Chapter 13 The Question of Causation

“Statistics are no substitute for judgment.”    — Henry Clay

13.1 Overview

You have tested for statistical associations and learned how to determine if a third variable moderates the strength and/or direction of a bivariate relationship. It is important to remember that your results should be considered in the context of observational data. Put simply, a significant association does not mean a causal relationship has been established. It is important not to use causal language in the presentation of your findings.

13.2 Lesson

Learn about confounding variables (i.e., those variables that explain or account for a significant association between two variables). Consider the ethical reasons behind why some experiments cannot be conducted. Compare and contrast a prospective study and retrospective study in helping determine if a causal relationship exists between two variables. Understand that lurking variables make a relationship between two variables only appear strong. See how some relationships cannot be examined by cause and effect experiments, but with enough non-experimental evidence from many different sources we can make confident causal conclusions. Click HERE to watch the video lesson.

---