Dynamics of Dyads in Social Networks: Assortative, Relational, and Proximity Mechanisms (Review)

MotivationA Typology of Mechanisms of Change in Social Networks

Conclusions

Dynamics of Dyads in Social Networks: Assortative,

Relational, and Proximity Mechanisms

J. Mejía1

1Facultad de Economía

Universidad de los Andes

May 26, 2015

Rivera, Soderstrom, Uzzi (2010) Annual Review of Sociology


Conclusions

Outline

1 Motivation

2 A Typology of Mechanisms of Change in Social Networks

Assortative MechanismsDynamics of Homophily

Dynamics of Heterophily

Relational MechanismsDynamics of Reciprocity

Dynamics of Repetition

Dynamics of Clustering and Closure

Dynamics of Degree

Proximity MechanismsDynamics of Proximity and Social Foci

3 Conclusions



Conclusions

Increasing Interest in Social Network Analysis

The capital of social network research has risen rapidly

20-year increase in the share of papers that list �networks� as akeywordIncreasing citation impact of network papers.

Conceptualizations of networks as static substrata of social

interaction are giving way.

They examine journal articles that explain how social networks

evolve over time.



Conclusions

The increasing role of network papers in sociology journals,

1970�present



Conclusions

Assortative MechanismsRelational MechanismsProximity Mechanisms

Outline

1 Motivation







Dynamics of Degree


3 Conclusions



Conclusions


Dynamics of Homophily

The greater the similarity between two individuals the more

likely they are to establish a connection

Self-similar alters mitigate the potential con�icts,misunderstandings, and monitoring costs that come withmaking connections

Homophily more strongly a�ects attachment in intimate

relations

Van Duijn et al. (2003) found that gender homophily waspositively associated with the formation of �real friends� butnot �friendly� or �neutral� relations.



Conclusions


Dynamics of Homophily

It is a rare attribute that has the strongest in�uence on

attachment

�[T]wo African Americans in a crowd of whites will tend tonotice and identify with each other because of their commonrace; however, when in a group of other African Americans,the same two people are unlikely to notice or identify witheach other�

People are more likely to maintain homophilous relationships

than they are to maintain heterophilous ones.

Felmlee et al. (1990) found that the hazard rate fordetachment of interracial romantic relationships was more thanthree times higher than that exhibited by racially homophilouscouples



Conclusions



It is also selfevident that diverse network relationships exist

Moody (2004): �in high-growth, fast changing specialties, wewould expect to see more coauthorship because it is easier tobring in a new author than it is to learn new material oneself.�People seek out others whom they believe to have valuable andcomplementary task-related skills

Heterogeneous relationships have instrumental nature

Collaborative ties between actors with complementaryattributes are often short term and oriented toward thecompletion of a project or goal



Conclusions


Multiform Heterophily

Network ties may be between individuals who are

simultaneously similar and di�erent

Casciaro & Lobo's (2008) �eld research suggests that peoplepreferentially collaborate with others who have complementaryspecializations but similar demographic traits that facilitatecommunication and trustOrganizations, too, appear to seek a balance between similarityon some dimensions, such as status (Podolny 1993), anddissimilarity on others, such as industry niche (Gulati 1995,Chung et al. 2000), strategic grouping (Nohria & Garcia-Pont1991), geographic location (Gulati & Gargiulo 1999, Chung etal. 2000), size (Shipilov et al. 2006), or embedded versusarm's-length ties (Lazerson 1995; Uzzi 1996, 1997).



Conclusions


Outline

1 Motivation







Dynamics of Degree


3 Conclusions



Conclusions


Dynamics of Reciprocity

The bestowing of a friendship from i to j tends to be quickly

followed by a reciprocal o�ering of a friendship from j to i.

While one-way friendships were signi�cantly more likely thannull relationships to blossom into reciprocated ties, they alsohad a high base rate of dissolutionThus, reciprocity is associated with stability. Over time,unreciprocated ties are inherently unstable, either becomingreciprocated or withdrawn.



Conclusions



In relationships that can be ongoing in time, an important

relational property is the frequency of repetition.

Coauthorship networks: rate of repeated ties varied from about50% to approximately 99%, which suggests that repeated tiesare common and, for some networks, are the norm.In networks of economic exchange and friendship, the ties thatexist at one point in time are likely to be repeated in thefuture.



Conclusions



A characteristic that distinguishes social networks from

biological and technological networks is clustering.

Clustering means that social networks tend to have a highdensity of closed triads, or, colloquially, people tend to becomefriends with the friends of their friends.

Rationales

1 People who spend time with a common third are likely toincidentally encounter each other, even if they are notexplicitly introduced.

2 Sharing a third party provides information about potentialconnections through referrals and gossip and that thisinformation decrease the uncertainty and risk of a newconnection



Conclusions



Clustering occurs even between individuals who are separated

by more than a single intermediary

Ties that connect people who share several common alters are

said to be structurally embedded and are not only more likely

to arise in a social network, but are also more likely to

withstand the test of time.

Clustering decreased after a shock perturbed the network.

Not all social networks exhibit clustering.

Kinship networks where ties represent genealogical descent arean obvious example of taboo norms against closure.



Conclusions


Dynamics of Degree

Most actors have only a few ties, while a small number have

extraordinarily many (Borgatti&Everett 1999, Albert&

Barab´asi 2002).

Facebook, sex...Central actors bene�t from a rich-get-richer phenomenon ofcumulative advantage

The relationship between centrality and tie accumulation is

likely to be diminished by factors that moderate an actor's

resources for tie acquisition or maintenance such as time,

money, or even natural aging.

Less extreme di�erences between the most and least centralactors than do networks of relationships where increasing one'sdegree requires only a small cost, or one-time cost



Conclusions


Dynamics of Degree

Not only are high-degree actors more likely to form

attachments than noncentral actors, but they are also more

likely to form attachments with each other

A literature is emerging on degree centrality and detachment.

Mad Men e�ect: a company's degree, as measured by thenumber of ad agencies it employed, was positively related tothe probability of breakup of the �rmagency relationship.



Conclusions


Outline

1 Motivation







Dynamics of Degree


3 Conclusions



Conclusions


Dynamics of Proximity and Social Foci

�Cupid may have wings, but apparently they are not adapted

for long �ights�

More than one in six couples �ling for a marriage license hadpreviously lived within a single city block of each other,providing early evidence for the important role of proximity onsocial attachment.Frequency and duration of 1.8 billion instant-messageconversations between 180 million people worldwide decreasedas geographic distance increased.



Conclusions


Dynamics of Proximity and Social Foci

Some individuals lost as much as one-third of their old

personal networks as they transitioned to a new location.

Coporate interlocks

Firms were more likely to share corporate board members ifthose �rms had headquarters in proximate locations.Kono et al. (1998) found that membership in exclusiveupper-class clubs in the same city as the headquarters was theactual driver of board interlocks.

Given advances in technologies that facilitate communication,

geographically dispersed social foci may be playing amore

important role in organizing attachments while geography is

playing a less important role???


Appendix

Conclusions

Researchers have historically approached the issue in three

ways: one focusing on actors' attributes, another on their

relationships, and a third on proximity.

Further research

Disambiguating causal relationships in the dynamics ofrelationship formation and dissolution.Explore interactions between assortative, relational, andproximity mechanisms.It is relatively unknown whether di�erent mechanisms playgreater or lesser roles as networks evolve.


Appendix For Further Reading

For Further Reading I

Rivera, M. Soderstrom, S. B. & Uzzi, B.

Dynamics of Dyads in Social Networks: Assortative, Relational,

and Proximity Mechanisms.

TAnnual Review of Sociology, 36, 91 - 115, 2010.


Dynamics of Dyads in Social Networks: Assortative, Relational, and Proximity Mechanisms (Review)

Documents

Transcript of Dynamics of Dyads in Social Networks: Assortative, Relational, and Proximity Mechanisms (Review)