Effects of missing data in social networks

By Gregory Kossinets. 2003

924 citations recorded by the source corpus.

We perform sensitivity analyses to assess the impact of missing data on the structural properties of social networks. The social network is conceived of as being generated by a bipartite graph, in which actors are linked together via multiple interaction contexts or affiliations. We discuss three principal missing data mechanisms: network boundary specification (non-inclusion of actors or affiliations), survey non-response, and censoring by vertex degree (fixed choice design), examining their impact on the scientific collaboration network from the Los Alamos E-print Archive as well as random bipartite graphs. The simulation results show that network boundary specification and fixed choice designs can dramatically alter estimates of network-level statistics. The observed clustering and assortativity coefficients are overestimated via omission of affiliations or fixed choice thereof, and underestimated via actor non-response, which results in inflated measurement error. We also find that social networks with multiple interaction contexts may have certain interesting properties due to the presence of overlapping cliques. In particular, assortativity by degree does not necessarily improve network robustness to random omission of nodes as predicted by current theory.

Subfields: Sociology and Political Science, Statistical and Nonlinear Physics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Sociology and Political Science

In Statistical and Nonlinear Physics

Citations in the corpus

Modeling relational events via latent classes (2010, background): “accounting for censoring, network boundaries, and so on [16].”
Missing and spurious interactions and the reconstruction of complex networks (2009, background): “Simulation studies have established that these inaccuracies can lead to fundamentally wrong estimates of network properties and to misleading conclusions (8), which is particularly worrisome at a time when social network analysis is being used for finding new friends and partners, singling out key i”
Joint Link Prediction and Attribute Inference Using a Social-Attribute Network (2014, background): “The missing-link problem has important rami.cations, as missing links can alter estimates of network-level statistics [Kossinets 2006], and the ability to infer these missing links raises serious privacy concerns for social networks.”
How Does the Data Sampling Strategy Impact the Discovery of Information Diffusion in Social Media? (2010, background): “Recent work has investigated sampling of large-scale graphs, with a focus on recovering topological characteristics such as degree distribution, path length etc. (Rusmevichientong et al. 2001) as well as analyzing the impact of missing data on social network properties (Kossinets 2006).”
A Social Network Analysis of Positive and Negative Gossip in Organizational Life (2010, background): “…response rate of 75% (a response rate this high limits the possible 188 Group & Organization Management 35(2) negative effects of missing data in social network analysis and is considered to be an acceptable response rate for a whole-network approach; see Kossinets, 2006; Wasserman & Faust, 1994).”
Self-similarity of complex networks (2005, background): “I Kossinets (2006) “Effects of missing data in social networks” [18]”
Social Network Analysis in the Science of Groups: Cross-Sectional and Longitudinal Applications for Studying Intra- and Intergroup Behavior (2015, background): “Not only does the researcher lose the data of non-participating network members, but their absence—especially in the case of central network members—can dramatically affect the structure of the entire social network (Kossinets, 2006).”
Too Many Friends: Social Integration, Network Cohesion and Adolescent Depressive Symptoms (2009, methodology): “We selected the cutoff of 70 percent based on the recommendation from recent research on non-response in social networks (”
The structure of information pathways in a social communication network (2008, background): “In sociology, this is known as the boundary specification problem [17, 20], and it is inherent in essentially any study of a social network embedded in some larger world.”
The statistical physics of real-world networks (2018, background): “[101] G.”
Correcting for missing data in information cascades (2011, background): “Related to our work here are the works that study the effects of missing data on measured properties of social networks [11] and the study of biases when obtaining a graph of the Internet based on measurements [13, 1].”
Analysis of topological characteristics of huge online social networking services (2007, background): “Recently, extensive simulations are performed for several network sampling methods [10] and the effect of missing data in social network analysis is studied [2].”
Social network analysis of animal behaviour : a promising tool for the study of sociality (2008, background): “diameter), while others generate misleading artefacts (Kossinets 2006).”
Measuring Large-Scale Social Networks with High Resolution (2014, background): “Since we only record data from a finite number of participants, our study population is also a subset, and every network we analyze will be sampled in some way, see [117] for a review on sampling.”
Link Prediction in Complex Networks: A Survey (2010, background): “Kossinets and Watts [14] analyzed a large-scale social network, suggesting that two students having many mutual friends are very probable to be friends in future.”