sodiummuffin
No bio...
User ID: 420
sodiummuffin
12d ago
On the other hand, people who get more sun live years longer.
Observational studies without clear mechanisms of action are almost completely worthless. Reading the study it's just the classic thing where they controlled for a handful of factors that they thought of and then declared whatever was left over the effect of the thing they're studying. Rather than it being any of the countless variations across the population that aren't included in the arbitrary list of controls. (I think Scott has a post somewhere where he mentioned how little he trusts studies like this.) For instance this is how they controlled for comorbidity:
As a measure of comorbid illness at the start of the study, we created a dummy variable termed ‘comorbidity’ to identify women who had been treated with antidiabetic or anticoagulant drugs or medication for CVD for more than 1 month.
I wonder if any illnesses not prescribed those drugs might both increase mortality and decrease sunbathing? Or general variation in health below the level of actual illness?
This is in contrast to the skin cancer risk where the mechanism of action is very straightforward. It seems like a serious failure of both science and science communication that this sort of largely-meaningless observational study gets put on the same level.
sodiummuffin
16d ago
Not all randomized control trials are blinded randomized control trials. All you need for a randomized control trial is to randomly assign a group of patients that gets the treatment and a group that doesn't. As far as I know, no long-term randomized control study of gender transition has ever been conducted, in either children or adults.
Non-RCT's are if anything even worse than euphemisms like "moderate-quality" make them seem, reading something like Scott's ivermectin post might help give a sense for it. That's why fields like nutrition, where long-term randomized control trials are impractical, are so terrible despite far more quantity and quality of research than a small field like gender dysphoria.
As a result of the GRADE approach, we read things like this in the report:
There was one high quality study, 25 moderate quality studies and 24 low quality studies. The low quality studies were excluded from the synthesis of results.
No, it's way worse than that, the high/moderate/low quality ratings were based on the cited meta-study and seem if anything too lenient. Reading the meta-study, many of the studies only looked at physical outcomes like "is puberty suppressed", they made no attempt to measure psychological outcomes to determine whether suppressing puberty actually provided any benefit. This is the supposed single "high-quality" study. It isn't a randomized control study, it compares patients who have been given puberty blockers to ones who just started the assessment process. (It also compares to a "cisgender comparison group", such comparisons tend to be even more worthless.) Among other potential problems, this means the results are very plausibly just regression to the mean or benefits from the other mental-health care provided. If you think the parents of children with worse self-reported "internalizing, suicidality, and peer relations" are more likely to seek treatment than the parents of children who are currently doing fine, which the study itself shows, then improvement over time is the expected result even if you don't do anything. And then here are the detailed explanations of why they considered the other studies to be even worse.
sodiummuffin
22d ago
It's only unchangeable if we allow the combination of the eugenics taboo and a lack of long-term planning to deny us the use of current technologies like embryo-selection and potential future technologies like polygenetic genetic-engineering. (To say nothing of the possible individual enhancements opened up if we manage to achieve brain emulation.) Remember the general population of whites and asians is also less intelligent than intellectuals tend to assume, because they associate with a highly selected subset. A large fraction of the population struggles with tasks like "interpreting a simple bar graph". It's also getting worse, with current dysgenic trends. Just achieving and maintaining the sort of humanity that many people already assume exists requires transhumanism, for every race.
Afterwards we can look back on the statistics about stuff like intelligence and crime and obesity and depression the same way we currently look back on 50% infant mortality rates and widespread stunting from malnutrition. Though of course the biggest leap would be curing aging, if we ever achieve that I expect a lot of the other improvements would seem like a sideshow by comparison.
sodiummuffin
24d ago
You can argue that consistently using "anti-X" to refer to any restriction on X, even if the restriction is the lack of a special privilege and is something the speaker thinks is justified, would be a more objective way to use language. But it is not the standard way to use language, guesswho isn't out there talking about people arrested for dangerous driving as being "arrested under an anti-white law", so it seems understandable for Folamh3 to interpret guesswho as making a bolder and less semantic claim.
I don't think it would really be a better way to use language either, because it's so impractical to do consistently that nobody would do it. Nobody is going to use it for every hypothetical special privilege that could exist, at best it would be influenced by status-quo bias based on what laws already exist, and realistically personal bias would creep in immediately. It would just create a natural motte and bailey where people would use "anti-X" in some cases based on their biases, and then retreat to "it's a restriction on X so it's anti-X" when challenged.
sodiummuffin
24d ago
I think you're missing the point of her analogy. A law that restricts trans behavior is an "anti-lgbt law" regardless of the truth value of the underlying premise and how good the law is.
Is "white people aren't allowed to run red lights" an "anti-white law"? Would it become an anti-white law if it was overruling a lower level of government, like if some municipalities were allowing white people to run red lights and the state government passed a law saying they couldn't make racial exceptions? Yes white people are more restricted than if they got an exemption from traffic law, but nobody describes the lack of such an exemption as anti-white, not even white supremacists. But this means that describing a law that restricts X group as "objectively an anti-X law" is just a way to smuggle in assumptions about what laws are reasonable. I think Folamh3 assumed the implicit argument was that those laws were unreasonable, not that they were anti-transgender in the same way that "Chinese-Americans need to pay income tax" is anti-Chinese, because otherwise the argument doesn't make sense.
Notice that guesswho didn't describe segregation of sports by sex as anti-male, despite men and boys being the overwhelming majority of those restricted, likely due to believing that the segregation is reasonable except for when it applies to people who identify as transgender.
No they weren't, The Guardian just made that up. It's not a prediction, it's a brief outline of a hypothetical written by two non-scientists (both self-professed futurists working for the consulting firm Global Business Network) who specifically state that it is extreme and unlikely. The point is not that they think it is likely to happen, but that they think such unlikely but extreme scenarios should be considered and prepared for by the Pentagon.
An Abrupt Climate Change Scenario and Its Implications for United States National Security
These are the steps they propose be taken:
Notice that reducing CO2 emisssions isn't even mentioned because their scenario is so abrupt that it would be too late, rather they are talking about preparing ways to mitigate the damage and/or do emergency geo-engineering, in case an unlikely scenario like that happens.
More options
Context Copy link