3 ms·
Instead of quoting the article, how about we quote the actual proposed regulation? The proposed regulation provides that when EPA develops regulations, includi
by refurb 8y ago
Instead of quoting the article, how about we quote the actual proposed regulation?
The proposed regulation provides that when EPA develops regulations, including regulations for which the public is likely to bear the cost of compliance, with regard to those scientific studies that are pivotal to the action being taken, EPA should ensure that the data underlying those are publicly available in a manner sufficient for independent validation.[1]
Seem eminently reasonable to me. And seems aligned with the "open science" movement to me.
And in terms of privacy...
EPA believes that concerns about access to confidential or private information can, in many cases, be addressed through the application of solutions commonly in use across some parts of the Federal government. Nothing in the proposed rule compels the disclosure of any confidential or private information in a manner that violates applicable legal and ethical protections.
[1] https://www.federalregister.gov/documents/2018/04/30/2018-09078/strengthening-transparency-in-regulatory-science https://www.federalregister.gov/documents/2018/04/30/2018-09...
- Retric 8y agoChanging the rules after the fact is generally a terrible idea. Suppose zero studies that satisfy that criteria study lead. Should the EPA suddenly say lead must be safe in any concentration?
- refurb 8y agoThese standards are no different than any reputable scientific journal. Try and publish a paper in Nature and tell them “I can’t share the data”.
- smsm42 8y agoWhy would there be zero studies that study lead and have source data? What happened to those that studied lead - did they destroy their data for some reason? Are the data about lead somehow unique so that, unlike any other data, they can't be published? Did some weird catastrophe occur which erased all data regarding to lead studies, and nobody knows about it? What would be the reason that basic scientific norms would not be preserved in this case? I guess if it happened, we'd need to conduct new studies - and find people who screwed up the old ones and make sure they are not responsible for any new studies ever again, to avoid future screwups.
- Retric 8y agoScience is old. Good luck tracking down data from ~50 year old studies published in 1970. Further, their is little point in replicating old studies so often that's all we got.
- smsm42 8y ago1970 is not a Stone Age, and civilization didn't collapse since then. Why would it impossible to have data? Also, does it mean if a study in 1970 made a mistake, we're now doomed to carry on with this mistake forever? And if not, how would we discover it if not by trying to replicate/re-analyze data?
- ouid 8y agoNew research is overwhelmingly more likely to hurt industrial interests than help them, so adding any hoops at all is likely going to benefit industry. In fact, all that this does, from a regulatory perspective, is provide a pathway for the EPA to accept and reject studies at its own discretion, without having to ever actually call the result into question.
- refurb 8y agoThe EPA already weights the value of studies in a public way. In terms of new research are you saying that it’s ok to keep the raw data a secret? Why? We don’t do that for any other type of science.
- ouid 8y agoYes we do, and it doesn't matter. Data isn't relevant to the reproduction discussion. The reproducibility crisis is a crisis of methodology, not falsified or misinterpreted data. If I p-hack a result, then no amount of transparency is going to demonstrate that that's what I've done. The only standard of evidence for reproducibility is reproduction, and they are only using the "reproducibility crisis" as a diversionary tactic to increase the cost of doing science to an unacceptable level, at the same time as pulling EPA funding of science.
- refurb 8y agoGive me some examples of science where we make decisions and keep the raw data a secret. I can’t think of any except ones involving national security. Yes, the best way to prove reproducibility is to reproduce it, but that can’t always be done and at any rate is no excuse for keeping raw data a secret.
- jhayward 8y ago> Nothing in the proposed rule compels the disclosure This is blatant disingenuousness. What they are saying is "you need not disclose private information, but we won't use your study if you don't".
- refurb 8y agoNo they aren’t. They are saying “we already have ways of sharing data and not violating privacy”.
- jhayward 8y agoYes, they are. Do you have any familiarity with studies conducted under a human study IRB permission? Please cite one that would permit sufficient sharing with the public of the detailed information needed to replicate the study. Even just sharing one personal detail, such as gender, or age, is almost always completely forbidden. Also, please don't be so credulous. This rule has been promulgated, invented, and proagandized by people who are trying to hamstring the EPA from using scientific studies to regulate industry. It's not new, it's just newly empowered. You appear to be falling for the plausible cover story, which is disinformation. People are really bad at being able to discern the intent behind "plausibly good" organizations and movements.
- refurb 8y agoThere are plenty of published human studies where raw, anonymized patient level data is shared.
- smsm42 8y agoNope, they are saying "disclose information in ways that do not violate privacy, ask us how if you don't know how to do it, we're already doing it. But if you don't bother, only then we won't consider your study".