What Even Are Facts? How to Read a Scientific Paper

Am I the only one who feels like saying something is a “fact” doesn’t mean anything anymore?  I don’t have to give an example of two opposing sides using “facts” to steamroll one another–the examples are endless.

And honestly, I am about to freak out about it.

Which is why I want to take 10 minutes and talk about how to find scientific facts. Real, all natural, organic, farm-to-table, primary source facts, to help give you confidence that you are informed.

Before I explain how to read a scientific paper, I want to point out some important aspects about scientific research.

First: Where to find facts?

The internet (obviously).  

Many facts are born out of scientific studies, or even better, groups of studies that come up with the same result.

You can’t always type a topic into Google and get scientific papers as results; although sometimes you may find summaries of papers (known as abstracts).

In order to find the entire paper you often need to use search engines that are built specifically for scientific literature. Some search engines only cover one discipline, but some are more general.

Many (but not all) of these databases are paid subscriptions and if you’re a student you almost always have access through your university. Some reputable databases for the natural sciences are:

• Web of Science
• Wildlife and Ecology Studies Worldwide
• Fish, Fisheries & Aquatic Biodiversity Worldwide

Or if you want a more general database you can use

• Google Scholar (free!)
• ScienceDirect

Look at the journal it was published in.

When you find a paper with an eye-catching title (Lol! The titles are meant to precisely describe the study, and are not newspaper headlines), sniff around the journal it was published in.

Scientific literature gets published by individual journals and not all of them are created equally. Do a bit of research to find out if the journal is well respected.

Nature, Science, and PNAS come to mind as some of the most prestigious science journals.

Keep in mind that scientists pay a fee to have their papers published. There are predatory journals that exploit new scientists and will publish anything.

Look to see if the paper is peer-reviewed.

There are lots of PhDs with wacky views, and they can publish articles all day long. That doesn’t mean we should believe them.

The peer review process means that other people considered to be experts (often with PhDs) have read the paper and criticized it.

The reviewers don’t know who the original author is, which helps eliminate bias. The author of the original paper has to respond to those critiques and either fix the errors or justify why they belong before a good journal will publish it.

The journal should say upfront whether their articles are peer-reviewed.

Beware the preprint!

There is a growing practice known as pre-publishing, where the result of studies are released to the public before the time-consuming process of peer-reviewing and publishing begins.

There are obvious concerns about data no one else has analyzed, and major news outlets regularly report the findings of preprints without really explaining what that means. Sometimes the article will plainly identify the source as preprint, but also keep an eye out for language like, “preliminary results suggest.”

That said, there are completely legit projects that have pre-published their data, to the benefit of other researchers. The Human Genome Project is an example.

Opinions on the practice of pre-publishing are far ranging, but the point is: be aware that pre-publishing is a thing, and it means the data hasn’t been reviewed and critiqued by qualified experts yet.  

Who was the study funded by?

Science is expensive and sometimes people can be swayed by the agenda of the organization that pays for their research.

However it’s important to note that this likely isn’t the norm and nothing would ever get done if scientists couldn’t seek funding opportunities from a wide variety of sources.

Can you find other articles that come to similar conclusions?

Or is this some lone wolf paper in a sea of lambs? If so, question why that might be the case.

Notice the publication date.

Cancer research, for example, has advanced tremendously in the past 20 years. I wouldn’t rely solely on a paper from 2001 to inform my decisions.

Once you’ve examined these items it’s time to actually read the article.

Scientific Article Format

It’s the same every time; it includes an abstract, introduction, methods, results, and discussion.

The abstract is your friend.

The abstract is a short summary of the study and will likely include most of what you want to know.

The introduction will follow and will give you a little background information about the study including the purpose behind the study.

Methods: The broccoli of the paper.

The methods section can be the hardest one to get through because it can be long and tedious and include techniques and processes you don’t understand.

But it is extremely important to digest the methods if you want your opinion to grow up big and strong.

If the paper says it estimated the blood calcium levels of the average American woman by sampling blood from 2 women who worked at the university, this should raise some red flags about the validity of the conclusions.

Size Matters. For sample groups in scientific studies anyway. Generally, the bigger the sample size, the more reliable the findings will be.

Likewise, did the scientist choose methods that will provide the desired information? Did they say they sampled blood calcium by taking hair samples from those two women? That’s an odd methodology.

Most of the time the errors won’t be that glaring however. Which means you will likely need to spend LOTS of time understanding their processes in detail. A beer might help at this point.

Results and Discussion

The results section gives the hard and fast numbers generated by the study, and the discussion section explains what those numbers mean.

Do you agree with the conclusions drawn by the author? Don’t assume because this person is an expert that their conclusions can’t be questioned.

Scientists should be thought of more like “the man behind the curtain” than “the great and powerful Oz.”

If all else fails, contact the author.

If you’re really stuck on understanding a paper, often authors will have their email address right in the article. If they’re cool they’ll be happy to help you understand their article. They should be really flattered you give a crap about their niche-ass study anyway.

Fact: Comprehending scientific literature can be REALLY hard.

Scientific literature is not written for lay audiences. There can be a lot of jargon and techniques you’re unfamiliar with (especially if the paper involves higher math).

It isn’t always taught to undergraduates, and even graduate students can struggle. A common practice among graduate students and professors is to form reading groups that discuss a paper in their field. And yes, a lot of the discussion can sound like “wait so what is this graph showing?”, “I don’t understand their methods—like, at all.”

It will definitely take time to get good at reading scientific literature. Be prepared to spend A LOT of time looking up new vocabulary.

But don’t let this discourage you! One of the most helpful things my graduate advisor ever said to me was “why aren’t you working harder?” “embrace the stupid.” Meaning, don’t beat yourself up if you get bogged down. You are stupid. But we’re all stupid. Embrace it. Get comfy with it.

Because getting scientific facts right has real consequences.

The truth matters. Now go out and find it!