Single Time Point Interventions in Network-Dependent Data

  • Oleg Sofrygin
  • Elizabeth L. Ogburn
  • Mark J. van der Laan
Chapter
Part of the Springer Series in Statistics book series (SSS)

Abstract

Consider a study in which we collect data on N units connected by a social network. For each unit i = 1, , N, we record baseline covariate (W i ), exposure (A i ), and outcome of interest (Y i ) information. We also observe the set F i that consists of the units in {1, , N}∖{i} that are connected to and could influence the unit i. The set F i constitutes “i’s friends”. We allow | F i |, the total number of friends of i, to vary in i. In addition, we assume that | F i | goes to zero when scaled by 1∕N. For example, F i could represent the set of all the friends of i in a social network, or the set of all of i’s sexual partners in a study of the effects of early HIV treatment initiation.

References

  1. S. Aral, D. Walker, Identifying social influence in networks using randomized experiments. IEEE Intell. Syst. 26(5), 91–96 (2011)CrossRefGoogle Scholar
  2. S. Aral, D. Walker, Tie strength, embeddedness, and social influence: a large-scale networked experiment. Manag. Sci. 60(6), 1352–1370 (2014)CrossRefGoogle Scholar
  3. P.M. Aronow, C. Samii, Estimating average causal effects under interference between units. ArXiv e-prints, May (2013)MATHGoogle Scholar
  4. A.-L. Barabási, R. Albert, Emergence of scaling in random networks. Science 286(5439), 509–512 (1999)MathSciNetCrossRefMATHGoogle Scholar
  5. G.W. Basse, E.M. Airoldi, Optimal design of experiments in the presence of network-correlated outcomes. ArXiv e-prints, July (2015)Google Scholar
  6. J. Bowers, M.M. Fredrickson, C. Panagopoulos, Reasoning about interference between units: a general framework. Polit. Anal. 21(1), 97–124 (2013)CrossRefGoogle Scholar
  7. D.S. Choi, Estimation of monotone treatment effects in network experiments. ArXiv e-prints, August (2014)Google Scholar
  8. N.A. Christakis, J.H. Fowler, The spread of obesity in a large social network over 32 years. N. Engl. J. Med. 357(4), 370–379 (2007)CrossRefGoogle Scholar
  9. N.A. Christakis, J.H. Fowler, Social contagion theory: examining dynamic social networks and human behavior. Stat. Med. 32(4), 556–577 (2013)MathSciNetCrossRefGoogle Scholar
  10. A.P. Dawid, V. Didelez, Identifying the consequences of dynamic treatment strategies: a decision-theoretic overview. Stat. Surv. 4, 184–231 (2010)MathSciNetCrossRefMATHGoogle Scholar
  11. L. Liu, M.G. Hudgens, Large sample randomization inference of causal effects in the presence of interference. J. Am. Stat. Assoc. 109(505), 288–301 (2014). ISSN 0162-1459Google Scholar
  12. R. Lyons, The spread of evidence-poor medicine via flawed social-network analysis. Stat. Politics Policy 2(1) 1–26 (2010)CrossRefGoogle Scholar
  13. E.L. Ogburn, T.J. VanderWeele, Vaccines, contagion, and social networks. ArXiv e-prints, March (2014)MATHGoogle Scholar
  14. E.L. Ogburn, O. Sofrygin, M.J. van der Laan, I. Diaz, Causal inference for social network data with contagion. ArXiv e-prints, October (2017)Google Scholar
  15. J.M. Robins, Addendum to: “A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect”. Comput. Math. Appl. 14(9–12), 923–945 (1987)MathSciNetCrossRefMATHGoogle Scholar
  16. P.R. Rosenbaum, Interference Between Units in Randomized Experiments. J. Am. Stat. Assoc. 102(477), 191–200 (2007)MathSciNetCrossRefMATHGoogle Scholar
  17. O. Sofrygin, M.J. van der Laan, R. Neugebauer, Simcausal R package: conducting transparent and reproducible simulation studies of causal effect estimation with complex longitudinal data. J. Stat. Softw. 81, 2 (2017)CrossRefGoogle Scholar
  18. O. Sofrygin, M.J. van der Laan, tmlenet: targeted maximum likelihood estimation for network data (2015)Google Scholar
  19. O. Sofrygin, M.J. van der Laan, Semi-parametric estimation and inference for the mean outcome of the single time-point intervention in a causally connected population. J. Causal Inference 5(1), 20160003 (2017)Google Scholar
  20. C. Steglich, T.A.B. Snijders, M. Pearson, Dynamic networks and behavior: separating selection from influence. Sociol. Methodol. 40(1), 329–393 (2010)CrossRefGoogle Scholar
  21. E.J. Tchetgen Tchetgen, T.J. VanderWeele. On causal inference in the presence of interference. Stat. Methods Med. Res. 21(1), 55–75 (2012)MathSciNetCrossRefGoogle Scholar
  22. P. Toulis, E. Kao, Estimation of causal peer influence effects, in Proceedings of The 30th International Conference on Machine Learning (2013), pp. 1489–1497Google Scholar
  23. M.J. van der Laan, Causal inference for a population of causally connected units. J. Causal Inference 2(1), 13–74 (2014a)Google Scholar
  24. T.J VanderWeele, Sensitivity analysis for contagion effects in social networks. Sociol. Methods Res. 40(2), 240–255 (2011)Google Scholar
  25. T.J. VanderWeele, Inference for influence over multiple degrees of separation on a social network. Stat. Med. 32(4), 591–596 (2013)MathSciNetCrossRefGoogle Scholar
  26. T.J. VanderWeele, W. An, Social networks and causal inference, in Handbook of Causal Analysis for Social Research (Springer, Berlin, 2013), pp. 353–374CrossRefGoogle Scholar
  27. T.J. VanderWeele, E.L. Ogburn, E.J. Tchetgen Tchetgen, Why and when “flawed” social network analyses still yield valid tests of no contagion. Stat. Polit. Policy 3(1), 2151–2160 (2012c)Google Scholar
  28. T.J. VanderWeele, E.J. Tchetgen Tchetgen, M.E. Halloran, Interference and sensitivity analysis. Stat. Sci. 29(4), 687–706 (2014b)Google Scholar
  29. D. Walker, L. Muchnik, Design of randomized experiments in networks. Proc. IEEE 102(12), 1940–1951 (2014)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  • Oleg Sofrygin
    • 1
  • Elizabeth L. Ogburn
    • 2
  • Mark J. van der Laan
    • 3
  1. 1.Division of BiostatisticsUniversity of California, BerkeleyBerkeleyUSA
  2. 2.Department of BiostatisticsJohns Hopkins Bloomberg School of Public HealthBaltimoreUSA
  3. 3.Division of Biostatistics and Department of StatisticsUniversity of California, BerkeleyBerkeleyUSA

Personalised recommendations