Computational social sciences 4010-ONSa

Module 1: Introduction — Digital Traces and the Foundations of Computational Social Science

The module consists of three parts. The first introduces the “Big Data” paradigm in the social sciences and the key inferential limitations of working with observational, found data: representativeness (who is present online and who is missing), the data generating process for “found data,” reactivity (behavioral change in response to being measured), selection errors and coverage bias, the role of metadata, and typical “dirty data” pathologies (duplicates, drift, missing fields, ambiguous units of analysis). The second part covers core practices of responsible digital data processing: contextual privacy, the principle of data minimization, the distinction between anonymization and pseudonymization, and the intuition behind differential privacy as a technique for reducing disclosure risk.
The third part focuses on computational simulations in the social sciences (when and why they are useful) and on the distinction between predictive and prescriptive analytics in the context of policy-oriented modeling (case studies).

Module 2: Network Analysis
The module consists of two parts. The first introduces the network perspective in the analysis of social phenomena and the basic properties of social networks. It covers classic intuitions and empirical findings: the “six degrees of separation” hypothesis, small-world networks, and the problem of “searching in networks” as a graph task. Key generative models are presented: Watts–Strogatz (small-world), Barabási–Albert (preferential attachment / scale-free networks), and Kleinberg (navigability and the conditions for efficient decentralized search).
The second part discusses mechanisms and social interpretations of network structures: the strength of weak ties (Granovetter), social capital (bonding vs. bridging), assortativity, and embeddedness. It also introduces the analysis of individuals’ positions in networks through different centrality measures, as well as approaches to statistical inference in networks: null models, randomization procedures, and significance tests for network properties and node-level metrics.

Module 3: Social Processes in Networks
The first part covers basic micro-mechanisms shaping the structure and dynamics of relations: triadic closure and structural balance, followed by homophily understood as the result of social selection or social influence (the identification problem: selection vs. influence). This part also introduces simple formalizations of social processes (cellular automata and simulations) as tools for studying how local interaction rules generate macro-level patterns (e.g., clustering, segregation, structural stability).
The second part discusses processes of diffusion, mobilization, and polarization in networks. Topics include information cascades and virality, threshold models of mobilization, echo chambers, simple vs. complex contagion, and “information disorder” (mis-/dis-/mal-information) in relation to the attention economy and the design of feeds and recommender systems. The module also introduces exposure metrics that go beyond “reach” (e.g., intensity, repetition, source diversity, and the network neighborhood of exposure).

Module 4: Organisational networks - innovation, power, management. During the module, concepts of multiplex networks, structural holes, formal and informal networks, silos and boundary spanning, brokers, and bipartite networks will be presented. The classes will be supplemented with a workshop component, during which students will work with data from organisations.

Module 5: Biases and systemic inequalities. The module will consist of two parts; in the first, various biases in social research and ways to avoid them will be shown. In the second part, problems related to the technical dimensions of inequality will be discussed, such as algorithmic bias, machine fairness, the perpetuation of stereotypes in language models, and bias in the data collection process. The module includes a workshop component.

The order of the topics addressed and the details of their implementation may be subject to minor changes.

- Lazer, D., Pentland, A., Adamic, L. A., Aral, S., Barabasi, A.-L., Brewer, D., Christakis, N., Contractor, N., Fowler, J., Gutmann, M., Jebara, T., King, G., Macy, M., Roy, D., & Van Alstyne, M. (2009). Computational Social Science. Science, 323(5915), 721–723. https://doi.org/10.1126/science.1167742
- Wu, A. X., Taneja, H., Boyd, D., Donato, P., Hindman, M., Napoli, P., & Webster, J. (2020). Computational social science: On measurement. Science, 370(6521), 1174.2-1175. https://doi.org/10.1126/science.abe8308
- Conte, R., Gilbert, N., Bonelli, G., Cioffi-Revilla, C., Deffuant, G., Kertesz, J., Loreto, V., Moat, S., Nadal, J. P., Sanchez, A., Nowak, A., Flache, A., San Miguel, M., & Helbing, D. (2012). Manifesto of computational social science. The European Physical Journal Special Topics, 214(1), 325–346. https://doi.org/10.1140/epjst/e2012-01697-8
- Watts, D. J., & Strogatz, S. (1998). Collective dynamics of small-world networks. Nature, 393, 440–442.
- Newman, M. E. J., Watts, D. J., & Strogatz, S. H. (2002). Random graph models of social networks. Proc. Natl. Acad. Sci. USA, 99, 2566–2572.
- Barabási, A. L., & Albert, R. (1999). Emergence of scaling in random networks. Science, 286, 509–512.
- Newman, M. E. J. (2002). Assortative mixing in networks. Phys. Rev. Lett., 89, 208701.
- Girvan, M., & Newman, M. E. J. (2002). Community structure in social and biological networks. Proc. Natl. Acad. Sci. USA, 99, 7821–7826.
- Newman, M. E. J. (2006). Modularity and community structure in networks. Proceedings of the National Academy of Sciences, 103(23), 8577–8582. https://doi.org/10.1073/pnas.0601602103
- Ronald Burt: Structural Holes: The Social Structure of Competition – absolutny fundament teorii brokera.
- David Krackhardt: Informal networks: The company behind the chart

The student obtains credit for the classes based on:
- attendance in classes (15%),
- partial reports from the classes (20%),
- a final paper (65%).

The final grade for the subject is determined by the sum of points from the 3 components above.