ABSTRACT
Coefficient estimates in regression analyses come with standard errors, which indicate the degree of uncertainty in the estimates. This chapter demonstrates how to use and interpret standard errors, confidence intervals (a range of likely estimates), hypothesis tests, and p-values. There is a focus on proper interpretations of the hypothesis tests and p-values, which are often misunderstood. In particular, there is a discussion on the Bayesian critique of the p-value, which is that the p-value is not indicative of how likely the statistical relationship is real, as that likelihood requires extra information, such as the prior probability that there is a statistical relationship in the first place. And, the chapter uses research on the “hot hand” in basketball (along with an analogy on the search for aliens) to demonstrate how insignificant estimates have been misinterpreted by famous academics. The chapter ends with an emphasized point that the goal of research is not statistical significance, but rather objective and honest research.
