Drew Bailey on backward causal questions and forward causal inference

Following up on my paper with Guido on backward causal questions and forward causal inference, education researcher Drew Bailey writes:

(1) Some disagreements between social scientists or between social scientists and the public arise when one side is in “forward causal inference” mode and the other side is in “backward causal question” mode;

(2) Individuals or entire subfields can develop blind spots when they spend too much of their time in one of these modes and not enough in the other; and

(3) Students should get practice flexibly switching back and forth between these two modes.

His longer writeup of these ideas is here.

7 thoughts on “Drew Bailey on backward causal questions and forward causal inference

  1. Could one recover a forward-looking obtained causal effect by using backward causal questions/methods when given some data? Is there any work on this? And isn’t this difference more a matter of the information one has available at hand? Thank you :D

  2. “An advantage of spending some time in forward causal inference mode is that we can narrow the scope of the argument substantially. If you and I have different stories about why people come to be good at algebra (understanding of the equal sign, motivation, school funding, having an effective teacher), but our stories make identical predictions about what will happen to the posttest distribution of algebra test scores conditional on a long list of hypothetical interventions, maybe our stories aren’t as irreconcilable as they seem to us. Or, if our stories are substantively different, maybe it doesn’t matter as much as we thought it did. For example, if one person argues that “common environmental factors” have a large influence on adults’ test scores and another argues, “no, genes have a large influence,” but both agree that being randomly assigned to an upbringing in a somewhat more advantaged household might raise adult test scores by, say, around 1/3 of a population standard deviation on average, maybe there’s not too much to argue about after all. Or, to the extent there’s a disagreement, it’s more about rhetoric than about human development.”

    This is where I believe social science, including the most thoughtful social scientists, is ignoring important variation because the measurements are not adequate to disentangle the causal variation. The quote above dismisses the importance of equifinality with, “maybe there’s not too much to argue about after all.” Simply because two pieces land on the same square does not imply any similarity beyond that fact without a lot more information and analysis.

  3. “A reverse causal question does not in general have a well-defined answer, even in a
    setting where all possible data are made available.”

    Every example in the “paper with Guido” involves some murky question in social science or some other field that attempts to generalize from inconsistent human behavior. It would be really easy to show a well-defined answer if I get to pick the example!

    That is a minor nit, though. The real issue is that after reading the paper three times now, I am still looking for the place where the quoted statement above gets validated in some way rather than just restated as received knowledge, or shown using murky examples where nothing is remotely binary.

    In short: with many real world problems, the plausible root causes can be evaluated with results that are at least close to – if not fully – binary. If all plausible root causes are strongly refuted by binary test results except one, which is strongly supported by binary test results, the conclusion about causation is more robust than any forward causation result in the social sciences.

  4. Andrew –

    As a fan of Hill’s Criteria, I’m wondering if you see any crossover here with this (via Wikipedia):

    -snip-

    Some authors consider, also, Reversibility: If the cause is deleted then the effect should disappear as well.

    Or perhaps, “direction of causality?”

Leave a Reply to Adrian Keister Cancel reply

Your email address will not be published. Required fields are marked *