We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.SI

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Social and Information Networks

Title: Using Ego-Clusters to Measure Network Effects at LinkedIn

Abstract: A network effect is said to take place when a new feature not only impacts the people who receive it, but also other users of the platform, like their connections or the people who follow them. This very common phenomenon violates the fundamental assumption underpinning nearly all enterprise experimentation systems, the stable unit treatment value assumption (SUTVA). When this assumption is broken, a typical experimentation platform, which relies on Bernoulli randomization for assignment and two-sample t-test for assessment of significance, will not only fail to account for the network effect, but potentially give highly biased results.
This paper outlines a simple and scalable solution to measuring network effects, using ego-network randomization, where a cluster is comprised of an "ego" (a focal individual), and her "alters" (the individuals she is immediately connected to). Our approach aims at maintaining representativity of clusters, avoiding strong modeling assumption, and significantly increasing power compared to traditional cluster-based randomization. In particular, it does not require product-specific experiment design, or high levels of investment from engineering teams, and does not require any changes to experimentation and analysis platforms, as it only requires assigning treatment an individual level. Each user either has the feature or does not, and no complex manipulation of interactions between users is needed. It focuses on measuring the one-out network effect (i.e the effect of my immediate connection's treatment on me), and gives reasonable estimates at a very low setup cost, allowing us to run such experiments dozens of times a year.
Subjects: Social and Information Networks (cs.SI); Applications (stat.AP)
Cite as: arXiv:1903.08755 [cs.SI]
  (or arXiv:1903.08755v1 [cs.SI] for this version)

Submission history

From: Guillaume Saint-Jacques [view email]
[v1] Wed, 20 Mar 2019 21:52:11 GMT (6574kb,D)

Link back to: arXiv, form interface, contact.