Databases Reference
In-Depth Information
the OK Cupid platform within a week of Frank's original email. Note
we'd be giving a “1” to women who ignore his emails but for some
separate reason send him an email with their number. It would also be
hard to check that the number isn't fake.
The nodes in a causal graph are labeled by these sets of confounders,
treatment, and outcome, and the directed edges, or arrows, indicate
causality. In other words, the node the arrow is coming out of in some
way directly affects the node the arrow is going into.
In our case we have
Figure 11-5
.
Figure 11-5. Causal graph with one treatment, one confounder, and
one outcome
In the case of the OK Cupid example, the causal graph is the simplest
possible causal graph: one treatment, one confounder, and one out‐
come. But they can get much more complicated.
Definition: The Causal Effect
Let's say we have a population of 100 people that takes some drug, and
we screen them for cancer. Say 30 of them get cancer, which gives them
a cancer rate of 0.30. We want to ask the question, did the drug cause
the cancer?
To answer that, we'd have to know what would've happened if they
hadn't taken the drug. Let's play God and stipulate that, had they not
taken the drug, we would have seen 20 get cancer, so a rate of 0.20. We
typically measure the increased risk of cancer as the difference of these
two numbers, and we call it the
causal effect
. So in this case, we'd say
the causal effect is 10%.
