Design and Analysis of Experiments in Networks: Reducing Bias from Interference

By Dean Eckles, Brian Karrer, Johan Ugander. 2014

336 citations recorded by the source corpus.

Abstract Estimating the effects of interventions in networks is complicated due to interference, such that the outcomes for one experimental unit may depend on the treatment assignments of other units. Familiar statistical formalism, experimental designs, and analysis methods assume the absence of this interference, and result in biased estimates of causal effects when it exists. While some assumptions can lead to unbiased estimates, these assumptions are generally unrealistic in the context of a network and often amount to assuming away the interference. In this work, we evaluate methods for designing and analyzing randomized experiments under minimal, realistic assumptions compatible with broad interference, where the aim is to reduce bias and possibly overall error in estimates of average effects of a global treatment. In design, we consider the ability to perform random assignment to treatments that is correlated in the network, such as through graph cluster randomization. In analysis, we consider incorporating information about the treatment assignment of network neighbors. We prove sufficient conditions for bias reduction through both design and analysis in the presence of potentially global interference; these conditions also give lower bounds on treatment effects. Through simulations of the entire process of experimentation in networks, we measure the performance of these methods under varied network structure and varied social behaviors, finding substantial bias reductions and, despite a bias–variance tradeoff, error reductions. These improvements are largest for networks with more clustering and data generating processes with both stronger direct effects of the treatment and stronger interactions between units.

Subfields: Education, Statistics and Probability

How researchers see this paper

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

In Education

In Statistics and Probability

Citations in the corpus

Social Data: Biases, Methodological Pitfalls, and Ethical Boundaries (2019, background): “For example, a conversation on a topic may include (re)shared content or hashtags and, thus, one user’s use of a term may have an eﬀect, sometimes called network interference , on the utility observed by others in an online community (Eckles et al., 2017; Olteanu et al., 2017b).”
Handbook of Big Data (2016, methodology): “Combined pan/zoom based on focus rectangles is described in Section 6.4.3.”
Algorithmic amplification of politics on Twitter (2021, methodology): “In this study, we chose to not employ intricate causal inference machinery that is often used to approximate causal quantities (24), as this would not guarantee unbiased estimates in the complex setting of Twitter’s home timeline algorithm.”
Almost-Matching-Exactly for Treatment Effect Estimation under Network Interference (2020, background): “Some of this work makes explicit use of observed network structure to randomly assign treatment so as to reduce interference (Ugander et al., 2013; Toulis and Kao, 2013; Eckles et al., 2016, 2017; Jagadeesan et al., 2019).”
Elements of estimation theory for causal effects in the presence of network interference (2017, background): “With the rise of interest in social networks, many authors have proposed that observed network of interactions or relationships between units may inform how causal effects can be transmitted between units [Ugander et al., 2013, Eckles et al., 2014, Athey et al., 2015].”
Opportunities and challenges in search interaction (2018, background): “5 Many of these lessons apply in domains beyond Web search.”
Causal inference with misspecified exposure mappings: separating definitions and assumptions (2021, background): “Eckles, Karrer, and Ugander (2017) discuss strategies to minimize bias introduced by violations to no-interference assumptions.”
Identification and Estimation of Spillover Effects in Randomized Experiments (2017, background): “On the other hand, Eckles, Karrer, and Ugander (2017) study the bias of the global ATE estimator under different modeling assumptions and experimental designs.”