11

You need to distinguish between uses of propensity scores for matching of cases versus for more general adjustments. The discussion on this page suggests that there isn't much of a use case for propensity score matching. Among other problems, there is seldom much to be gained by throwing away information. Yet that is what matching cases does, with additional ...


10

Propensity score (PS) analysis has many problems in general, and matching is especially problematic. I prefer covariate adjustment for a spline function of the logit of PS if you need propensity scores, and you must also include pre-specified individual strong covariates to absorb outcome heterogeneity. If the sample size is large in relationship to the ...


9

Your assumption is that conditioning on a variable (i.e., $X_4$) blocks all paths through that variable, but that is not so. Conditioning on a variable opens a path between the antecedents of the variable. $X_1$ and $X_2$ are d-connected after conditioning on $X_4$. $X_4$ is a collider of $X_1$ and $X_2$.


5

Re-reading your question, my understanding is that you are asking if emmeans() does G-computation as part of what it ordinarily does. And based on my very limited understanding of causal models and G-computation, I would say the answer is NO. That is simply because we don't treat covariates in any special way. For a numerical covariate, the default action is ...


4

Both the cited finding that, "People who trust the media more are more knowledgeable about politics and the news" and the cited finding that, "The more people trust science, the more scientifically literate they are", are results found in observational data. It is well established that it is difficult to infer causality from ...


4

Complementary to the the answers from EdM and Frank Harrell (+1 to both). One might want to consider extensions to using Propensity Scores as the direct probability of treatment group assignment. Usually such work aim to re-weight our sample at hands such that certain features are "balanced". A prime example of that is entropy balancing, (...


4

The answer to your first question is yes. With the bidirected arc, the correlation between $Z$ and $Y$ is composed of two components, the correlation through the bidirected arc and the causal path from $Z$ to $X$ to $Y$. A simple regression of $Y$ on $Z$ gives you the whole thing, which cannot separate the two components to give you only the causal component....


3

No they do not need to be similar, if you control for that variables, as you did. That is the whole point of using control variable apart from the dummy that you are interested in.


3

It means, that changing one of those variables may, as well as may not, lead to changing the other. We can not apriori say which is how much likely. Imagine, that you can force somebody to increase the knowledge about science. If evidence is correlational it may or may not result in changing the trust. If there existed causal evidence, that knowledge causes ...


3

Evidence being "correlational" (more often called "observational") means that this was passive observational evidence of a statistical association between things, without any controlled experimentation. For example, consider the claim that "[t]he more people trust science, the more scientifically literate they are." Presumably ...


2

The distinction between direct and total effect often concerns the role of mediators. In this DAG, we can try to discern "comorbidity" as the exposure of interest, and mortality as the response or outcome. "Age" fulfills the role of confounder, as it is causal of both comorbidity and mortality. Physical functioning is a mediator because ...


2

Trust in Science -> Scientific Literacy is observationally equivalent with Scientific Literacy -> Trust in Science. Similarly, Education -> Trust in Science & Education -> Scientific Literacy is also observationally equivalent with Trust in Science -> Scientific Literacy. So when they say "even if this evidence remains correlational&...


2

Sadly, there is no closed-form solution for the ATT except in certain cases. The formula for the ATE is the combined coefficient on the A when evaluating the predictors at their means, i.e., -3 + 5*.55 - 10*.3 + 15*(.55*.3) which does equal -.775 as you have figured out. (Note that the final term should include the mean of x1*x2, which in this case happens ...


2

You used causal effects in the topic title but did not provide any background information that would make us believe that you are using a causal design. If you did not randomize the exposures, and you have no believable causal diagram, the best you can do is to estimate an association that has accounted for known factors. And you have postulated a ...


2

I'm writing a paper about this very topic, so I'll just summarize here and update with a link to the paper when it's ready. In short, the ATE, ATT, and ATC can be described as follows: The ATE is the average effect of mandating a policy of treatment for everyone vs. mandating a policy of control for everyone The ATT is the average effect of withholding ...


2

BART is a method of estimating $E[E[Y|A=1,X]] - E[E[Y|A=0,X]]$ in a highly flexible way, where $Y$ is the outcome, $A$ is the treatment, and $X$ are covariates. BART is one of many such methods that estimate the same quantity, including inverse probability weighting, propensity score methods, TMLE, causal random forests, etc. This is a totally separate ...


2

A t-test doesn't construct a certainty interval around the actual treatment difference, even the CI can't be interpreted that way. No NHST method will tell you the actual treated difference is 1000. However, both models should be powered to reject the null hypothesis that income does not differ by treatment assignment. In the self-selection scenario, age is ...


1

When you add M into the model, things can go either way: the extra predictor can make a previously non-significant X become significant or vice versa. BUT those are for really very different reasons and only one relates to mediation. Adding M can "tidy up Y" so that you can observe the X->Y relationship better -- and that can make X significant ...


Only top voted, non community-wiki answers of a minimum length are eligible