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’d drag my embarrassing mountain of overdue books back without late fees. Data Science needs a space where we can share our past mistakes without consequences and ask questions that we might be embarrassed about, without judgement. 

Once a week Heather Krause picks a story, question or comment to address and posts a video discussion of the topic online. You can check them out each Wednesday on our youtube channel and add your comments and thoughts!

Submit stories and questions in advance via our anonymous link (below) or by emailing Heather. We go through data science concepts step by step, fostering real understanding rather than nervously pretending you know everything about everything! No question is too stupid. Heather examines data mess-ups from her own work and from the anonymous submissions so we can have a good laugh and learn together.

The most useful questions and outrageous stories from each session 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.


– Heather Krause

These are the best questions, comments, and mistakes from the Data Amnesty live sessions. They made it in either by being very valuable learning opportunities for us all, or by being very hilarious. Sometimes both.
What’s the difference between a mean and a median and how do I ever actually know which one to use?

I’ve always told everyone that the difference between qualitative and quantitative data boils down to the number of respondents. Right?

Accidentally switched the codes for gender mid-way through the project!

We gathered 10 rounds of individual’s data without using a unique ID. Oops.

I sent a survey to 4200 people. Only eleven people answered. What can I do?