Studying Automated Hiring from the Margins (Fall 2021)
Automated hiring solutions have become a controversial topic at the intersection of data science and society. On the one hand, these data-driven systems promise gains in accuracy, efficiency, and speed when sorting through and ranking large numbers of applicants and applications. Companies like HireVue, Pymetrics, myInterview, and Curious Thing offer a range of AI-based tools for automating résumé screening, interviewing, and other stages of the hiring process. On the other hand, these very systems have been criticized for their potential to deepen and perpetuate existing biases or bring on new ones. Especially the opacity associated with these tools has raised demands for accountability and oversight.
Yet, while much recent work has focused on the technological design of systems and the companies who use them, much less is known about the people who are subject to their operation. How do job candidates and applicants make sense of these new tools? What challenges do they encounter in their everyday lives? What strategies and tactics do they use to address the the problem?
The goal of this project is to make sense of automated hiring systems through the eyes of those who have to live with them. Through semi-structured interviews with students who have gone through different rounds of automated hiring, we explore the lived experience of working with these systems and their operators from the margins. A better understanding of these problems will not only help us contribute to current research, but also offer insights for policy-makers, hiring managers, and developers.
Cornell IRB #2109010533 | Informed Consent Form | Research Team
Mapping Data Flows & Frictions at Cornell (Spring 2020/2021)
We all know how to navigate a campus like Cornell, moving back and forth between the different units, groups, departments, offices, and dining halls. But how does our data travel as we enter the institution? How does an organization as large and complex as Cornell manage all the different forms and streams of data—and what challenges may occur?
This project seeks to provide an in-depth understanding of what it takes to manage data flows and frictions in a large organization. Through a combination of hands-on research and reflections, we explore data practices in areas like health, policing, recruiting, and IT. Specifically, the project asks two questions:
- How, when, and by whom is student data at Cornell acquired, collected, managed, shared, and used?
- What problems, challenges, and frictions do occur along the way—and how are they resolved in practice?
In order to explore these questions, we are interviewing Cornell employees to better understand data policies and practices and how they vary across the institution. A large group of employees are being interviewed, from diverse backgrounds, fields, and levels, including personnel involved in the collection, storage, management, and use of personal data as well as the building of tools to manage such data. The goal of the research is to produce an exploratory map of the data flows and frictions at in a large organization.
Cornell IRB #2001009338 | Informed Consent form | Research Team