Course Title: Data Practices (in the 21st century)
- Pamela Rivière (pdriviere at ucsd dot com)
- Sean Trott (sttrott at ucsd dot com)
Faculty sponsor: Gary Cottrell
Scientific practice generally relies upon some kind of data. “Data” are collected through various means, transformed and pre-processed (e.g. “cleaned”), analyzed (e.g. through statistical inference, qualitative analysis, etc.), interpreted (usually with respect to some theoretical framework or paradigm), and sometimes even made publicly available.
Questions must be confronted and considered along every step of this pipeline. Some of these questions concern the practice of science as an epistemological tool: How ought our data-collection practices be designed to license the appropriate inferences, what kind of data “counts” as evidence, and how should this data be analyzed? Other questions concern the ethics of data collection and publication: how should researchers balance the recent push for data transparency with the importance of privacy, and who is entitled to claim “ownership” of data? What kinds of infrastructure––legal, technological, etc.––can help support appropriate scientific practice?
In this course, we will hear from speakers from a variety of disciplines, all of whom have confronted these questions in some form or another:
- Week 1: Tal Golan (History)
- Week 2: Kamala Visweswaran and Burgundy Fletcher (Ethnic Studies)
- Week 3: Julian McAuley (CSE)
- Week 4: Brad Voytek (Cog Sci)
- Week 5: Ilkay Altintas (San Diego Supercomputer Center)
- Week 6: Saiba Varma (Anthropology)
- Week 7: Molly Roberts (Political Science)
- Week 8: Kip Kantelo and Pegah Parsi (Health Sciences, Privacy)
- Week 9: Andrew Gelman (Columbia University: Statistics and Political Science)
- Week 10: Eric Baković (Linguistics)
This course is intended to be taken Pass/No Pass.
The main requirements for the course include:
- Reading the required readings each week.
- Writing and turning in 2 questions by noon on Fridays each week, on the topic of that week’s reading / talk.
- Attend all talks and discussions.
- Participate in group discussions.
- A paper (anywhere from 5-10 pages) either: 1) presenting an opinion (backed by evidence and argumentation) regarding one of the central questions tackled in the course; or 2) relating one or more of the themes of the course to the treatment of data in your particular field.