Chapter 2. Building Causal Diagrams From Scratch

Now that we are familiar with the elements of causal diagrams, let’s explore the iterative process to build a CD from scratch1. Because at this point we’re only dealing with observational data, it will not be a completely satisfactory process; we’ll revisit this process later in the book after having explored experiments and mediation and we’ll see how they can strengthen our understanding of causal relationships and help us draw better CDs.

As we go through the process, please keep in mind our ultimate goal of understanding what drives behaviors so that we can draw relevant and actionable conclusions for our business. Our objective is not to build a complete and precise knowledge of the whole world. Shortcuts and approximations are fair game, and everything should be assessed according to a single criterion: is this helping me with my business goal?

In addition, ...

Get Behavioral Data Analysis with R and Python now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.