Data Literacy

In an increasingly data-centric world, a working understanding of data analytics and quantitative methods is essential, for all members of society. When presented with claims in the media that are accompanied by statistics, diagrams, and outputs from technologies like artificial intelligence and machine learning, how can we learn to separate useful information from pseudoscience? In other words, how can we learn to not be fooled by statistics? The aim of this course is to improve students' data literacy, through a largely non-technical introduction to some of the foundational concepts in statistical thinking. The course will teach students from all backgrounds how to interpret and critically appraise claims made by machine learning and quantitative data science methods, and understand both the possibilities and pitfalls of these emerging sciences. It assumes no technical background and is taught largely through case studies of applications of data science outside of academia. The course teaches some fundamental quantitative methods for dealing with and interpreting data, as well as visualisation techniques using simple spreadsheets. Topics include: how to translate mathematical jargon into understandable language; measuring and talking about uncertainty using probability; how to easily make clear charts and data visualisations; demystifying fundamental statistical ideas (correlation versus causation, distinguishing between significant and important results); explaining and predicting with statistical models; the ethics of data science.