confounder
A factor that influences both a putative cause and its effect, which must be adjusted for to obtain a valid causal estimate (e.g., a hypothesized gene affecting both smoking and cancer).
As defined in 196 | Judea Pearl on Cause and Effect
Where it comes up · 3
- 164 - Cause Effect3 claims
- Anupam Bapu Jena on Random Acts of Medicine2 claims
- 196 | Judea Pearl on Cause and Effect2 claims
Most important claims · 7 of 7Search all claims for “confounder” →
- Bringing knowledge in the form of a causal diagram lets you do more than predict the effect of forced action—it tells you which factors you must adjust for (such as gender, family history, or addiction in the family) in order to obtain a causal answer without running an experiment, effectively replacing a randomized experiment with a piece of knowledge.196 | Judea Pearl on Cause and Effect
- The statistician Ronald Fisher, himself an avid pipe smoker, used his statistical expertise to argue the smoking-cancer correlation might be due to a hidden genetic confounder causing both nicotine craving and cancer risk; his resistance prolonged the debate for over a decade, during which many millions died, until the Surgeon General resolved it by showing the required confounder effect was implausibly large.164 - Cause Effect
- The 1960s smoking-cancer controversy could not be settled by a randomized experiment (which would be unethical), and Fisher argued a hidden gene could make people both crave nicotine and develop cancer; the debate was resolved by bringing knowledge of plausibility to bear—calculating that such a gene would have to make carriers eight times more likely to smoke, a strength judged implausible.196 | Judea Pearl on Cause and Effect
- No purely statistical argument can establish causation no matter how much correlational data you have; to reach a causal conclusion you must supply some causal assumption or an experiment ('no causes in, no causes out'), a principle that has been insufficiently appreciated.164 - Cause Effect
- Temporal order is the usual but not sufficient signature of causal influence: the rooster's crow reliably precedes and correlates with the sunrise yet does not cause it, showing that 'B follows A in time' cannot by itself establish that A causes B.164 - Cause Effect
- Statistical significance (the result is unlikely due to chance given sample size) must be distinguished from economic or clinical significance (the magnitude is large enough to matter in practice); a result can be statistically significant yet trivial, or so large that it signals a data error or an unobserved confounder rather than a true effect.Anupam Bapu Jena on Random Acts of Medicine
- To credibly claim a natural experiment, one must first demonstrate that the exposed and unexposed groups are similar on all observable characteristics—demographics, chronic conditions, procedures performed, staffing levels—because any imbalance suggests the exposure was not truly as-good-as-random, and even then a valid experiment establishes the effect but not the mechanism.Anupam Bapu Jena on Random Acts of Medicine