Thread by @ginasue

Our 2017 paper Critique and Contribute: A Practice-Based Framework for Improving Critical Data Studies and Data Science is now one of the most cited in Big Data. Open Access here: liebertpub.com/doi/full/10.10…
With Anissa Tanweer @BrittaFiore @lilaavati

We show how people working with data grapple with pragmatic ethical choices in their day-to-day work. We do this by looking at Data Science for Social Good teams, academic data science teams, and energy engineering teams.

We look at four common critiques of big data
data are inherently interpretive
data are inextricable from context
data are mediated through the tools that produce them
data become a chance to negotiate values.

We see these versions of these critiques voiced by the data scientists and model makers whom we study.

The implication is we can't just teach data scientists an ethics course and be done with it. That's not what the data science teams that we studied need.

What they need is help with the process of identifying and clarifying the best practices of making data and doing data science as they reflect on their work. And they need help in designing new kinds of ethical institutions to support this kind of work.

At least among the values-driven teams that we studied these data science and engineering teams already want to be ethical. Their problem was figuring how to make it so.

<I can't resist @lilaavati>

We call for two practical things going forward to, um, make it so:
1)  Create new opportunities for data science practitioners and data science scholars to engage in sensemaking together

(2)  Support new kinds of organizational arrangements to foster a culture of ethical data science practice

For those of us working for fair and accountable data systems, we have to think seriously about work and organizations if we want to think about the ethics of large scale data.

We believe that genuine collaboration between people who are working in data science and people who understand social theory can push for more ethical, and better, ways of knowing in increasingly data-saturated societies.

We hope you like our paper! Let us know what you think. liebertpub.com/doi/full/10.10…