Matthew E. Brashears

Social Isolation in America: Changes in Core Discussion Networks over Two Decades

Have the core discussion networks of Americans changed in the past two decades? In 1985, the General Social Survey (GSS) collected the first nationally representative data on the confidants with whom Americans discuss important matters. In the 2004 GSS the authors replicated those questions to assess social change in core network structures. Discussion networks are smaller in 2004 than in 1985. The number of people saying there is no one with whom they discuss important matters nearly tripled. The mean network size decreases by about a third (one confidant), from 2.94 in 1985 to 2.08 in 2004. The modal respondent now reports having no confidant; the modal respondent in 1985 had three confidants. Both kin and non-kin confidants were lost in the past two decades, but the greater decrease of non-kin ties leads to more confidant networks centered on spouses and parents, with fewer contacts through voluntary associations and neighborhoods. Most people have densely interconnected confidants similar to them. Some changes reflect the changing demographics of the U.S. population. Educational heterogeneity of social ties has decreased, racial heterogeneity has increased. The data may overestimate the number of social isolates, but these shrinking networks reflect an important social change in America

Models and Marginals: Using Survey Evidence to Study Social Networks

Fischer (2009) argues that our estimates of confidant network size in the 2004 General Social Survey (GSS), and therefore the trend in confidant network size from 1985 to 2004, are implausible because they are (1) inconsistent with other data and (2) contain internal anomalies that call the data into question. In this note, we assess the evidence for a decrease in confidant network size from 1985 to 2004 in the GSS data. We conclude that any plausible modeling of the data shows a decided trend downward in confidant network size from 1985 to 2004. The features that Fischer calls anomalies are exactly the characteristics described by our models (Table 5) in the original article.

Small networks and high isolation? A reexamination of American discussion networks

Abstract Recent findings from the General Social Survey suggest that discussion networks have shrunk, and social isolation has increased, since 1985. These data have been threatened by possible artifacts, however, and may be unreliable. This paper reports on a nationally representative experiment that estimates discussion network size and evaluates several proposed explanations for the shrinking networks finding. It finds that modern discussion networks have decreased in size, but that social isolation has not become more prevalent. It also disconfirms several explanations for shrinking networks, and validates a method for increasing the number of names provided in response to name generators.

Humans use Compression Heuristics to Improve the Recall of Social Networks

The ability of primates, including humans, to maintain large social networks appears to depend on the ratio of the neocortex to the rest of the brain. However, observed human network size frequently exceeds predictions based on this ratio (e.g., “Dunbar’s Number”), implying that human networks are too large to be cognitively managed. Here I show that humans adaptively use compression heuristics to allow larger amounts of social information to be stored in the same brain volume. I find that human adults can remember larger numbers of relationships in greater detail when a network exhibits triadic closure and kin labels than when it does not. These findings help to explain how humans manage large and complex social networks with finite cognitive resources and suggest that many of the unusual properties of human social networks are rooted in the strategies necessary to cope with cognitive limitations.

Anomia and the sacred canopy: Testing a network theory

This article evaluates the Durkheim/Berger argument that integration in a network of co-religionists protects against anomia. The 1985 General Social Survey network instrument is used to evaluate the effect of integration on anomia and the probability of unhappiness. Results indicate that contact with religiously homogeneous others paired with personal religious belief reduces anomia and the likelihood of unhappiness. Additionally, while ego/alter closeness is important, alter/alter closeness is not. These results suggest that individuals benefit from religious association more so than religious community. Additional analyses indicate that these results are unlikely to be due to homophily.

The microstructures of network recall: How social networks are encoded and represented in human memory☆

Abstract A growing number of studies indicate that aspects of psychology and cognition influence network structure, but much remains to be learned about how network information is stored and retrieved from memory. Are networks recalled as dyads, as triads, or more generally as sub-groups? We employ an experimental design coupled with exponential random graph models to address this issue. We find that respondents flexibly encode social information as triads or groups, depending on the network, but not as dyads. This supports prior research showing that networks are stored using “compression heuristics”, but also provides evidence of cognitive flexibility in the process of encoding relational information.

Sex and network recall accuracy

Abstract How does an individuals sex influence their recall of social relations? Extensive research has shown that social networks differ by sex and has attempted to explain these differences either through structural availability or individual preferences. Addressing the limitations of these explanations, we build on an increasing body of research emphasizing the role of cognition in the formation and maintenance of networks to argue that males and females may exhibit different strategies for encoding and recalling social information in memory. Further, because activating sex roles can alter cognitive performance, we propose that differences in recall may only or primarily appear when respondents are made aware of their sex. We explore differences in male and female network memory using a laboratory experiment asking respondents to memorize and recall a novel social network after receiving either a sex prime or a control prime. We find that sex significantly impacts social network recall, however being made aware of ones sex does not. Our results provide evidence that differences in male and female networks may be partly due to sex-based differences in network cognition.

Error correction mechanisms in social networks can reduce accuracy and encourage innovation

Abstract Humans make mistakes but diffusion through social networks is typically modeled as though they do not. We find in an experiment that high entropy message formats (text messaging pidgin) are more prone to error than lower entropy formats (standard English). We also find that efforts to correct mistakes are effective, but generate more mutant forms of the contagion than would result from a lack of correction. This indicates that the ability of messages to cross “small-world” human social networks may be overestimated and that failed error corrections create new versions of a contagion that diffuse in competition with the original.

In the organization’s shadow: How individual behavior is shaped by organizational leakage

Individuals who join an organization often adopt its characteristic behaviors, but does the same effect extend to nearby nonmembers, and is this process impeded or enhanced by the competition between organizations? This article argues that organizations influence the behavior of both members and proximate nonmembers in a process we term “organizational leakage” and that competition between organizations moderates the impact of any one of them on individual behavior. This article finds, using the Add Health data, that an individual’s location in an organizational ecology is an important predictor of his or her behavior, even while controlling for other factors, including membership.

Gangs, clubs, and alcohol: The effect of organizational membership on adolescent drinking behavior.

How does adolescent organizational membership in general, and simultaneous membership in distinct types of organizations in particular, impact drinking behavior? While past studies have focused either on the learning effect of involvement with gangs or on the constraining influence of conventional organizations on adolescent problem behavior, we explore the possibility that conventional school clubs can serve as socializing opportunities for existing gang members to engage in drinking behavior with non-gang club members. Using the Add Health data, we show that gang members drink more often, and engage in more binge drinking, than non-members. More importantly, individuals who are members of both gangs and school clubs drink alcohol at greater levels than those who are solely involved in gangs. In addition, non-gang adolescents who are co-members with gang members in the same school club are more likely to drink alcohol than non-members. This result has important implications for understanding the role of organizations in adolescent behavior and suggests that the study of delinquent behaviors would benefit from devoting more attention to individuals who bridge distinct types of organizations.