpus‐based work has also had little to say about individuals whose linguistic styles defy population‐level gender patterns. To identify such individuals, we train a statistical classifier, and measure the classifier confidence for each individual in the dataset. Examining individuals whose language does not match the classifier's model for their gender, we find that they have social networks that include significantly fewer same‐gender social connections and that, in general, social network homophily is correlated with the use of same‐gender language markers. Pairing computational methods and social theory thus offers a new perspective on how gender emerges as individuals position themselves relative to audiences, topics, and mainstream gender norms.     

What is gained and what is left to be done when content analysis is added to network analysis in the study of a social movement: Twitter use during Gezi Park

As social movements relying on the weak ties found in social networks have spread around the world, researchers have taken several approaches to understanding how networks function in such instances as the Arab Spring. While social scientists have primarily relied on survey or content analysis methodology, network scientists have used social network analysis. This research combines content analysis with the automated techniques of network analysis to determine the roles played by those using Twitter to communicate during the Turkish Gezi Park uprising. Based on a network analysis of nearly 2.4 million tweets and a content analysis of a subset of 5126 of those tweets, we found that information sharing was by far the most common use of the tweets and retweets, while tweets that indicated leadership of the movement constituted a small percentage of the overall number of tweets. Using automated techniques, we experimented with coded variables from content analysis to compute the most discriminative tokens and to predict values for each variable using only textual information. We achieved 0.61 precision on identifying types of shared information. Our results on detecting the position of user in the protest and purpose of the tweets achieved 0.42 and 0.33 precision, respectively, illustrating the necessity of user cooperation and the shortcomings of automated techniques. Based on annotated values of user tweets, we computed similarities between users considering their information production and consumption. User similarities are used to compute clusters of individuals with similar behaviors, and we interpreted average activities for those groups.     

Mobility types,transnational ties and personal networks in four highly skilled immigrant communities in Seville (Spain)

In this study, we explore how different mobility patterns influence the composition and structure of the transnational social support networks and how personal networks allow us to elicit insightful data of mobile individuals. Ninety-five mobile individuals were selected from four distinct communities based in Seville (Spain), namely: Erasmus students, Flamenco artists, musicians from the symphonic orchestra and partners of European Commission researchers. Data were collected through an electronic survey sent by email with multiple name generators and a structured face-to-face interview utilizing a network visualization tool, VennMaker. Two distinct methods, namely qualitative case studies and cluster analysis were used to characterize mobility types. Findings reveal a heterogeneous foreign population, in which different forms of mobility are reflected in the personal networks of mobile individuals. Respondents who were settled in the city were more likely to have networks in which social support was mainly derived by hosts and people in the host location and with whom they communicated predominantly through face-to-face communication. Those who were in the host location for a study exchange, knowing that return to the county of origin is imminent were more likely to have networks linked to the home location. They relied heavily on strong transnational ties in the home country using social media to sustain their relationship. Respondents with an itinerant mobility profile, also had networks dominated by strong transnational ties, however, such ties had a higher degree of geographical spread due to previous international mobility. Participants who had a high number of hosts in the network but low connection between the ties were more likely to be linked to a specific subculture in the host society. The integration in the host location follows a different pattern to other settled individuals, mainly because their connection in the city tends to be community specific.     

Social complex contagion in music listenership: A natural experiment with 1.3 million participants

Can live music events generate complex contagion in music streaming? This paper finds evidence in the affirmative—but only for the most popular artists. We generate a novel dataset from a music tracking website to analyse the listenership history of 1.3 million users over a two-month time horizon. We show that attending a music artist’s live concert increases that artist’s listenership among the attendees of the concert by approximately 1 song per day per attendee (p-value < 0.001). Moreover, this effect is contagious and can spread to users who did not attend the event. However, whether or not contagion occurs depends on the type of artist. We only observe contagious increases in listenership for popular artists (∼0.06 more daily plays per friend of an attendee [p < 0.001]), while the effect is absent for emerging stars. The contagion effect size increases monotonically with the number of friends who have attended the live event.     

Cross-border credit networks,banking risk contagion and suppression effects

From the perspective of risk contagion, it is difficult to balance the benefits and costs of banking globalization. We use data on cross-border syndicated loans intended to build heterogeneous credit networks among banking sectors in various countries to explore the introduction of banking risks into the network. Via social network analysis (SNA), we find that the global credit network has phased characteristics, and it is a source of exposure to global banking risks. However, the more important network position, that is, the greater centrality, identified shows that the banking industry is more stable. To explain these findings, we present and estimate the banking risk contagion model and the suppression effect model of contagion. Evidence suggests that the international credit network is the channel of risk contagion, and the long-term loan network and nonfinancial borrower network have a stronger contagion effect. Exogenous banking supervision and network endogenous closeness centrality can inhibit risk contagion. From a policy perspective, these findings indicate the need to supervise high-risk cross-border loans and maintain independence in the international credit network.Data Availability StatementThe data that support the findings of this study are available from the corresponding author, [Heng Zhao, email: zhao0501@mail.dlut.edu.cn], upon reasonable request.     

Effects of Social Anxiety on Emotional Mimicry and Contagion: Feeling Negative,but Smiling Politely

Socially anxiety may be related to a different pattern of facial mimicry and contagion of others’ emotions. We report two studies in which participants with different levels of social anxiety reacted to others’ emotional displays, either shown on a computer screen (Study 1) or in an actual social interaction (Study 2). Study 1 examined facial mimicry and emotional contagion in response to displays of happiness, anger, fear, and contempt. Participants mimicked negative and positive emotions to some extent, but we found no relation between mimicry and the social anxiety level of the participants. Furthermore, socially anxious individuals were more prone to experience negative emotions and felt more irritated in response to negative emotion displays. In Study 2, we found that social anxiety was related to enhanced mimicry of smiling, but this was only the case for polite smiles and not for enjoyment smiles. These results suggest that socially anxious individuals tend to catch negative emotions from others, but suppress their expression by mimicking positive displays. This may be explained by the tendency of socially anxious individuals to avoid conflict or rejection.     

An agent-based model of network effects on tax compliance and evasion

Agent-based models are flexible analytical tools suitable for exploring and understanding complex systems such as tax compliance and evasion. The agent-based model created in this research builds upon two other agent-based models of tax evasion, the Korobow et al., 2007, Hokamp and Pickhardt, 2010 models. The model utilizes their rules for taxpayer behavior and apprehension of tax evaders in order to test the effects of network topologies in the propagation of evasive behavior. Findings include that network structures have a significant impact on the dynamics of tax compliance, demonstrating that taxpayers are more likely to declare all their income in networks with higher levels of centrality across the agents, especially when faced with large penalties proportional to their incomes. These results suggest that network structures should be chosen selectively when modeling tax compliance, as different topologies yield different results. Additionally, this research analyzed the special case of a power law distribution and found that targeting highly interconnected individuals resulted in a lower mean gross tax rate than targeting disconnected individuals, due to the penalties inflating the mean gross tax rate in the latter case.     

Estimating network structure via random sampling: Cognitive social structures and the adaptive threshold method

This paper introduces and tests a novel methodology for measuring networks. Rather than collecting data to observe a network or several networks in full, which is typically costly or impossible, we randomly sample a portion of individuals in the network and estimate the network based on the sampled individuals’ perceptions on all possible ties. We find the methodology produces accurate estimates of social structure and network level indices in five different datasets. In order to illustrate the performance of our approach we compare its results with the traditional roster and ego network methods of data collection. Across all five datasets, our methodology outperforms these standard social network data collection methods. We offer ideas on applications of our methodology, and find it especially promising in cross-network settings.     

Modelling a disease-relevant contact network of people who inject drugs

This study uses social network analysis to model a contact network of people who inject drugs (PWID) relevant for investigating the spread of an infectious disease (hepatitis C). Using snowball sample data, parameters for an exponential random graph model (ERGM) including social circuit dependence and four attributes (location, age, injecting frequency, gender) are estimated using a conditional estimation approach that respects the structure of snowball sample designs. Those network parameter estimates are then used to create a novel, model-dependent estimate of network size. Simulated PWID contact networks are created and compared with Bernoulli graphs. Location, age and injecting frequency are shown to be statistically significant attribute parameters in the ERGM. Simulated ERGM networks are shown to fit the collected data very well across a number of metrics. In comparison with Bernoulli graphs, simulated networks are shown to have longer paths and more clustering. Results from this study make possible simulation of realistic networks for investigating treatment and intervention strategies for reducing hepatitis C prevalence.