▪Negative ties and signed networks

[UNDER CONSTRUCTION]

This page provide online resources to complement the chapter “Negative Ties and Signed Networks.”

Please refer to the chapter for more details on the theory. Cite: Agneessens, F. (Forthcoming) “Negative Ties and Signed Networks.” In: The Sage Handbook of Social Network Analysis. (Second Edition). Edited by John McLevey, Peter J. Carrington, and John Scott. Sage Publications

BASIC CONCEPTS

Defining negative ties

Negative ties generally refers to ties that capture “negative feelings, cognitive judgments, and/or behaviors or behavioral intentions directed toward other people” (Agneessens, Forthcoming).

  • Negative affective relations: e.g., dislike, jealousy, hate
  • Negative behavior or behavior intent includes: e.g., bullying, avoiding, providing misinformation, spreading negative gossip about the person
  • Negative cognitive ties: e.g., negative evaluations such as distrust, considering the person incompetent, unreliable.
  • Negative role relations: e.g., enemy – tend to combine the last three dimensions.

Signed networks

Negative ties are often considered the opposite of positive ties. When considering both at the same time, this is considered a signed network. There are three types of signed networks:

  • {P|N}-signed networks: signed networks where the absence of a negative tie (N) implies the presence of a positive tie (P) and vice versa.
  • {P|O|N}-signed networks: signed networks where positive and negative ties are exclusive, but a third option exist: a neutral tie (O)
  • {P+N}-signed networks; signed networks where a person can simultaniously have a positive and a negative tie

NODAL LEVEL MEASURES OF POSITION

Degree-based measures

Possible ways to generate degree-type measures for negative and signed networks:

  • For negative ties: reverse the degree value so that low become high (subtract from maximum)
  • For {P|O|N} and {P+N}-signed networks: weight the importance of positive and negative ties
  • For {P|O|N}-signed networks: take proportion of negative out of total number of positive and negative ties
  • For {P+N}-signed networks: take the proportion of positive only or overlap

Indirect connections involving negative (and positive) ties

Some possible ways to consider negative ties being important in an indirect way

  • Negative ties as “negative resources”: distance of a node to all negative ties (using positive ties)
  • The flow of (mis)information argument: whether at least one negative tie exist between ego and alter (or amount of)
  • The status argument: Beta centrality type measure
  • Attitudinal and behavioral adjustments: negative influence/contagion
  • Power-dependence argument: PII-index
  • PN-index

Structural holes for signed networks

Two versions of structural holes for signed networks:

  • being positively connected to two others who are negatively connected
  • being negatively connected to two others who are positively connected

WHOLE NETWORK LEVEL MEASURES

Cohesion

Bicliques

Structural balance

  • Measuring the level of structural balance for {P|N}-signed networks
  • Measuring the level of structural balance for {P|O|N}-signed networks
  • Weak balance for {P|O|N}-signed networks

Emergence of balance

SUBGROUPS AND BLOCKMODELING

Subgroups

  • Cliques
  • Balanced bicliques
  • Community detection

Blockmodeling

  • structural equivalence
  • blockmodeling

SOME REFERENCES

Recent overview papers:

  • Harrigan, N. M., Giuseppe (Joe) Labianca, & Agneessens, F. (2020). Negative ties and signed graphs research: Stimulating research on dissociative forces in social networks. Social Networks, 60, 1-10.
  • Offer, S. (2021). Negative social ties: prevalence and consequences. Annual Review of Sociology, 47, 177-196.
  • Labianca, G. J. (2014). Negative ties in organizational networks. In Contemporary perspectives on organizational social networks. Emerald Group Publishing Limited.
  • Yang, S. W., Trincado, F., Labianca, G. J., & Agneessens, F. (2019). Negative ties at work. In Social networks at work (pp. 49-78). Routledge.

Measures:

  • Bonacich, P., & Lloyd, P. (2004). Calculating status with negative relations. Social networks, 26(4), 331-338.
  • Everett, M. G., & Borgatti, S. P. (2014). Networks containing negative ties. Social networks, 38, 111-120.
  • Aref, S., & Wilson, M. C. (2018). Measuring partial balance in signed networks. Journal of Complex Networks, 6(4), 566-595.
  • Smith, J. M., Halgin, D. S., Kidwell-Lopez, V., Labianca, G., Brass, D. J., & Borgatti, S. P. (2014). Power in politically charged networks. Social networks, 36, 162-176.

Classic papers on balance:

  • Heider, F. (1946). Attitudes and cognitive organization. The Journal of psychology, 21(1), 107-112.
  • Cartwright, D., & Harary, F. (1956). Structural balance: a generalization of Heider’s theory. Psychological review, 63(5), 277.
  • Davis, J. A. (1967). Clustering and structural balance in graphs. Human relations, 20(2), 181-187.
  • Festinger, L., & Hutte, H. A. (1954). An experimental investigation of the effect of unstable interpersonal relations in a group. The journal of abnormal and social psychology, 49(4p1), 513.
  • Harary, F. (1959). On the measurement of structural balance. Behavioral Science, 4(4), 316-323.
  • Traag, V. A., & Bruggeman, J. (2009). Community detection in networks with positive and negative links. Physical Review E, 80(3), 036115.