Introduction

The second half of the book has a key chapter about data and how it really matters what data you collect

Assignment 1

For the first assignment, I would like everyone to read Chapter 4: What Gets Counted Counts. As with previous chapters, one group needs to sign up to present a summary and everyone else needs to submit discussion questions.

For the related case study assignment, find a news or other online article (blogs would be ok, if they are relevant) relevant to this article (on missing data, on data not collected, etc). The article link must also be turned into canvas. Be prepared to summarize it for the class.

 

Assignment 2

For the second half of today, we will finish the book and choose from several papers.  Since we will almost all have read something different, many groups will be presenting.  The choices are (you can signup on canvas):

• Chapter 5 of Data Feminism
• Chapter 6 of Data Feminism
• Chapter 7 of Data Femininsm
• “Everyone wants to do the model work, not the data work: Data Cascades in High-Stakes AI” by Nithya Sambasivan, Shivani Kapania, Hannah Highfill, Diana Akrong, Praveen Paritosh, Lora Aroyo (PDF on canvas or here)
• “Data Cards: Purposeful and Transparent Dataset Documentation for Responsible AI” by Mahima Pushkarna, Andrew Zaldivar, Oddur Kjartansson (PDF on canvas or arXiv)
• “Mitigating dataset harms requires stewardship: Lessons from 1000 papers” by Kenny Peng, Arunesh Mathur, Arvind Narayanan (PDF on canvas or arXiv)

Given how many choices you have, be ready for a clear presentation of your chosen paper or chapter!