Statistical Tests for Contagion in Observational Social Network Studies

Greg Ver Steeg, Aram Galstyan
Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics, PMLR 31:563-571, 2013.

Abstract

Current tests for contagion in social network studies are vulnerable to the confounding effects of latent homophily (i.e., ties form preferentially between individuals with similar hidden traits). We demonstrate a general method to lower bound the strength of causal effects in observational social network studies, even in the presence of arbitrary, unobserved individual traits. Our tests require no parametric assumptions and each test is associated with an algebraic proof. We demonstrate the effectiveness of our approach by correctly deducing the causal effects for examples previously shown to expose defects in existing methodology. Finally, we discuss preliminary results on data taken from the Framingham Heart Study.

Cite this Paper


BibTeX
@InProceedings{pmlr-v31-steeg13a, title = {Statistical Tests for Contagion in Observational Social Network Studies}, author = {Ver Steeg, Greg and Galstyan, Aram}, booktitle = {Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics}, pages = {563--571}, year = {2013}, editor = {Carvalho, Carlos M. and Ravikumar, Pradeep}, volume = {31}, series = {Proceedings of Machine Learning Research}, address = {Scottsdale, Arizona, USA}, month = {29 Apr--01 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v31/steeg13a.pdf}, url = {https://proceedings.mlr.press/v31/steeg13a.html}, abstract = {Current tests for contagion in social network studies are vulnerable to the confounding effects of latent homophily (i.e., ties form preferentially between individuals with similar hidden traits). We demonstrate a general method to lower bound the strength of causal effects in observational social network studies, even in the presence of arbitrary, unobserved individual traits. Our tests require no parametric assumptions and each test is associated with an algebraic proof. We demonstrate the effectiveness of our approach by correctly deducing the causal effects for examples previously shown to expose defects in existing methodology. Finally, we discuss preliminary results on data taken from the Framingham Heart Study.} }
Endnote
%0 Conference Paper %T Statistical Tests for Contagion in Observational Social Network Studies %A Greg Ver Steeg %A Aram Galstyan %B Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2013 %E Carlos M. Carvalho %E Pradeep Ravikumar %F pmlr-v31-steeg13a %I PMLR %P 563--571 %U https://proceedings.mlr.press/v31/steeg13a.html %V 31 %X Current tests for contagion in social network studies are vulnerable to the confounding effects of latent homophily (i.e., ties form preferentially between individuals with similar hidden traits). We demonstrate a general method to lower bound the strength of causal effects in observational social network studies, even in the presence of arbitrary, unobserved individual traits. Our tests require no parametric assumptions and each test is associated with an algebraic proof. We demonstrate the effectiveness of our approach by correctly deducing the causal effects for examples previously shown to expose defects in existing methodology. Finally, we discuss preliminary results on data taken from the Framingham Heart Study.
RIS
TY - CPAPER TI - Statistical Tests for Contagion in Observational Social Network Studies AU - Greg Ver Steeg AU - Aram Galstyan BT - Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics DA - 2013/04/29 ED - Carlos M. Carvalho ED - Pradeep Ravikumar ID - pmlr-v31-steeg13a PB - PMLR DP - Proceedings of Machine Learning Research VL - 31 SP - 563 EP - 571 L1 - http://proceedings.mlr.press/v31/steeg13a.pdf UR - https://proceedings.mlr.press/v31/steeg13a.html AB - Current tests for contagion in social network studies are vulnerable to the confounding effects of latent homophily (i.e., ties form preferentially between individuals with similar hidden traits). We demonstrate a general method to lower bound the strength of causal effects in observational social network studies, even in the presence of arbitrary, unobserved individual traits. Our tests require no parametric assumptions and each test is associated with an algebraic proof. We demonstrate the effectiveness of our approach by correctly deducing the causal effects for examples previously shown to expose defects in existing methodology. Finally, we discuss preliminary results on data taken from the Framingham Heart Study. ER -
APA
Ver Steeg, G. & Galstyan, A.. (2013). Statistical Tests for Contagion in Observational Social Network Studies. Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 31:563-571 Available from https://proceedings.mlr.press/v31/steeg13a.html.

Related Material