In search of causality

14 Nov, 2018 at 14:41 | Posted in Statistics & Econometrics | 3 Comments


One of the few statisticians that yours truly have on the blogroll is Andrew Gelman. Although not sharing his Bayesian leanings, I find his open-minded, thought-provoking and non-dogmatic statistical thinking highly recommendable. The plaidoyer below for ‘reverse causal questioning’ is typical Gelmanian:

When statistical and econometrc methodologists write about causal inference, they generally focus on forward causal questions. We are taught to answer questions of the type “What if?”, rather than “Why?” Following the work by Rubin (1977) causal questions are typically framed in terms of manipulations: if x were changed by one unit, how much would y be expected to change? But reverse causal questions are important too … In many ways, it is the reverse causal questions that motivate the research, including experiments and observational studies, that we use to answer the forward questions …

Reverse causal reasoning is different; it involves asking questions and searching for new variables that might not yet even be in our model. We can frame reverse causal questions as model checking. It goes like this: what we see is some pattern in the world that needs an explanation. What does it mean to “need an explanation”? It means that existing explanations — the existing model of the phenomenon — does not do the job …

By formalizing reverse casual reasoning within the process of data analysis, we hope to make a step toward connecting our statistical reasoning to the ways that we naturally think and talk about causality. This is consistent with views such as Cartwright (2007) that causal inference in reality is more complex than is captured in any theory of inference … What we are really suggesting is a way of talking about reverse causal questions in a way that is complementary to, rather than outside of, the mainstream formalisms of statistics and econometrics.

In a time when scientific relativism is expanding, it is important to keep up the claim for not reducing science to a pure discursive level. We have to maintain the Enlightenment tradition of thinking of reality as principally independent of our views of it and of the main task of science as studying the structure of this reality. Perhaps the most important contribution a researcher can make is revealing what this reality that is the object of science actually looks like.

Science is made possible by the fact that there are structures that are durable and are independent of our knowledge or beliefs about them. There exists a reality beyond our theories and concepts of it. It is this independent reality that our theories in some way deal with. Contrary to positivism, I would as a critical realist argue that the main task of science is not to detect event-regularities between observed facts. Rather, that task must be conceived as identifying the underlying structures and forces that produce the observed events.


In Gelman’s essay there is  no explicit argument for abduction —  inference to the best explanation — but I would still argue that it is de facto nothing but a very strong argument for why scientific realism and inference to the best explanation are the best alternatives for explaining what is going on in the world we live in. The focus on causality, model checking, anomalies and context-dependence — although here expressed in statistical terms — is as close to abductive reasoning as we get in statistics and econometrics today.


  1. The phenomena social sciences are asked to study are NOT “independent of our knowledge or beliefs about them”.
    Like it or not, it seems to be the way it is.

  2. Another way to look at things is to recognize that experiments and the like start from some dissatisfaction with the status quo that is normally of some kind of approximation to be approved upon. The result (barring mistakes or cheating) is normally a better approximation and it is generally a pretty good result given accepted theory and the data. But sometimes what one scientist accepts another (like Einstein) may question. For example, it was widely recognized in finance that ‘1 in 100 year events’ were much more common than the theory said they should have been. Economists might have deduced that their theories were not necessarily sound, and might have gone on to induce that they were probably wrong. They would then have been motivated to consider alternatives. To me, then, the key to ‘abduction’ is not that one uses some unfamiliar alternative way of reasoning, but that one keeps an open mind about one’s ‘assumptions’ and seeks to check them from time to time. (Unfortunately, people are often bad at this, as checking assumptions implies that they may have been wrong in the past. Tough.)

  3. “it is important to keep up the claim for not reducing science to a pure discursive level. We have to maintain the Enlightenment tradition of thinking of reality as principally independent of our views of it and of the main task of science as studying the structure of this reality.”
    Isn’t this itself just a narrative?
    I go out to immerse myself in nature, to try to listen and learn the stories of the twisted rocks, the bugs, the birds, the old growth trees. Thinking of them existing independently does not mean they don’t have stories that are interesting.
    Science tries to diminish the stories. Instead of a 200-year-old tree telling about the storms it weathered, its relations, its use of language to talk through its roots with fungi, science tries to objectify the tree in terms of measurable board-feet, so loggers can make money …

Sorry, the comment form is closed at this time.

Blog at
Entries and Comments feeds.