McCormick / NICO / Kellogg 2018: 4th Annual International Conference on Computational Social Science

Hosted by the Kellogg School of Management, Northwestern University, Evanston, IL USA

For IC2S2 attendees, we offer two options for experiential opportunities in advance of the General Session on Thursday July 12th: a series of Skills Workshop or a Datathon.

For social science researchers and data analyst enthusiasts who are new to computational methods or want to add new tools to their toolkit, we offer skills workshops on computational psychometrics, new advances in social networks analysis, visual communication, and more.

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. Space is limited.

Skills Workshops

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

Workshop participants, please download the workshop descriptions and pre-work instructions.

Session 1a) New Advances in Network Analysis | Roger Guimerà

After decades developing methods to model and characterize social networks, we are now in a position to use rigorous probabilistic approaches to make predictions from network data. For example, using network approaches we can anticipate conflict between team members, or predict whether individuals will cooperate or defect when facing a social dilemma. In this session, we will discuss the fundamentals and learn some Python tools for network inference.

Session 1b) Computational Psychometrics for Educational and Psychological Assessments | Alina von Davier, ACTNext by ACT and Pietro Cipresso, Instituto Auxologico Italiano

In this workshop, the instructors will use lecture, discussions, and software demos to introduce a new area, Computational Psychometrics (CP; Cipresso, 2015; von Davier, 2015), and the best assessment practices for data logging, data mining (DM), visualization, and machine learning (ML) techniques, as well as methods for evaluating results from the analyses of Big Data, even using virtual reality (VR). The session is designed for researchers with a background in measurement but less experience with data mining or machine learning.

Session 2a) Now they see it: Visual Communication of the Patterns in your Data | Steven Franconeri, Northwestern University

Within a well-designed graph or data visualization, the eyes can be a powerful tool for understanding patterns in data. But within a poorly-designed depiction of the same data, the same tasks can be inefficient, or even overwhelming. In this workshop Psychology Professor Steven Franconeri will combine an overview of data visualization techniques with hands-on exercises to illustrate how to clearly present your data to both your research colleagues and to non-technical audiences.

Session 2b) Using Web of Science Citation Data for “science of science” Studies on the Global Research Network | Joe Brightbill, Technology Lead, Custom Data

The Web of Science Core Collection is a vast citation network representing the global landscape of science since 1900. For decades, researchers have explored this rich dataset to answer key questions about the nature of scientific discovery and innovation and to empower big data analytics. The Web of Science Core Collection contains metadata on the entire scholarly publishing landscape including over 20,000 peer reviewed journals, 180,000 conferences, and 80,000 books across all sciences, social sciences, and arts & humanities. It contains over a billion cited references, forming a giant network of interconnections between scientists, their institutions, and their disciplines. The data represent science globally, enabling the exploration of changing R&D patterns as emerging economies grow their research footprint and collaboration trends evolve. Big data techniques have opened the analytical possibilities for citation data beyond the traditional bibliometric studies into network analyses that combine citation data with datasets from other disciplines. Citation data are used by researchers in information science, computer science, business, economics, public policy, and network studies to answer questions related to scientific funding, innovation, scientific workforce, team science, etc. We will discuss the characteristics of Web of Science Core Collection data, how it is structured, and common types of analysis conducted. A sample of the raw data will be provided in advance. Participants are encouraged to explore the data in advance and bring questions to the workshop. The instructor will run through common use cases using the data.

Session 3a) Creating Interactive Visualizations using R Shiny | Christina Maimone, Northwestern University Information Technology

Shiny is an R package that lets you build interactive web applications that can stand alone on a web page, function as a dashboard, or be incorporated into R Markdown documents. You can share your data and research in engaging ways or create tools to help students explore statistical concepts and data analysis. In this workshop we’ll explore what you can do with Shiny, and you’ll create your first Shiny application. Workshop attendees should be familiar with R and bring their laptop with R, RStudio, and packages Shiny and tidyverse installed and up to date.

Session 3b) Building multi-level agent-based models with NetLogo and LevelSpace | Arthur Hjorth, Nortwestern University and Bryan Head, Northwestern University

In this hands-on workshop, participants will learn to build multi-level agent-based models in NetLogo with the recently developed LevelSpace Extension. The workshop is led by members of the Center for Connected Learning and Computer-Based Modeling, and directed by Prof. Uri Wilensky, the inventor of NetLogo. We will run the workshop to accommodate a wide range of experience levels. No programming or modeling experience is required.

Multi-level Agent-Based Modeling (ML-ABM) enables modelers to easily expand on models by connecting them to other models. Typical use cases for ML-ABM include modeling interactions between levels by delegating each representational level to each own model, e.g. representing an organization as a collection of individual departments, each represented by a model; zooming in on particular event by designing higher spatial or temporal granularity event-specific models, e.g. a model for simulating a shipping accident embedded inside a larger time-scale logistics model; or connecting different types of models like agent based models and systems dynamics models.

LevelSpace is a recently developed ML-ABM extension to NetLogo, one of the most widely used ABM environments. In this workshop, participants will learn about ML-ABM, and use NetLogo and LevelSpace to build and/or expand on models. Participants are encouraged to bring their own models if they have them, but we will provide interesting models for participants as well.

Datathon

Recommended for researchers equipped with computational skills who are prepared to apply computational methods to datasets and compete.  Teams will make presentations at the conference and a prize will be awarded to the winning team.

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 first place team will be awarded $1500. The theme of the datathon this year is "Culture in the Age of Intersectionality: How are categories used to include and exclude?” Our judges include Aymar Jean "AJ" Christian, Northwestern University, Taha Yasseri, University of Oxford, and Taylor Brown, Duke University.

Datathon participants: please download the datathon description and instructions.