“10 Things You Need to Know About Causal Effects”

Macartan Humphreys pointed me to this excellent guide.

Here are the 10 items:

1. A causal claim is a statement about what didn’t happen.
2. There is a fundamental problem of causal inference.
3. You can estimate average causal effects even if you cannot observe any individual causal effects.
4. If you know that, on average, A causes B and that B causes C, this does not mean that you know that A causes C.
5. The counterfactual model is all about contribution, not attribution.
6. X can cause Y even if there is no “causal path” connecting X and Y.
7. Correlation is not causation.
8. X can cause Y even if X is not a necessary condition or a sufficient condition for Y.
9. Estimating average causal effects does not require that treatment and control groups are identical.
10. There is no causation without manipulation.

The article follows with crisp discussions of each point. My favorite is item #6, not because it’s the most important but because it brings in some real social-science thinking:

6. X can cause Y even if there is no “causal path” connecting X and Y.

Sometimes we think of causal chains of the form: A causes Z by causing B which in turn causes C which causes D and so on, where each pair of elements in the chain are connected to each other in time and space. You can gain confidence that A causes Z by seeing the effects on B, C, and D along the way. . . This way of thinking can however be very misleading. Consider an example given in [2]. Person A is planning some action Y; Person B sets out to stop them; person C intervenes and prevents person B from stopping person A. In this case Person A may complete their action, producing Y, without any knowledge that B and C even exist; in particular B and C need not be anywhere close to the action. Nevertheless, C caused Y since without C, Y would not have occurred. And this despite the fact that there is no “spatiotemporally continuous sequence of causal intermediates” between C and Y.

This sort of thing comes up a lot in political science: the threat is more powerful than the execution, and all that.

Also I have some minor comments:

Following Rubin, I would alter item 1 slightly. We can talk about causal claims for things that have not yet happened (as Rubin would say, “potential outcomes”). For example, if I consider the effect of giving you treatment A or treatment B but we haven’t picked the treatment, then I wouldn’t quite say that either “didn’t happen.” They haven’t happened yet, which is slightly different. And I’m not thrilled with item #8 which is framed a bit too deterministically for my taste. And I don’t quite agree with #10. Sure, I know what they mean here and I basically agree, but “manipulation” seems a bit strong to me. All sorts of things can happen without manipulation.