"In turn, this suggests that the contributions of internet access and use on well-being were independent of the covariates we selected for, and thus might indicate causal relations. (Although we highlight the evidence here for causal claims is less than thin.)"
I think it would be more useful, especially on a forum known to have a strong opinion on a topic, to establish alternate theories that match these observations and independence relations.
I'm not convinced that this study proves what is claimed.
He is not.
It’s fine to disagree, but at least write a sentence to say why. If you can’t say why, then just take it as an interesting anecdote.
The study shows ～X
I am never wrong
Therefore, the study must be wrong.
However, I find it to be extremely unlikely that the design of this study is sound. I'm not convinced that the scientific community even knows what all of the variables that determine your level of well being are, much less having the ability to account for them in a study like this.
As I have repeatedly stated elsewhere in this thread, in regions of the world with pervasive Internet access there are many, many factors that lead to increased well being, and conversely, in regions of the world without pervasive Internet access there are many, many factors that lead to decreased well being.
This study has accounted for some of the obvious variables here, but certainly not all of them.
If you can't provide concrete debunks then your complain is nothing but hot air.
[0] - https://tmb.apaopen.org/pub/a2exdqgg#s2
[1] - https://web.archive.org/web/20220319202108/https://news.gall...
Per your links they accounted for just 6 variables, not nearly enough to make the stated conclusion with any confidence:
> In addressing our second research question, our aim was to approach the independent contributions of internet access and use on well-being. To that end, we adjusted for plausible (and available) covariates in our models that might otherwise mask or bias any independent contributions of internet access on well-being or create spurious associations. We chose six variables to represent such potentially confounding factors that have been previously considered important in the literature on well-being (): the respondent’s income (e.g., ), educational (), work (), and relationship statuses (), their ability to meet basic needs for food and shelter, and whether or not they reported having health problems ().
Really, the math isn't any harder than what's necessary to land a rocket screaming along at mach 8 or reverse a couple terabytes of voltage readings into a nice clean volumetric medical image.
This is definitely evidence, but I do share the OP's concerns around controls and would regard this finding as somewhat likely to be biased by the kinds of people who both a) have internet and b) are willing to respond to a really, really long poll.