[estearum]

Why don't you read the study and raise the specific confounders you think are missing?

The answer is universally "no", there's no such thing as "controlling for [all] confounding variables" without blinded randomization. But we unfortunately have to rely on observational studies to try to understand the world. At least to understand what specific questions can justify the cost and complexity of a randomized study.

So yeah, of course there are confounders. Some are controlled in this study and others are not. Which are the important ones you think are missing?

[rahimnathwani]

Sure, a drive-by "did they control for confounders???" without naming any is low-effort, and it's reasonable to ask GP to be specific.

BUT you overshot in the other direction! "There's no such thing as controlling for confounders without blinded randomization" isn't fair. Sure, randomization is the cleanest identification strategy, but it's not the only one. Instrumental variables, regression discontinuity, difference-in-differences, sibling or twin fixed effects, and natural experiments are all used in economics, and they exist because randomization is often impossible or unethical, and you have to rely on natural experiments.

(Separately, a phone-ownership study of 9yo kids has an obvious potential for selection bias. So in a sense I'm sympathetic to GP asking about confounders without naming any.

[estearum]

If you actually copy/pasted the "quoted" text from my comment, you will have found I said something different from what you claimed.

Virtually any study has an obvious potential for selection bias. That's why people write up how they did the study and then other people can interrogate and criticize it.

Might as well just paste a link to https://en.wikipedia.org/wiki/Bias_(statistics)#Types in every comment section ever.