Rejoignez 15,5 milliers d’autres abonnés

John Oliver on the dangers of junk science

06 takeaways from this video:

00:00:21 Science is extremely complex, and new studies and findings are constantly being released. These “studies find X can/may/might Y” splattered across your television and trickling down your social media profiles may seem contradictory and ridiculous at first glance…

…and is enough to make you jaded; questioning the legitimacy of the entire scientific field.

All scientific studies are equal, but some are more equal than others.

00:02:31 With the unfortunate reality of tenure, reputation, and funding at risk, all the while mainstream media’s business model of glorifying new and different content as they push that content to the less scientifically-saavy, general population, scientists seem to have found themselves in the same boat as an online blogger seeking views for monetization. “My success as a scientist depends on me publishing my findings as frequently as possible and in the most prestigeous outlet.” –Brian Nosek, PH.D at the Center for Open Science

Even prestigious organzations such as TED have featured talks which have created discord, for example scientists Paul Zak and Molly Crockett’s disagreement over the implications of oxytocin.

00:03:11 These demands, unfortunately, have given rise to junk science – spurious or fraudulent scientific data, research, or analysis which has been incorrectly gathered, analyzed, measured, tested, or interpreted but is nevertheless presented as scientific fact, especially for personal gain and advancement, for example in a court of law.

Consciously or unconsciously, ways scientists can manipulate scientific studies results include:

  • Controlling the study’s duration or sample size
  • P-hacking, or data dredging: the data mining process of ‘finding’ seemingly significant correlations between causes without first attempting to identify the underlying cause for the correlation.

[EDITOR’S NOTE: Recall in the documentary End:CIV. How Our Industrial Society is Leading Towards an Ecological Apocalypse that companies often hide their environmentally UNfriendly products behind scientific studies in order to increase profit. Greenwashing is “is a form of spin in which green public relations (PR) or green marketing is deceptively used to promote the perception that an organization’s products, aims or policies are environmentally friendly.”

00:04:24 Science’s best defense against junk science are replication studies where other scientists repeat the original study to compare the results and test congruency. Unfortunately,

  • “Replication studies are so rarely funded and so under-appreciated that they never get published, therefore no one wants to do them. There’s no reward system in place, so you only have exploratory studies out there taken as fact that has never actually been confirmed.” –Elizabth Iorns, PH.D at Science Exchange
  • The media’s obsession with new and different content means they will actively diffuse original exploratory studies, but won’t actively share replication studies which disprove the original study. So the junk science results are what stand in the general population’s minds and get shared.

Individual results shouldn’t be taken seriously until they are put into context with all the work as a whole.

[EDITOR’S NOTE: Recall in Brad Frost’s lecture Death to Bullshit that 90% of everything is crap and Karl Popper’s lecture How & When to Override The Autonomous Mind that information pollution in the internet age is that, with freedom of speech combined with the ability to instantly publish anything you want immediately and without verification, as information accumulates in any field of study, high-quality information in a particular subject tends to grow in a linear fashion, while the total amount of information on that same particular subject tends to grow as a cube, meaning as more and more information accumulates on that particular subject, it becomes harder and harder to rummage through the useless, polluted information in order to locate the high-quality information because it’s unavoidably mixed in with crap. This will only get worse.

Also, recall in Tyler Cowen’s TED talk Why You Should Be Suspicious of The Stories You Hear that stories act as an information filter allowing the storyteller to pack a lot of information and social power into a brief narrative by ignoring certain bits of information while highlighting others. Humans are biologically programmed to respond to stories. But the more inspiring a story makes you feel, the more nervous you should become because the best stories are often the trickiest ones. When information is conveyed to us in the form of a story, we often place more importance on it, remembering it when perhaps we shouldn’t.

Lastly, recall in Robert Sapolsky’s lecture Human Behavioral Biology: The Dangers & Challenges of Categorical Thinking that although humans tend to think categorically because doing so is the simplest, easiest, and most convenient way to make sense of something and to then communicate that something to someone else there are a bunch of problems when it comes to categorical thinking, namely:

  • When you pay too much attention to category boundaries, you don’t see the big picture; all you see are categories.
  • When you think in categories, you underestimate just how different two facts can be when two things “fall under” the same category. Likewise, you over-estimate how different two facts are when they “fall outside” of the same category.]

00:09:30 Many studies are conducted on lab rats, and the majority of these tests that worked on rats failed to work on humans.

00:13:29 Is all this media coverage on junk science dangerous? Consider…

  • The dehydrated while driving as dangerous as drunk driving study (mentioned above) was based on a study of 12 men and was funded by Coca-Cola, a company who would directly benefit financially from the diffusion of these test results.
  • The tobacco industry spent years claiming “the scientific results that smoking causes cancer weren’t conclusive.”

Were a person to take at word the results of all these studies and apply them to his or her life in order to be ‘as healthy as scientifically possible,’ a person might come to the conclusion that everything causes cancer and kills you, that all doctors and pharmaceutical companies are corrupt and lying to them, and that brands and news medias agencies will say anything to gather information and get more clicks to monetize their website throughth native advertising, so therefore there is nothing humans can do to protect themyselves from cancer or stay healthy.

All this would lead to a compounded lack of trust for the overall integrity of real science, causing people to further deny the proven fact that humans are destroying the planet by climate change because the “studies are still inconclusive.”

“A study says…” should include sourcing or context, or it shouldn’t be mentioned at all.

[EDITOR’S NOTE: Recall in Deepak Malhotra‘s lecture 20 Tips to Better Negotiate Your Job Offer & Compensation Package at Harvard University that at every stage in a discussion, whenever someone asks you a question, or says something ambiguous or that you weren’t expecting, investigate and ask for clarification so you understand the situation. The better you understand the situation, the more wisely you can respond to it.]

4 réponses à “201. John Oliver’s Dangers Of Not Challenging Junk Science”

  1. […] 201. John Oliver on the dangers of not challenging junk science […]

  2. […] NOTE: Recall in John Oliver‘s show The Dangers Of Not Challenging Junk Science that mainstream and online media’s monetization business model of glorifying makes discrediting […]

  3. […] Lastly, for more on how junk science and statistics seep into your life and ruin your logic and critical thinking skills, check out John Oliver On The Dangers Of Not Challenging Junk Science.] […]

  4. […] NOTE: Recall in John Oliver On The Dangers Of Not Challenging Junk Science that individual results shouldn’t be taken seriously until they are put into context with all the […]