Assessment of relationship between turnover and balance

Science is teamwork. Teams can be modeled as social networks. Numerous social science theories that take a social network analytic perspective have been developed to model, describe, explain, and predict the formation of ties among scholars [1, 2], including preferential attachment [3], triadic closure [4], and structural balance [5]. While it has been recently shown that insights about preferential attachment and triadic closure might be biased due to methodological [6] and pre-processing choices [7, 8], recent empirical work on structural balance has further confirmed the original theory [9, 10]. Structural balance basically evaluates the valence of links in triads to determine if a triad in a graph is balanced or not. Valence here often means positive versus negative relationships, where this assessment might be asymmetric between any two people who share a connection [11]. Imbalanced triads are assumed to modify one or more of their links to move towards a balanced state. Since these changes can ripple through a graph, balance is a micro-level process with potential macro-level implications for a graph. However, we have an insufficient understanding of the dynamics and patterns of turnover in triads. Therefore, in this project, we are expanding our prior research to study patterns of turnover – and by extension, stability – in social networks.

References:

[1] Easley, D. and Kleinberg, J. Networks, crowds, and markets. Cambridge Univ Press, New York, NY, USA, 2010.

[2] Newman, M. Networks: an introduction. Oxford Univ Press, Oxford, United Kingdom, 2010.

[3] Barabási, A. and Albert, R. Emergence of Scaling in Random Networks. Science, 286, 5439 (1999), 509-512.

[4] Granovetter, M. The strength of weak ties. American Journal of Sociology, 78, 6 (1973), 1360-1380.

[5] Cartwright, D. and Harary, F. Structural balance: a generalization of Heider's theory. Psychological review, 63, 5 (1956), 277-293.

[6] Clauset, A., Shalizi, C. R. and Newman, M. E. Power-law distributions in empirical data. SIAM review, 51, 4 (2009), 661-703.

[7] Kim, J. and Diesner, J. Over-time measurement of triadic closure in coauthorship networks. Social Network Analysis and Mining, 7, 9 (2017).

[8] Kim, J. and Diesner, J. Distortive effects of initial‐based name disambiguation on measurements of large‐scale coauthorship networks. Journal of the Association for Information Science and Technology (2015).

[9] Aref, S., Dinh, L., Rezapour, R. and Diesner, J. Multilevel structural evaluation of signed directed social networks based on balance theory. Scientific reports, 10, 1 (2020), 1-12.

[10] Diesner, J. and Evans, C. Little Bad Concerns: Using Sentiment Analysis to Assess Structural Balance in Communication Networks. City, 2015.

[11] Almaatouq, A., Radaelli, L., Pentland, A. and Shmueli, E. Are you your friends’ friend? Poor perception of friendship ties limits the ability to promote behavioral change. PloS one, 11, 3 (2016), e0151588.

Performance Period: July 2024 through September 2026

Grant No.: 2024-62192-SSR