Yongnam Kim I don't think it makes sense to think of grade causing gender or vice versa. Assuming there's a cohort effect, we can think of 'year born' as causing both grade and gender. In which case controlling for gender would block the non-causal path through 'year born' and 'gender'

Methodological pluralism is commendable. But platitudinous pluralism (the knee-jerk response that no single approach can do everything) doesn't really help anything. This piece fails to illuminate what DAGs are and what they are used for. doi.org/10.1093/ije/dy…

In a survey, the 2nd top reason of some social scientists’ pessimism about their disciplines was “too much causal inference.” Hard to believe it but anyway, it sounds like #causalinference seems to bother many social scientists, which is a good signal.

Yongnam Kim However, note that "causal inference" is lumped together with "experiments". I'd be very interested in those numbers. Personally, I see more good reasons for being skeptical about experiments than causal inference (for which I see mainly bad reasons) in the Social Sciences.

“Causal effects are not identified without random assignment” (Rosenbaum, 2020). How about DiD & front-door adjustment? Do we have more?

What in Hell is this? I'm dumbfounded by how a policy could be both so malicious and so idiotic.

"Race, COVID Mortality, and Simpson's Paradox." Dana Mackenzie, my co-author in #Bookofwhy, has discovered another case of Simpson's Paradox in COVID data, and has posted an interesting analysis of its implications here: ucla.in/3gy17m8

These definitions may help to resolve some confusion b/t #DIF (association) and item bias (causation) in the psych measurement literature. Interpretation of spurious differential item functioning: psyarxiv.com/qyzwx

Typo correction. <Can two variables have the same values but different impacts? Revisiting Card & Krueger (1994) and Lord’s (1967) paradox> A new perspective & a new answer on Lord’s paradox

I saw earlier posts on DiD, so I thought some of you might be interested in my recent work in causal inference! Our paper illustrates how DiD can be used, albeit positivity violation: osf.io/preprints/psya… Go check it out if interested! #DiD #causalInference #lordsParadox

DAGs can be used to understand the mechanics of linear interaction analysis. Easy to see the zero correlation between the first-order and interaction terms after centering (though this is not the reason for centering). See more here: bpspsychub.onlinelibrary.wiley.com/doi/10.1111/bm…

Why does centering X1 only (not X2) change the coef on X2​ while leaving the coefs on X1 and the (centered) interaction term unchanged? For those who enjoy DAGs, here is a slightly modified visual explanation we provided (tinyurl.com/32b5kbuz)