EUSci Podcast Extra - Sean Roberts on spurious correlation
Having teased you with an excerpt in podcast 66, here is the full version of our interview with Sean Roberts.
Correlation is the statistical relationship involving dependence of one variable to the other. Scientist observe and use correlations every day to generate hypotheses and interpret raw data. But apart from the well know mantra "correlation doesn't imply causation", the explanatory power of correlational studies is limited and we should be aware of those limitations. In this special episode of the podcast our producer Chris is talking to Sean Roberts from the University of Edinburgh about "spurious" correlations: how almost any variable can be correlated to any other. To prove that point Chris challenges Sean to find a correlation between two seemingly random variables: the basic word order of the language people use, and the number of children they have.