When you think of statistics, do you think of tools that help analyze the real world, or do you think of the phrase “lies, bad lies, and statistics”? Regardless of your answer to that question, Jeremy Weber’s new book I wrote Statistics for public policy, for you. In this episodehost Russ Roberts welcomes Weber to talk about it.
Weber argues that no statistics textbook includes integrating context, purpose, and audience into statistical analysis. That’s a problem. Roberts praises Weber’s use of illustrations rather than equations and explains how he thinks statistics in college is similar to cooking class. Weber thinks of them like vocational education. Both are great examples!Which is the most effective? you?
Of course, there are many more questions I’d like to ask… As always, I’ll limit myself to a few, but I’d appreciate it if you could take a moment to share your thoughts. As Russ says, we’d love to hear your thoughts.
1- Is learning statistics in school similar to taking a cooking class or learning how to use a chainsaw? What? mistaken What about these ideas about statistics? What does Weber mean when he compares concept-dependent and context-free concepts? (Perhaps think of statistics and physics.)
2- Is statistical analysis often used as a weapon in the political process or in the search for truth? How do you think politicians will receive Weber’s book and why?
3- Roberts asks to what extent we can look at data without considering theory. How does Weber describe the proper relationship between data and theory? With today’s increased computing power and data volumes, how is analysis performed? Harder• How can today’s analysis be both more data-backed and shallower at the same time?
4- The (in)famous distinction between correlation and causation is brought up towards the end of the conversation. Roberts said one of his favorite things about Weber’s book is that it makes much deeper points than that. What’s the point?how should we think about size What about causation and how do analysts use statistical significance as a cue?
5- So far, the focus is on: mistaken About statistics and how to teach them. What did we learn from this conversation about how to improve? Is a good understanding of statistics a necessary part of civics education? Why or why not?
