Spotting and Understanding Junk Science Reporting
Junk science reporting is everywhere. Here's how you can fight back.
In the world of science writing, a certain level of integrity must be maintained. It’s unbelievably easy to misrepresent research, and the incentives in the American media system are a little twisted. I’m a fervent supporter of modern science, believing it’s the way forward out of the darkness of superstition that’s haunted the distant and not-so-distant human past alike. We no longer pray to the rain gods and hope for the best. Now, we understand weather patterns and likelihoods, we can build complex mathematical models, and we can engineer nature to help people thrive and survive.
These are all very good things.
But from the research lab to the printed paper (or digital website), the details are often lost in the sauce. One of the most common exploitations is the use of the word “link” when discussing scientific research. A link merely means that some mathematical connection was established between two things. A link can either be strong or weak, depending on the quality of the evidence.
In other words, a link could mean something, or it could mean nothing at all and just be an accident due to poor study methods (or interpretations).
Here are some excellent examples from Statology.org.1
Clearly, shark attacks don’t drive ice cream sales (unless sharks suddenly developed employable skills, started earning paychecks, and devouring ice cream as a desert after taking a big bite out of human beings, but it’s doubtful).
It’s not that shark attacks cause ice cream sales. It’s a correlation, not a causation. As you can see, the chart above is in months, so it’s safe to say that people go to the beach in summer (and are more likely to be bitten by sharks) and eat more ice cream in summer.
Same with master’s degrees and box office revenue, though I sense there’s no relevance here except for the population increase. More people live today than there were in 1910, and more movies are made, thus, more people are getting master’s degrees and purchasing tickets, and more seats are being sold.
Yep. Nuclear energy doesn’t cause drowning (nor does drowning kickstart nuclear energy production). Maybe this has something to do with the population boom, you think?
Again, while this is technically a “link,” there’s no causal relationship between the two things, nuclear energy production and drownings. We don’t need to go drowning people as human sacrifices to the gods of nuclear power.
I’ll let you caption this one.
Go ahead, and give it your best shot. Feel free, whatever comes to mind. Consider it a statistical Rorsarch test.
And finally…
Okay, okay, maybe high school and pizza might have something to do with one another. Still, I can assure you that simply graduating high school doesn’t transform our physiology into pizza-scarfing hedonists.
Again, we have a correlation that likely has to do with the population expansion.
But it’s easy to see how all of these figures can be easily used by unscrupulous media outlets to mislead people. And it happens all the time.
I can’t tell you how often I’m doing research for The Science of Sex here, and I just cringe when I read the study and then read the reporting on the study afterward.
Yeah, I have weird hobbies, laugh if you will. #ScienceNerdFTW
Panty Fetishes & Brain Damage
My favorite example of this shoddy research is a case study that linked a man’s panty fetishes with decreased blood flow to the brain. Just to be clear, small studies have looked into whether or not brain flow affects our personalities. But studies like this one with 18 people don’t really tell us much. The sample sizes are so small that it could all just be a weird coincidence. If 2 out of 100 people are allergic to peanut butter (2%), and if you only study 18 people, there’s a chance that you could wind up with a study that says 44% of the population is allergic to peanut butter! That would happen if you just so happened to get 8 people with the allergy out of the 18 you picked for the study.
Then it could get picked up by the major news outlets looking to capitalize on fear-clicks: Study finds 44% of the population allergic to peanut butter! Then everyone freaks out and throws away their peanut butter, calling for bans, conspiracy theory groups crop up on Facebook, talking about how the government is slowly poisoning us with peanut butter. Yeah, stuff gets crazy sometimes.
Back to blood flow to the brain and panty fetishes.
These two things have absolutely nothing to do with one another. Fetishes are complicated behaviors that require a blend of nature and nurture, things like imprinting, access to the taboo (a man from a country where women are required to dress in head-to-toe coverings might develop foot fetishes because that’s the only part of a woman’s body they can see), and much, much more. It’s not as simple as saying, welp, a lack of blood flow is the culprit.
Nonetheless, the study says:
Few reports have described the association between fetishism and brain function. We examined this association using single-photon emission CT (SPECT) in a 24-year-old male patient who was arrested for stealing underwear and referred to our hospital for evaluation. The patient had stolen women's underwear on multiple occasions since the age of 11 years, although he showed no interest in underwear that belonged to his mother, sister and girlfriends. He was an academically strong student. SPECT examination revealed bilaterally decreased cerebral blood flow in the temporal and occipital lobes. Other neurological conditions were ruled out, and he was diagnosed with fetishism on the basis of history and examination and successfully treated with behavioural therapy. Klüver-Bucy syndrome is characterised by temporal lobe dysfunction and altered sexual behaviour. We believe that decreased function of the temporal lobe may have been associated with fetishism in our patient.
I added the emphasis because this is couched in several different euphemisms for uncertainty, saying, “We believe…may have been…associated…” is totally different from “He had low blood flow to the brain, which caused him to steal panties.”
There’s not a single mention of it being a cause in the study.
Plus, it’s a case study. Sticking with the peanut butter allergy analogy (that just sounded good, didn’t it?), the sample size isn’t low—it’s one single guy. They found one guy who presumably has low blood flow to the brain and also happens to have a panty fetish that’s gone to tragic lengths.
Yet, that didn’t stop the site Medical Daily from boldly proclaiming in its headline:
Underwear Fetishes Are Caused By Decreased Blood Flow In The Brain, Or At Least That's The Conclusion Of A Japanese Study
That wasn’t the conclusion of the study, clearly.
But it gets clicks.
This kind of crap reporting has become commonplace, sadly, which is part of why I write this Substack—to clarify subjects for you all when people—scientists or journalists—try to mislead you.
Relating to the young man in question, I think this speaks volumes about our current climate of media and how broken it’s become.
You fight back by approaching these kinds of news articles with a skeptical eye. When you see a “link,” take it with a big, fat grain of salt. I’ve almost entirely stopped using search engines. They pay zero mind to the quality of information they deliver. Their job is to find websites, not answers. Realizing this helped me tremendously.
Whatever happened to the idea that boys will sometimes do bad things sometimes, and those bad behaviors need to be corrected by parents and society? No, this isn’t a “boys will be boys” argument, but I worry that we’re trying to overly medicalize very natural things (that should be dissuaded and punished). Young people (before the age of 30 or so, went the pre-frontal cortex is fully developed) make bone-headed decisions often.
My twenties weren’t exactly my finest hours and would certainly bar me from running for political office in any capacity when people found out how much I drank and hung out with friends instead of doing more important things like philanthropy.
It goes without saying boys shouldn’t go around stealing panties, not by any stretch of the imagination, but we’re humans, evolved for a very different world from the civilized one we live in.
Not every bad behavior needs to be medicalized.
Cases like his need role models, ethics, and teaching—not to be labeled with a condition or disease. He needs an explanation as to why it’s wrong not to be labeled as a bad person or, as would likely happen here in America, imprisoned and labeled as a sex offender for the rest of his life for what’s essentially just theft.
Not to mention, we can’t extricate this case study (from Japan) from the very sex-shy culture in which it took place. We should loosen our sexual morals by expanding our sexual understanding and not shun people with kinks and alternative ways of practicing their sexuality so long as no one is harmed. In this case, laws were broken, and harm was done. In no way am I justifying this behavior.
But I can’t help but wonder what would happen if we had a freer society where we were all a little bit more tolerant of sexual deviations from the norm (within the boundaries of consent), one that facilitated unorthodox individuals and gave them safe, consensual spaces to exercise what tickles their fancy without judgment.
Correlation is the Causation! [Not really :)]
It makes me laugh over and over to watch some folks that are anti-science scream bias or use this method thinking they are smart. When they aren't.
Great descriptions ... Ice Cream anyone while we watch Sharkando