How Science Is Made: A Glimpse of the People, Institutions and Money Behind It

This article appeared in the June 2021 issue of Discover magazine as "Show Me the Science." Subscribe for more stories like these.

Some scientists wish to uncover truths of the natural universe — to learn the properties of distant stars, or deep-sea creatures, or the interior of our cells. Others seek solutions, hoping to better our lives or undo the damage we’ve done to our environment. The list of motivations runs long, depending on who you talk to. But most people don’t know any scientists personally. In 2017, about 4 out of 5 Americans polled couldn’t name a single living scientist, according to Research America. Of those who could, top answers were Stephen Hawking (27 percent), who died in 2018; Neil deGrasse Tyson (19 percent), who last published research in 2008; and Bill Nye (5 percent), who quit his job as an engineer in 1986. Yet 1.5 million-plus Americans are currently working as scientists, which is more than the number of elementary school teachers.

We don’t know their names because they’re mostly behind the scenes, trying to resolve questions, bit by bit. Few will ever do work that makes the news. Even fewer will garner enough publicity that people begin to recognize them. Regular Discover readers may know names like astrophysicist Avi Loeb, or Jennifer Doudna, the 2020 Nobel Prize winner for her work in CRISPR gene-editing. But before we could edit genes with CRISPR, people were accumulating decades of data on microbiology and genetics. Pull any researcher today out of a hat, and we can only speculate how their work might change our lives.

Despite their power to improve the world, modern scientists face the realities of polarization and politicization. “Those of us who study science see this as a really unsettling time,” says Allan Brandt, a historian of science at Harvard University. “We’re alarmed at the erosion of scientific legitimacy and authority, because it’s so crucial to solving the world’s problems.”

Recent history illuminates how strategic corporate interests and politics can undermine science, beginning on a large scale in the 1950s. That’s when Big Tobacco began sowing seeds of doubt with tactics that many industries still use today. Shortcomings in academia and medicine also have harmed public trust, such as rare incidents of fraud and the many ways that racial, gender and other biases have informed research and public health; those blunders have especially hurt women, people of color, Black and Indigenous communities and LGBTQ+ people. In 2020, fractured trust ruptured into a public health disaster, as scores of Americans revealed that they believed the COVID-19 pandemic was either a hoax or purposefully and drastically exaggerated for political gain — despite constant assertions to the contrary from health officials, doctors and researchers.

Listen to scientists and you might hear that climate change could be mitigated, vaccines save lives or genetically engineered crops can help combat starvation without adverse health effects. Why should we believe them? The answer can only come from an examination of the process of science itself, which unfolds along a web of funding, research standards and public perceptions.

Carlos Mariscal, a philosopher of science at the University of Nevada, Reno, thinks a big part of science’s public perception problem is poor communication. “We haven’t done a good enough job of bringing people behind the scenes to the process of science, as opposed to the product of science,” says Mariscal.

Take nutrition science and eggs. If you just read the headlines, you’d think eggs were a staple food one day, a cholesterol-filled death trap the next, and then back to being a healthy option before you even had a chance to find a new favorite breakfast. If you looked deeper, Mariscal explains, what looked like a flip-flop on the surface was really just scientists doing what they do best: learning. When researchers identified cholesterol’s role in heart disease, and cholesterol’s high levels in eggs, they warned people to be cautious about their egg consumption. Then when they discovered that there are two types of cholesterol and eggs have the healthier kind, eggs went back on the menu.

“I think that it’s genuinely one of the most impressive and one of the best features of science,” says Mariscal. “When it goes wrong, it fixes itself. It can correct itself.” Researchers are constantly following up on what current evidence suggests is true, and pushing the boundaries of what’s known. The field is designed to both challenge our current understanding and wade into questions that have no solid answers — at least, not yet.

The formal process of science typically begins when a researcher thinks of a specific, testable question and seeks to answer it (typically with a team of experts). The approach goes through multiple reviews, approvals and, often, failed attempts, to secure funding; we’ll tumble down that money hole shortly. The question may get tweaked along the way, and research involving animals or humans is subject to an additional review board and committee. If money is secured, the lab or fieldwork begins and the team documents their findings. When it’s time to share them with the world, they write a research paper and select a scientific journal that might publish it. Options include many smaller, subject-matter-specific journals and some bigger names, such as Science or Nature. That’s when the crucial peer-review phase kicks in.

After submission to a specific journal, the main editor will give a paper first look. If it seems to be a good fit, that editor will identify typically three other scientists in the relevant field of research (the eponymous peers in the process) and ask them to review the paper. That means multiple experts examine the work ultra-carefully, looking for anything that seems amiss. They may say: I think this datapoint is a glitch; you used the wrong analysis here; or, your results are fine, but the conclusions you drew are flawed. Any given paper goes through at least one round of edits between the authors and the reviewers — typically more — until everyone involved agrees that the paper is sound. This thorough vetting and scrutiny is the reason peer-reviewed journals form the bedrock of solid science.

Money, too, is embedded in this foundation, a reality that can draw criticism and scrutiny. If you want research, you need funding. So, who’s forking it over?

Simply put, research and development (R&D) is funded by numerous businesses, organizations and government bodies. While nailing down the figures gets sticky — and it depends on how you define science research — $580 billion was invested in R&D in the U.S. in 2018, according to the National Center for Science and Engineering Statistics. Barring more than half of that spent on experimental development (which includes a lot of private R&D — such as an auto company improving their car engines — that never lands in journals or public view),roughly $211.5 billion went toward basic and applied science research. From that total, businesses contributed about 43 percent of the funds. The rest came from federal money (38 percent) plus state governments, nonprofits and other institutions.

The bulk of federal science funding gets dispersed as grants to universities and institutions through the National Science Foundation (NSF), the National Institutes of Health (NIH) and other agencies. At an agency like the NSF, a panel made up of external people — experts in the relevant fields — reviews each proposal and makes a recommendation for which projects get funding. Typically, the granting body isn’t involved with the research once it begins. Scientists remain independent to do their work, but share a few progress reports to the funding institution along the way.

“People think that if someone is funding something then they’re basically buying that research, and the people doing the research, and that’s not quite the case,” says Antoinette Serrato, a climatologist at the University of Nevada, Reno.

Things can, however, get a bit complicated when for profit industries invest considerable money into research, according to Mariscal. “Definitely in the majority of cases, the funding has little to no noticeable effect,” he says. “[But] you have these really malicious uses of funding, as with the tobacco industry, that went out of their way to fund lots of research,” he says.

For instance, the tobacco industry invested in research about the genetic factors of lung disease. They also built legal cases around asbestos to fend off plaintiffs who smoked and got cancer. None of that means the research itself was conducted improperly, even if the motivation for and use of the research was shady. “They just wanted to muddy up the waters,” Mariscal says. Essentially, research could be used like a shield to protect tobacco sales.

Even when funding sources aren’t directly participating in the research process, they do have one ability that gives them power: choosing what to fund. This was a core strategy of the tobacco industry’s disinformation campaign that lasted for half of the last century. They funded scientists to study other causes of cancer besides tobacco, and more.

“It’s a pretty dramatic story,” says Brandt, the science historian. Top tobacco company executives gathered at New York’s Plaza Hotel in December of 1953. It was a crisis moment for them, with major findings beginning to connect the dots between smoking and lung cancer. So, they called in a public relations expert, John Hill, the head of one of the biggest PR firms at the time.

“Hill [basically] said to them, ‘Don’t try to dispute this. Don’t ignore it. If you don’t like the science that’s coming out, produce your own science.’ And so the industry set up an industry research program,” Brandt says.

The strategy has come to be known as the “tobacco industry playbook.” Others, like the oil and beverage industries, have followed suit. The main goal? Fund as much research as possible that distracts from your harmful product, as well as any research that might demonstrate your product is safe. That way, you can point to the other research and say that the link between the product and the harm is not clear.

In 1981, just a few years after Exxon scientists found a convincing link between fossil fuels and climate change, company executive Roger Cohen wrote an internal memo warning that the continued consumption of fossil fuels could be catastrophic, “at least for a substantial fraction of the population.” Yet external communications from the company maintained a different message. Even years later, then-chief executive Lee Raymond said in a 1996 speech at the Detroit Economic Club: “Currently, the scientific evidence is inconclusive as to whether human activities are having a significant effect on the global climate.”

Today, this type of messaging is still the bread and butter of oil and other lobbyists, and can be repeated by U.S. politicians and PR consultants alike. In some instances, campaigns have realized that simply saying “the evidence is inconclusive” is enough to achieve their goals — without actually funding or presenting research.

These tactics can complicate the playing field when it comes to individual pieces of research. But the long-term trajectory of science has ways of sifting out and correcting work that is infected by ulterior motives or human error.

There are ways to identify corporate interests, especially in modern work. Most scientific journals require authors to disclose any conflicts of interest and their funding sources before publishing results; anyone reading scientific findings can look for this disclaimer, usually at the bottom of a paper, before the works cited.

And although it may seem that researchers are incentivized to falsify their work, the incentives to not conduct research fraud can be much stronger. Researchers caught conducting unethical research can lose their jobs and be blacklisted from the field.

In 2005, a professor in the Department of Medicine at the University of Vermont was the first American researcher to go to jail for falsifying data — and he was not the last. Since 2009, the NSF — through federal law — requires that all institutes that receive NSF funding mandate Responsible Conduct of Research training for their researchers. It’s like driver’s ed or sex ed, but for scientists, complete with dramatized videos depicting the consequences of unethical practices.

Sometimes, scientists do make an honest mistake that slips through — like misidentifying an insect specimen, failing to notice the cat walked across the keyboard and changed a number, or choosing the wrong model to analyze their data. If a researcher down the line notices the error, the paper can be retracted — essentially, unpublished. Or future studies can show why something was wrong, and the earlier paper becomes obsolete, even if it’s still published. This is one reason why research is an ongoing, cumulative process: One bad study won’t typically cause significant harm. “There are some times where there’s stuff that gets done that gets overturned. But that’s fine, it’s what we expect,” says Mariscal.

Ultimately, the whole process is designed to ensure that science unfolds ethically and accurately. But science also doesn’t happen in a vacuum. The impact it has on the world rests not only on human scientists, but on the interplay between policymakers, media and society. Communicating findings through this tangle of opinions and channels complicates matters drastically.

Journalists at Discover and other publications play a role in this chain, poring over science journals, parsing compelling findings and talking to the researchers. Distilling all the info, and discerning what the public needs and wants to know, is more art than science. Some days, the news is “Study Says Eggs Are Back on the Menu.” Other days, science communication is explaining the arrival and approval of a vaccine created in record time — and poised to save countless lives around the globe.

“Academia and industry and government can produce effective, excellent science for human good,” Brandt says, pointing out humanity’s current hope against COVID-19. In that sense, the same global pandemic that has fueled rampant misinformation — and revealed the extent of some people’s distrust — also offers a striking example of science and its process working properly.

With so much information drifting around us, it can be hard to tell what’s real and what’s bunk. When it comes to scientific research, here are six questions that can help you decide when to trust a study’s claims, and when to remain skeptical.

1. Is this info peer-reviewed?

All papers are not created equal, even if they run in a bona fide journal. Sometimes you can find preprint papers that haven’t been fully vetted. Letters from experts or editors also appear in journals. Examine the language at the top and bottom of papers to understand what you’re looking at.

2. Who did the study?

On any scientific paper, you’ll find a list of authors and their institutional affiliation. Look for trusted universities and institutes. Take note if the researchers work for a for-profit industry or a nonprofit with a policy agenda. This information is typically right underneath the author names at the top of a paper, sometimes hidden in a drop-down menu. If it’s not there, it might be tucked at the end of the paper somewhere.

3. Who funded the study?

Research rarely gets published anymore without a disclosure about research funds. Look for federal agencies like the National Science Foundation or the National Institutes of Health. Notice if the funding came from a for-profit industry or a nonprofit with a policy agenda. Look for a funding disclosure and conflict of interest statement near the acknowledgement section at the end of the paper.

4. What were the parameters?

Consider whether test subjects in studies were animals or humans. Sample size is also a critical component. If you want to know if a new drug is safe, would you be satisfied with results on just one person? Or 100 mice? Or 100,000 people? Although different types of studies require differing sample sizes to get satisfactory results, trust your gut when a number seems low. Look at the methods section of a paper to see sample size. It’s often explicitly stated with the letter n (as in “n = 100”).

5. Do the results support the conclusions?

There are a lot of reasons why researchers occasionally write a paper where the results don’t exactly support the written conclusions, and they’re not all malicious. But this is a key distinction that can separate good studies from bad. Be wary of conclusions and claims that exaggerate the actual findings or go beyond the scope of the data collected in the study. This one requires some expertise or practice, as well as a thorough look through the whole paper.

6. Do other studies agree?

Finally, any single study is just that — one study, typically in a vast field of similar work. Before a research finding starts to be accepted as a possible fact, countless other studies need to confirm it and try to disprove it. Although you might be able to find a few studies that claim carbon emissions from humans don’t cause climate change, that wouldn’t negate the thousands of others that show the opposite. Whenever deciding what to believe, look at as many studies as possible. What does the larger body ofevidence, as a whole, suggest? The best shortcut to this is to find a review paper or what’s called a meta-analysis. These are papers written by experts that summarize numerous studies and all the findings on a subject to date.

In 2018, $580 billion was spent on science research and development (R&D) in the U.S., compared to $548 billion in 2017. The total includes the categories of experimental development and basic and applied research. The vast majority of development funding (85 percent, or $314 billion in 2018) came from businesses. Much of that work is proprietary for products, goods and processes. A total of $211.5 billion went toward basic and applied research. From that total, businesses contributed about 43 percent; federal money funded 38 percent; and state governments, nonprofits and higher education institutions invested 19 percent.

U.S. R&D Funding by Source and Category, 2018

Anna Funk is an ecologist and science writer based in Kansas City, Missouri.