How to: Read a scientific research paper and come away smarter


Unfortunately that needs step 0: have access to more than the abstract, which is something I’ve found tricky for some fields I’m trying to learn about, and can’t imagine managing for everything that might be interesting. Which I suspect is a key reason people rely on abstracts in the first place.

More than the abstract: Research libraries are wonderful things. Many colleges will let members of the public into theirs.

(Yeah, I know, I’m not going there.)

It doesn’t even need to be a research library. You’d be surprised what journals a public library card will give you access to. Minneapolis Public Library comes with online access to bunches.

1 Like

That’s a good tip, and is a thing I’ve been doing, but needing a special trip still counts as being hard to access for a lot of people. I mean yes, if you’re trying to become an expert or write about something it’s not really any trouble, but I got the vibe that this was advice for everyone.

Each field has its own style and jargon. And statistical methods vary by field - the statistical analysis of a plant genetics paper will be nothing like the statistics of a human nutrition study. And immunology has its own style, because most genes and proteins have original names that are nothing like the what is later determined to be their actual function (which is often the opposite of the name).

Also, some papers are good and some papers are bad. To tell them apart, you need to be familiar because the techniques in each field, and what a well designed experiment looks like compared to a badly designed experiment. A 2x2 climical trial with 600 subjects? I don’t think so!

So if you want to read scientific papers, focus on a specific subject. Spend a couple days reading a paper, then start over with a fresh hard copy. It will look completely different the second time through. Keep doing this, and in as little as 6 months you will see a noticeable improvement in your reading comprehension.

This topic was automatically closed after 5 days. New replies are no longer allowed.