ABSTRACT
So far in this book we’ve mainly used effect size measures that rely on interval scaling-including means, Cohen’s d, and correlations. In this chapter we take a step back to require only nominal scaling and focus on the proportion, which is a highly useful ES measure based on frequencies, or simple counts. We’ll consider research questions for which a proportion provides the answer and, as we’ve done so many times with other ES measures, we’ll focus on the CI as well as the point estimate. Then we’ll discuss research questions where we need the difference between two proportions and the CI on that difference. My first example investigates the possibility of telepathy, which is communication with another person using a psychic power, through “the power of the mind”.
