Here
and
Now

opinions

6 tips to help you detect fake science news

16 Comments
By Marc Zimmer

The requested article has expired, and is no longer available. Any related articles, and user comments are shown below.

© The Conversation

©2021 GPlusMedia Inc.

16 Comments
Login to comment

Some players have a vested interest or are sponsored by a company to make favorable research.

9 ( +9 / -0 )

Reporting in the news on science can be horribly inaccurate. What the public gets in digested form is often not what the original research shows.

That said, peer-reviewed research is often of deplorable quality. Efforts like the Reproducibility Project have been showing in recent years how bad it is. In many fields, the majority of research published in respected journals cannot be replicated. With most research, no one even tries to replicate it. Many peer reviewers never closer look at the methodology and data, but just jump to the results and discussion. The use of complex statistical models that the researchers themselves do not fully understand, but that helps them show significant findings amidst a lot of noise in the data, is often at fault. Peer-reviewed is no guarantee of quality research. It lives up to that standard less than half the time.

Academic publication is necessary for professional advancement. Much of what is published is about getting tenure, winning grant money, and building a professional reputation. The quality of the actual science is, sadly, distant behind these other concerns.

8 ( +8 / -0 )

@mikeylikesit, some good points about peer review. Even when done "properly", it's good to find out exactly what has been reviewed. That may depend on the publication. In some cases, it is only a check that accepted scientific procedures have been followed, but without any attempt to reproduce or falsify the results. Of course, that's not always practical (e.g. if it involves collecting data from the bottom of the ocean).

Meanwhile, regarding the correlation/causation thing, I'm sure many of us try to live longer in order to drink more red wine. :-)

6 ( +7 / -1 )

Funding and conflict and interest must be mentioned on most of the scientific paper.

Of course one can lie about it, but once caught the reputation will be badly damaged.

6 ( +6 / -0 )

The posters name is usually the best indication,

4 ( +5 / -1 )

There are two types of fake news. The first are theories. Flat land. Reptilians around. An accurate forecast of earthquakes is impossible. Car engine on water. And then let him write whatever he wants. People have the right to make mistakes in thinking. The second fake news is practical medicine advice. Tips that people can use directly on themselves. And here it is necessary to punish harshly and immediately for deception. To fine two people a large sum for falsifying scientific results - no one else will lie.

4 ( +6 / -2 )

6 tips to help you detect fake science news

Does your science info come from someone;

Who's claim to fame is from hawking a pillow?

Goes bare-chested and wears a fur hat with horns?

Thinks Alien Lizards have infiltrated our government?

Believes Hugo Chavez is still alive and stealing votes?

That COVID is caused by 5G network towers?

Thinks Colorado is on our southern border?

3 ( +6 / -3 )

Also, as I often mentioned, there are 100's of studies investigating the effectiveness of hydroxychloroquine. If you take out all those that involve conflicts of interest (links to big pharma), you are essentially left with studies that all show a favorable effect.

No, that would let you without any studies.

Conflicts of interests are only one of the many kinds of bias a study can have, it may be as simple as one scientists too in love with a drug, endlessly searching for anything to use it, or it can be as complicated as a relationship with sponsors to develop something that is never going to bring profit (at least in money).

HCQ is a very good example of how bad science took the spotlight by doing things fast and sloppy, while the good science ended up a few steps behind with a much more stronger conclusion, HCQ has no use for COVID.

Conspiracy zealots love to think the whole scientific and medical community is on the pocket of some powerful company but that makes no sense. Only people that cannot see anything apart from money it would make sense a conspiracy wher doctors and scientists hide a supposedly live-saving drug (that could be used to save their friends and family) just so some rich owners of a company can make a lot of money, it makes no sense. Specially when other dirt cheap drugs are perfectly fine being accepted as effective, like dexamethasone.

At this point HCQ is long abandoned, even Raoult, the scientist that ended up making unethical studies with manipulated results to support his pet drug has ended up accepting HCQ (with or without azithromycin) does nothing to reduce the deaths of patients. Eventually all fake science ends up being discarded, even by its own proponents.

3 ( +8 / -5 )

Also, as I often mentioned, there are 100's of studies investigating the effectiveness of hydroxychloroquine. If you take out all those that involve conflicts of interest (links to big pharma), you are essentially left with studies that all show a favorable effect.

Actually what kicked it off outside China was a completely rubbish study (of 24 patients) by Didier Raoult. All the talk of 'studies' showing dramatically effective results circled right back to him. Long before COVID he had an extensive track record of ethical violations, wacky science, bullying, and intimidation, so it should have been a surprise to no one that he claimed near perfect results for his treatment, produced data and conclusions that could be shot to bits without even trying, and angrily attacked anyone who dared to question him.

The science has left him behind, and HCQ has been found both ineffective and unreasonably risky as a COVID treatment. The book's closed on that one, but the Raoult cult is still open for anyone gullible enough.

3 ( +7 / -4 )

No, there are many studies that are not linked to big pharma, and they almost all show that HCQ has a positive effect.

No, funding is not related, good studies, with higher statistical power and valid, ethical protocols show that HCQ has not usefulness in the disease. On the other hand the more limited is a study, with less participants, deficient blinding, no randomization (and less ethical authors) the more it is likely it will give false results, some of those false results would give the appearance of effect, but once the limitations are corrected this stops and the true (zero) value is reached.

I mean, for Rault, the person that went to the extremes of manipulating and falsifying results to push for this drug to finally recognize it does not save patients it means it became impossible to keep pushing for this, even with fake science.

3 ( +8 / -5 )

These points are all important. But he left one out: follow the money. It can be tough, but it's worth finding out who funds research in contentious fields to get a clearer picture on the motivations behind it.

0 ( +5 / -5 )

I have two words to show these points do not work: Haruko Obokata.

0 ( +0 / -0 )

No, there are many studies that are not linked to big pharma, and they almost all show that HCQ has a positive effect.

All your "studies" are funded by Big HCQ and Big Vitamin D Pills, for whom you are a propagandist.

-2 ( +2 / -4 )

Yes, but that’s only for people who need such an advice. Luckily, I am my own probability estimator, statistics interpreter and science study proofreader. And in most cases this is very much more than only sufficient. Remains point 2 as the hurdle, admitted....lol

-3 ( +2 / -5 )

Also, as I often mentioned, there are 100's of studies investigating the effectiveness of hydroxychloroquine. If you take out all those that involve conflicts of interest (links to big pharma), you are essentially left with studies that all show a favorable effect.

No, that would let you without any studies.

No, there are many studies that are not linked to big pharma, and they almost all show that HCQ has a positive effect.

But considering all of your other posts, of course I am not surprised that you consider only the pharma-linked studies to be "good studies" and those that are not linked to big pharma to be "bad studies". I think this illustrates very well how to detect fake science information.

-5 ( +4 / -9 )

zichi and Gooch bring up a very important point, that of the funding of research.

Much of the research is funded directly or indirectly by big pharma. And some papers will be reviewed and approved by big pharma-linked researchers. When we submit a paper, we can often suggest people to review our manuscript, as well as provide names of people who should not be asked to review it. Lancet-gate is a good example of how all this can go terribly wrong.

Also, as I often mentioned, there are 100's of studies investigating the effectiveness of hydroxychloroquine. If you take out all those that involve conflicts of interest (links to big pharma), you are essentially left with studies that all show a favorable effect.

-8 ( +4 / -12 )

Login to leave a comment

Facebook users

Use your Facebook account to login or register with JapanToday. By doing so, you will also receive an email inviting you to receive our news alerts.

Facebook Connect

Login with your JapanToday account

User registration

Articles, Offers & Useful Resources

A mix of what's trending on our other sites