Influence of measurement errors on networks: Estimating the robustness of centrality measures

Research output: Journal contributionsJournal articlesResearchpeer-review

Authors

Most network studies rely on a measured network that differs from the underlying network which is obfuscated by measurement errors. It is well known that such errors can have a severe impact on the reliability of network metrics, especially on centrality measures: a more central node in the observed network might be less central in the underlying network. Previous studies have dealt either with the general effects of measurement errors on centrality measures or with the treatment of erroneous network data. In this paper, we propose a method for estimating the impact of measurement errors on the reliability of a centrality measure, given the measured network and assumptions about the type and intensity of the measurement error. This method allows researchers to estimate the robustness of a centrality measure in a specific network and can, therefore, be used as a basis for decision-making. In our experiments, we apply this method to random graphs and real-world networks. We observe that our estimation is, in the vast majority of cases, a good approximation for the robustness of centrality measures. Beyond this, we propose a heuristic to decide whether the estimation procedure should be used. We analyze, for certain networks, why the eigenvector centrality is less robust than, among others, the pagerank. Finally, we give recommendations on how our findings can be applied to future network studies.

Original languageEnglish
JournalNetwork Science
Volume7
Issue number2
Pages (from-to)180-195
Number of pages16
ISSN2050-1242
DOIs
Publication statusPublished - 01.06.2019

    Research areas

  • Business informatics - centrality measures, measurement error, missing data, robustness

Documents

DOI

Recently viewed

Publications

  1. How can problems be turned into something good? The role of entrepreneurial learning and error mastery orientation
  2. The effect of psychotherapy for depression on improvements in social functioning
  3. Numerical Investigation of the Effect of Rolling on the Localized Stress and Strain Induction for Wire + Arc Additive Manufactured Structures
  4. Making an impression with open strategy
  5. Glitch(ing)! A refusal and gateway to more caring techno-urban worlds?
  6. Set-Oriented and Finite-Element Study of Coherent Behavior in Rayleigh-Bénard Convection
  7. New developments in extrusion of profiles with variable curvatures and cross-sections
  8. The role of tree crown on the performance of trees at individual and community levels
  9. One planet
  10. Data quality assessment framework for critical raw materials. The case of cobalt
  11. The geometry of habitat fragmentation
  12. How to Measure the Speed of Enterprise IT?
  13. Alignment of the life cycle initiative’s “principles for the application of life cycle sustainability assessment” with the LCSA practice
  14. Determinants and Consequences of Executive Compensation-Related Shareholder Activism and Say-on-Pay Votes
  15. Irish English and Variational Pragmatics
  16. Alcohol intake can reduce gambling behavior
  17. Rapid upwards spread of non-native plants in mountains across continents
  18. In vivo degradation of binary magnesium alloys - A long-term study
  19. Jenseits des Elfenbeinturms
  20. How Individuals React Emotionally to Others’ (Mis)Fortunes
  21. BBS futur 2.0
  22. Building a digital anchor
  23. The Lawless Vitality of Sculpture, c. 1960-80. Formalism’s Monsters, Cybernetic Breakdowns and the Joys of Deviation
  24. Too Scared to Fight Back? Affective Job Insecurity as a Boundary Condition Between Workplace Incivility and Negative Mood States in Temporary Agency Workers
  25. LivingCare - An autonomously learning, human centered home automation system
  26. Long-term effects of historical heathland farming on soil properties of forest ecosystems
  27. Ein Jahr geht noch