When you penalise wrong answers, you build in a sense of shame and failure to being wrong that most people never get over. It leads to cheating, to covering up of mistakes, and to avoiding doing things where being wrong is a possibility. How about, instead, we make it the default that you assume that you will be wrong in numerous ways. We make it a fundamental part of the process to figure out those ways, and even reward the finding of those mistakes. In doing so, we give people the freedom to explore, to try new things, and, above all, to learn without fear.
Author: Dr Linda McIver
Dr Linda McIver pioneered authentic Data Science and Computational Science education with real impact for secondary students and founded the Australian Data Science Education Institute in 2018. Author of Raising Heretics: Teaching Kids to Change the World, Linda is an inspiring keynote speaker who has appeared on the ABC’s panel program Q&A, and regularly delivers engaging Professional Development for Primary, Secondary, and Tertiary Educators across all disciplines. A passionate educator, researcher and advocate for STEM, equity and inclusion, with a PhD in Computer Science Education and extensive teaching experience, Linda’s mission is to ensure that all Australian students have the opportunity to learn STEM and Data Science skills in the context of projects that empower them to solve problems and make a positive difference to the world.
Lies, Damned Lies, and AI
In which I rant about tech companies marketing chatbots that are not fit for purpose. People keep telling me this tech can only improve, so I gave it the benefit of the doubt, and threw it one of the tests that often causes me grief in my attempts to dine out or at people's houses. Is this product gluten free?
When simple values are complex
How do you measure temperature? It's more complicated than you expect.
Axes of awful
This is another example of how there are no absolute rules in data science (except for: there's no such thing as a perfect dataset - that one holds inviolable!). Everything is context. The y-axis not starting at 0 is sometimes ok. Pie charts are sometimes a great way to compare values. A line graph is sometimes useful for discrete data.
What is wrong with this data analysis?
So someone is using quantitative data to justify something. How can you figure out whether the analysis is valid, or whether there are holes in it you could drive a truck through? It's not always easy. Sometimes it's not even possible, without access to the raw data! But here are some starter questions you can ask about the data analysis, to help figure out where the issues are.