What questions can you answer with causal mapping?

What kind of object is a causal map and how do you answer questions with it?

The chapter after this one, Individual questions — introduction, is a catalogue: one page for each question you can put to a causal map.

A map you can query#

Coding narrative data leaves you with a table of links. A table can be interrogated. See Causal mapping produces models you can query to answer questions and The product of (causal) qualitative coding can be a model you can query.

Outputs of QDA puts causal mapping alongside other QDA approaches.

What follows from what#

A map shows that somebody said A influenced B, and somebody said B influenced C. Whether you may go on to say that A influences C is a venerable argument: The most controversial feature of causal maps is transitivity and The transitivity trap.

Beyond transitivity, Ways to causal inference surveys the routes that lead from a map to a claim about the world. We can reason about causal maps using a logic of evidence. Causal maps are knowledge graphs, but with wings connects this business to the knowledge-graph literature.

Whether to believe the answer#

Every query returns some result or ogher. Whether that answer is worth putting in a report depends on several things. Quality assurance at each step of the causal coding workflow works through the workflow stage by stage. Assessing quality or robustness of evidence for a causal link based on a bundle of coterminal causal claims narrows down to a single link and asks how strongly it is supported.

Finally, Automating the standard questions — slide creation functions shows how MapCat and the Causal Map app can help you decide which analyses you need, and run them for you.