Keep your data inclusivity without diluting your results.

It’s important to include nuanced and comprehensive options on surveys. You don’t want your question about race/ethnic background to be ‘White or Other’. You don’t want your preferred language question to say ‘English or Other’. Trying to be inclusive while gathering data on sexual orientation might mean including 10 or 15 specific sexual orientations. The desire to be inclusive is admirable, but in data science there’s a challenging equity trade-off. 
If you have so many categories, you may have such a small number of each type of respondent that after analysis you can’t say anything statistically meaningful about them. Or worse, only the majority category has enough data to count, forcing you to combine smaller categories, or ignore them altogether.

How can we negotiate the dilemma between trust-building inclusivity, meaningful analysis results, and giving weight to every response?

Get the details here.

Data Amnesty Salon is February 11th
Got questions about data that you're too embarrassed to ask? Find yourself wondering what a word you're seeing a lot in the data reports you're getting actually means? We've got you.

The Data Amnesty Salon is back.
We'll be live on Tuesday, February 11th 11am Eastern Time.

This month we have a much-improved registration process. (Still free, of course, but now you'll get reminders!) 

Submit an anonymous question here.

When I was a kid, one of the most important dates circled on my calendar was Amnesty Day at my local library. On these days, I could return my mountain of overdue books and have the late fees forgiven. Data Science needs a place to offer the same opportunity to learn from our mistakes without penalties or judgement. 

The We All Count Data Amnesty Salons are a place to share anonymous stories about what went wrong in your data projects, alert people to mistakes you’ve made in the past, and ask questions you might be embarrassed about. They are a judgement-free zone where we can all improve together.

Hosted by Heather Krause, the Salons are live-streamed and you can submit stories and questions in advance via our anonymous link (above) or by emailing Heather; or you can join the live discussion on the day! Unless your mistake is, like, career-endingly bad, in which case we DEFINITELY want to hear it, but I’d submit it anonymously. The best (and by best we mean worst) data gaff of each salon goes in the Data Amnesty Hall of Fame – a place to celebrate the difficulty and frequent absurdity of the world of Data Science. It’s part of our commitment to Demystify, Democratize and Demonstrate Data for Equity.

Great Resource Alert: 
The coin model of privilege and critical allyship: implications for health.
I was in Vancouver last week and met many amazing people working on equity and ethics issues in data. One of them was Kim Van Der Woerd, the Principal at Reciprocal Consulting. She pointed us in the direction of this article which included many conceptual and practical steps on embedding equity into data and research. More on this to come!
Feminist Data Book Club - Author Q&A
The Feminist Data Book Club has officially launched. We've got a line up of three books based on your feedback and in super exciting news, the author of this month's book, Dr. Catherine Harnois is answering questions live next Friday February 7th at 11am.

You can join the club and read along here. 

If you have any trouble joining, just send Heather an email.

Book 1: Feminist Measures in Survey Research
Book 2: White Logic, White Methods
Book 3: The brand new book Data Feminism

We're looking for equity problems and successes.

If you have a story or idea you want to share, send me a note by replying to this email.

Project for Equity in Data Science
Copyright © 2020 Datassist, All rights reserved.

Did someone forward this email to you? Sign up here to get the next one directly to your inbox.

Want to change how you receive these emails?
You can
update your preferences or unsubscribe from this list.