A pinboard by
Jacqueline Ng

PhD candidate in Industrial Engineering & Management Sciences, Northwestern University


The role of social media in both enhancing and inhibiting the efficacy of organizational work teams.

Today’s digital world has shaken the foundation of organizational structure, shifting the traditional functional hierarchical structure of organizations to one that is implemented around networks of teams. According to a Deloitte 2016 Human Capital Trends report, the number one issue on leaders’ minds is redesigning the way we work in teams to enable these networks of teams to coordinate their activities, share information and work together effectively. Meanwhile, the rapid proliferation of social media technologies is offering unprecedented opportunities for people to post, reach, and transmit information with diverse groups of people in a visible manner that persists over time. These advances in digital tools offer new ways for people to communicate and collaborate within and across teams that can facilitate ease of coordination, information sharing and teamwork. Yet at the same time, this persistent visibility and awareness of content and activity may lead to attention allocation problems in the workplace. Due to the potential for information overload, team members may develop strategies to manage their hyper awareness and selective unawareness to social media. Thus, an important question that one might ask is how should team members allocate their attention so that they are able to balance their abundant awareness with selective unawareness?

The availability of server-side and digital trace data of people’s interactions and communication with each other provide a novel way to address this important question. In my current research, I leverage a unique data set from a multi-city online real estate company that uses Slack, a widely used team-based communication platform, to organize their internal communications within the firm. Using a multi-method approach that involves social network analysis, survey methods, and regression, I first examine how team members strategically regulate information flow through the use of different heuristics. Then, I examine the relationship between each of these strategies and productivity, to determine how team members can optimally configure their memberships in team networks to promote more effective teamwork and performance outcomes.

Ultimately, my goal is to seek an improved understanding of how organizations are structuring their work around networks of teams, and how team members can take advantage of the widespread availability of social media technologies to accomplish their work effectively.


Causal Inference Under Network Interference: A Framework for Experiments on Social Networks

Abstract: No man is an island, as individuals interact and influence one another daily in our society. When social influence takes place in experiments on a population of interconnected individuals, the treatment on a unit may affect the outcomes of other units, a phenomenon known as interference. This thesis develops a causal framework and inference methodology for experiments where interference takes place on a network of influence (i.e. network interference). In this framework, the network potential outcomes serve as the key quantity and flexible building blocks for causal estimands that represent a variety of primary, peer, and total treatment effects. These causal estimands are estimated via principled Bayesian imputation of missing outcomes. The theory on the unconfoundedness assumptions leading to simplified imputation highlights the importance of including relevant network covariates in the potential outcome model. Additionally, experimental designs that result in balanced covariates and sizes across treatment exposure groups further improve the causal estimate, especially by mitigating potential outcome model mis-specification. The true potential outcome model is not typically known in real-world experiments, so the best practice is to account for interference and confounding network covariates through both balanced designs and model-based imputation. A full factorial simulated experiment is formulated to demonstrate this principle by comparing performance across different randomization schemes during the design phase and estimators during the analysis phase, under varying network topology and true potential outcome models. Overall, this thesis asserts that interference is not just a nuisance for analysis but rather an opportunity for quantifying and leveraging peer effects in real-world experiments.

Pub.: 28 Aug '17, Pinned: 31 Aug '17