Science

The Problem With the March for Science

Our culture’s understanding of science is very, very broken, and on Saturday, it was impossible to ignore.

March for Science
Activists and protesters march to City Hall during the March for Science in Los Angeles, California, on Saturday.

Mark Ralston/AFP/Getty Images

Hundreds of thousands of self-professed science supporters turned out to over 600 iterations of the March for Science around the world this weekend. Thanks to the app Periscope, I attended half a dozen of them from the comfort of my apartment, thereby assiduously minimizing my carbon footprint.

Mainly, these marches appeared to be a pleasant excuse for liberals to write some really bad (and, OK, some truly superb) puns, and put them on cardboard signs. There were also some nicely stated slogans that roused support for important concepts such as reason and data and many that decried the defunding of scientific research and ignorance-driven policy.

But here’s the problem: Little of what I observed dissuades me from my baseline belief that, even among the sanctimonious elite who want to own science (and pwn anyone who questions it), most people have no idea how science actually works. The scientific method itself is already under constant attack from within the scientific community itself and is ceaselessly undermined by its so-called supporters, including during marches like those on Saturday. In the long run, such demonstrations will do little to resolve the myriad problems science faces and instead could continue to undermine our efforts to use science accurately and productively.

Let’s start with my contention that most “pro-science” demonstrators have no idea what they were demonstrating about. Being “pro-science” has become a bizarre cultural phenomenon in which liberals (and other members of the cultural elite) engage in public displays of self-reckoned intelligence as a kind of performance art, while demonstrating zero evidence to justify it. On any given day, many of my most “woke” friends are quick to post and retweet viral content about the latest on what Science (and I’m capitalizing this on purpose) “says,” or what some studies “prove.” But on closer look, much of what gets shared and bandied about is sheer bullshit and is diagnostic of one thing only: The state of science (and science literacy) in this country, and most of the planet for that matter, is woefully bad. For example, the blog IFLScience (IFL stands for “I f—ing love”) seems singularly committed to undermining legitimately good science half the time, while promoting it the other half—which, scientifically speaking, is a problem. Here’s a neat one that relays news about a study that suggested that beer hops may protect against liver disease. I’ll be sure to mention that to the next alcoholic with hepatitis and cirrhosis that I treat. To date that article has been shared 41,600 times. Very few of those readers, I should mention, were mice, though the research was carried out in, you guessed it, mice. (And of course, this type of coverage is not refined to cleverly named blogs.)

That’s not to say plans to cut back funding for research are wise (though so far most of those plans seem contained to a meaningless budget proposal). Nor should we tolerate it when our policies are poised to undercut genuine scientific expertise for politically expedient purposes (fair warning: That’s a Breitbart link; click at your own risk). It’s genuinely horrifying that national policy hinges on votes by congressional representatives who think the female body can “shut that whole thing down” in order to avoid pregnancy after rape or that evolution is, at best, a “speculative theory.”

But there is very little indication that what happened on Saturday will counter these misconceptions. Instead, the march revealed the glaring dissonance of opposing that trough of ignorance by instead accepting a cringe-worthy hive-mind mentality that celebrates Science as a vague but wonderful entity, what Richard Feynman called “cargo cult science.” There was an uncomfortable dronelike fealty to the concept—an oxymoronic faith that information presented and packaged to us as Science need not be further scrutinized before being smugly celebrated en masse. That is not intellectually rigorous thought—instead, it’s another kind of religion, and it is perhaps as terrifying as the thing it is trying to fight.

Indeed much of the sentiment of the March for Science seemed to fall firmly in the camp of people espousing a gee-whiz attitude in which science is just great and beyond reproach. They feel that way because, so often, the science they’re exposed to feels that way—it’s cherry-picked. Cherry-picking scientific findings that support an already cherished and firmly held belief (while often ignoring equally if not more compelling data that contradicts it) is epidemic—in scientific journals and in the media.

Let’s face it: People like science when it supports their views. I see this every day. When patients ask me for antibiotics to treat their common colds, I tell them that decades of science and research, let alone a basic understanding of microbiology, shows that antibiotics don’t work for cold viruses. Trust me, people don’t care. They have gotten antibiotics for their colds in the past, and, lo, they got better. (The human immune system, while a bit slower and clunkier than we’d like it to be, never seems to get the credit it deserves in these little anecdotal stories.) Who needs science when you have something mightier—personal experience?

Another example is the vocal wing of environmentalists who got up one day and decided that genetically modified organisms were bad for you. They had not one shred of evidence for this, but it just kind of felt true. As a result, responsible scientists will be fighting against these zealots for years to come. While the leaders of March for Science events are on the right side of this issue, many of its supporters are not. I’m looking at you, Bernie Sanders; the intellectual rigor behind your stance requiring GMO labelling reflects a level of scientific understanding that would likely lead for calls for self-defenestration from your own supporters if it were applied to, say, something like climate change.

But it does not stop there. Perhaps as irritating as people who know nothing about science are those who know just a little bit—just enough to think they have any idea as to what is going on. Take for example the clever cheer (and unparalleled public declaration of nerdiness):

What do we want?

Science!

When do we want it?

After peer review!

Of course, the quality of most peer-review research is somewhere between bad and unfair to the pixels that gave their lives to display it. Just this past week, a study published by the world’s most prestigious stroke research journal (Stroke), made headlines and achieved media virality by claiming a correlation between increased diet soda consumption and strokes and dementia. Oh, by the way, the authors didn’t control for body mass index, even though, unsurprisingly, people who have the highest BMIs had the most strokes. An earlier study that no one seems to remember showed a correlation of around the same magnitude between obesity and strokes alone. But, who cares, right? Ban diet sodas now! Science says they’re linked to strokes and dementia! By the way, Science used to say that diet sodas cause cancer. But Science was, perish the thought, wrong.

This leads to the other “big bad” in science research and, therefore, a widespread misunderstanding of science by the lay pseudo-intellectual community: allegiance to the all-mighty p-value. One sign I saw at a March for Science rally was the somewhat creepy inducement to “Show me your p-value!” The popular understanding of the low p-value (the common cutoff being less than 0.05) is that the data attached to it must be true. That’s a false understanding. The p-value actually means that the data in question has less than a 5 percent chance of being the result of chance if the underlying experimental null hypothesis is true. If you don’t know what that means, don’t refer to p-values. Further, a great number of studies have basically been manufactured simply by querying large data sets with a high number of possibly relevant questions. By the math of it, around 1 in 20 of these questions will yield a positive result that can be published. This happens approximately all of the time.

And the p-value isn’t the whole story. There are a dozen other statistical games that researchers can play to get statistical significance. Such ruses do not rise to anything approaching clinical relevance. Nevertheless, fun truthy ones like the diet soda study grab headlines and often end up changing human behaviors.

The reason this problem, what one of my friends delightfully calls statistical chicanery, is so rampant is twofold. First, academics need to “publish or perish.” If researchers don’t publish in peer-reviewed journals, their careers will be short and undistinguished. Second, large pharmaceutical companies have learned how to game the science system so that their patented designer molecules can earn them billions of dollars, often treating made-up diseases (I won’t risk public opprobrium naming those) as well as other that we, the medical establishment, literally helped create (opioid-induced constipation being a recent flagrancy).

Of course, the journals themselves have suffered because their contributors know the game. There are now dozens of stories of phony research passing muster in peer-review journals, despite being intentionally badly written. These somewhat cynical, though hilarious, exposés have largely focused on outing predatory journals that charge authors money in exchange for publication (assuming the article is “accepted” by the rigorous peer-review process; the word rigorous, by the way, now means “the credit card payment went through and your email address didn’t bounce”). But even prestigious journals have been bamboozled. The Lancet famously published fabrications linking vaccines and autism in 1998. and it took 12 years to retract the studies. Meanwhile, the United States Congress took only three years for its own inquiry to debunk any link. You know it’s bad when the U.S. Congress is running circles around the editorial board of one of the world’s most illustrious medical journals. Over the last couple of decades, multiple attempts to improve the quality of peer-review adjudication have disappointingly and largely failed to improve the situation.

While the scientific research community is in desperate need of an overhaul, the mainstream media (and social media influencers) could in the meantime play a tremendously helpful role in alleviating the situation. Rather than indiscriminately repeating the results of the latest headline-grabbing scientific journal article and quoting the authors who wrote the paper, journalists should also reach out to skeptics and use their comments not just to provide (false) balance in their articles but to assess whether the finding really warrants an entire article of coverage in the first place. Headlines should be vetted not for impact and virality but for honesty. As a reader, be wary of any headline that includes the phrase “Science says,” as well as anything that states that a particular study “proves” that a particular exposure “causes” a particular disease. Smoking causes cancer, heart disease, and emphysema, and that’s about as close to a causal statement as actual scientists will make, when it comes to health. Most of what you read and hear about turns out to be mere associations, and mostly fairly weak ones, at that.

I am glad that people believe science is a concept worth marching for. But the reality is that the state of affairs within the scientific community and literacy among its “fans” lies largely in shambles. The March for Science, and the somewhat mindless glee that was on display, is entirely antithetical to the idea of science as a whole. For one thing, scientists are legendary flip-floppers. That’s actually a strength of genuine science, not a weakness. When the facts change, opinions must follow suit, not vice versa. We need to work to fix the problems of poor research and science comprehension if we’re ever going to have a reasonable case against those who disparage science, and as the march showed, we still have plenty of work to do on this front.

This does not render science deniers correct on any particular topic. Far from it. They indeed remain the primary enemy and are wrong on the issues a staggering majority of the time. But in the long run, the propagation of bad science that feels like good science—and the inability or unwillingness of well-meaning progressives to distinguish it from the truth—only strengthens the hand of the opposition.

Disclaimer: The opinions expressed in this article are solely those of the author and do not reflect the views and opinions of Brigham and Women’s Hospital.