Recommended for researchers equipped with computational skills who are prepared to apply computational methods to datasets and compete. Begins Wednesday June 22 at 1:30PM and continues through Thursday June 23rd. Presentations will be made on June 24th and winners announced after Friday’s final keynote address at 5:20pm in the OLC at the Jacobs Center.

The IC2S2 Datathon is a working session in which participants collaborate to turn datasets into insight. Data will be provided to teams who will then develop research questions and [preliminary] findings. Individuals need not have a team identified prior to registering.

Presentations will be made before judges who will award prizes to the winners. The theme of the datathon this year is: "U.S. Election 2016: An Outsider’s Cycle?" Our judges include Adriana Crespo-Tenorio, Lead Researcher at Facebook, Rayid Ghani, Director, Center for Data Science & Public Policy at the University of Chicago and Kate Grossman, Director of Fellowships, Institute of Politics, University of Chicago.

For IC2S2 attendees, we offer two options for experiential opportunities in advance of the General Session. Skills workshops on Thursday, June 23 and a Datathon which runs Wednesday afternoon June 22 – Thursday evening June 23.

For social science researchers and data analyst enthusiasts who are new to computational methods, we offer skills workshops on topics such as performing online experiments, interactive visualizations on the web, and causal inference online systems.

For researchers equipped with computational skills who are interested in applying computational methods to datasets, we offer the IC2S2 Datathon.

Please be sure to indicate your interest in participating in either the optional Skills Workshops or the Datathon during the registration process. You may sign up for either of these pre-sessions not both.

Skills Workshops

Recommended for social science researchers and data analytics enthusiasts who are new to computational methods. Thursday, June 23

Session 1: Online Experiments with Volunteer Science and Virtual Labs

Online Experiments with Volunteer Science | David Lazer and Jason Radford | 10am-12pm

Volunteer Science was built to be an online laboratory, offering facilities for researchers to create online experiments and for subjects to participate in these experiments. In this session, we provide an overview of our vision for the online laboratory and the battery of experiments we replicated to validate our approach. We then walk participants through the process of starting a research team and creating, testing, and running experiments with volunteers and workers on Mechanical Turk.

Online Experiments with Virtual Labs | Andrew Mao | 1pm-3pm

For decades, physical behavioral labs have been the primary method for conducting controlled experiments of human behavior. However, the Internet now enables software-based "virtual labs" using online participants, extending the scope of experimental methods and allowing for studies of increasing complexity, size, and scope. Open-source, software-based experiments also allow for faster iteration and improved reproducibility of experimental research.

We will discuss the design and features of an open-source platform for virtual lab experiments aimed at making it easier to build and conduct large-scale, real-time experiments for studying collective behavior. We will review several different studies conducted successfully so far, including a complex collective intelligence task and an experiment on cooperation over a month of real time. We will show how the platform uses Meteor, a real-time web framework, to allow for simple programming of simultaneous interaction, live monitoring of participants, and digital one-way mirrors for the experimenter.

Although our platform addresses many of the logistical challenges in studying groups online, we will also discuss various additional issues to consider in designing such experiments. We highlight how open-source software speeds iteration and facilitates reproducibility by allowing researchers to share not only data, but entire experiment protocols. We will conclude with a brainstorming session on future directions for virtual lab research and the types of novel behavioral questions that it enables (computational) social scientists to answer.

Session 2: Simultaneous Workshops on Causal Inference and Visualization

3:30-5:30PM Attendees select which workshop to attend

Causal inference in online systems: Methods, pitfalls and best practices | Amit Sharma

From recommending what to buy, which movies to watch, to selecting the news to read, people to follow and jobs to apply for, online systems have become an important part of our daily lives. A natural question to ask is how these socio-technical systems impact our behavior. However, because of the intricate interplay between the outputs of these systems and people's actions, identifying their impact on people's behavior is non-trivial.

Fortunately, there is a rich body of work on causal inference that we can build on. In the first part of the tutorial, I will show the value of counterfactual reasoning for studying socio-technical systems, by demonstrating how predictive modeling based on correlations can be counterproductive. Then, we will discuss different approaches to causal inference, including randomized experiments, natural experiments such as instrumental variables and regression discontinuities, and observational methods such as stratification and matching. Throughout, we will try to make connections with graphical models, machine learning and past work in the social sciences.

The second half will be more hands-on. We will work through a practical example of estimating the causal impact of a recommender system, starting from simple to more complex methods. The goal of the practical exercise will be to appreciate the pitfalls in different approaches to causal reasoning and take away best practices for doing causal inference with messy, real-world data.

Those attending this workshop should have R, the R library dplyr, and Git installed on their laptops and should clone this repository:

Interactive Visualization on the Web with D3.js | Alessandro Febretti, Sr.

Web-based visualization is gaining popularity thanks to its wide availability and ease of use. In this workshop you will discover the basics of D3.js, a Javascript library that helps create powerful visualizations on the web. D3 lets you combine data from multiple sources and gives you full control on the final look of your data. We will go through the steps needed to set up a basic development environment, create a simple visualization and make it accessible to other people online. All you need to follow along are your laptop with your favorite browser and text editor installed.