PARTING THOUGHTS

You’re still here, so you must be a sucker for punishment—or maybe just someone who cares as deeply about this as I do! Either way, I do have a few more things to say—tips that didn’t quite fit anywhere else in the guide or ideas so critical they bear repeating. I’ll keep this brief (ish).

1. Read. A lot.
The best writers are voracious readers. You can’t expect to write great papers if you haven’t read great papers. You’re unlikely to invent a brilliant system for writing papers out of the blue. Instead, you’ll learn by emulating the patterns and decisions made in papers that clicked for you. It pays to be an avid consumer of the work you hope to create; in this case, I'm not even sure what the alternative would be.

2. Set out to write the paper you’d be excited to read.
Honestly? I hate to say it, but a lot of scientific papers I read aren’t enjoyable. Some are borderline unreadable! So, I set my bar here: Be better than average—in clarity, structure, logic, and yes, even tone. Science should work, but it can be pleasant to read, too.

3. Steal (structure, not content).
If the structure of a sentence, paragraph, or section makes good, repeatable sense to you, borrow it. That’s not plagiarism—it’s how complex skills are learned. Just like with a golf swing or violin solo, you improve by modeling the excellence displayed by the forms of others.

4. Learn to read for style as well as substance.
Relatedly, when you read papers, don’t just read them for what they say; take notes on how they say it. What makes good papers flow? What tricks of tone or pacing or structure are making them more readable? Start a folder for the greatest papers you find and consult them for ideas whenever you sit down to start a new writing project.

5. Seek out feedback. Often.
You often will never know what isn’t working for others if there are no "others" to inform you. Seek out someone who's willing to read your work as early as you can. This person needn’t be a mentor, either. Peers, friends, partners, and labmates can all tell you when you’re being confusing or boring, even if they aren’t familiar with the subject matter.

6. Give writing the time it actually takes.
Writing isn’t quick, easy, or painless—not for most of us, anyway. It’s iterative, it’s collaborative, and it’s full of fits and starts. I’ve spent weeks just doing a literature review before. Turning around a second draft with co-authors? That can take months. Build this reality into your timeline, and you won’t be so surprised when it becomes your reality.

7. Embrace the 85-15 rule.

You’ll get a lot of feedback during the writing process—from co-authors, mentors, peers, reviewers, editors, etc. Most of it'll be valuable; you should take the bulk of it seriously. 

But inevitably, some advice you'll get'll be confusing, contradictory, or irrelevant, and some edits will not match your voice or tone at all.

What do you do then? In those instances, I guide my decisions with what I call the “85–15 rule:” I assume roughly 85% of the feedback I get will be worth acting on—and the remaining 15% I should let lie.

As a grad student, I felt like I had to accept every suggestion I received, especially from mentors and reviewers. I wish I could go back and tell that version of me: “Sure, your advisors know more than you. That’s why you should defer to them a lot—but not always. It’s your research. It’s your paper. What are the odds they’re right about literally everything?”

If you're training to be an expert—or already are one—it’s okay to trust your gut sometimes. In those instances, feel empowered to push back. Just do it with gratitude and humility, and you’ll be fine.

8. Write for the diverse, global audience you want.

In science, your audience is (and should be) global. So write like it. Don’t assume your readers will understand your corner of the world as well as you do—or speak your language natively. Keep idioms, regionalisms, and jargon to a minimum. Don’t reach for “academic” words when more commonplace ones'll do. If you want your reader to think you’re smart (and let’s be honest, who doesn’t?), impress them by how simple, not how complex, you can make your research sound!

9. Negative results are still results.

It bears repeating: There's absolutely nothing wrong with so-called “negative results”—when outcomes don’t match your predictions or are inconclusive. A well-designed, well-intentioned research project always teaches us something. If your results don’t come out the way you’d hoped, own it. Tell us what you think those results mean, just like you would’ve for “positive” results. Without knowing what didn’t work, we’re doomed to repeat fruitless lines of inquiry. And don’t let anyone make you feel like any less of a scientist for getting “negative results” either!

10. Get organized.

One of the most challenging parts of writing a scientific paper for everyone, experts and novices alike, is managing sources. Personally, I still print my papers, highlight them to high heaven, and use those highlights to skim for relevant passages when I need evidence or context. I then store these highlighted versions in Zotero to produce bibliographies. Is this the best system? Nah. I hate large parts of it! I think every scientist I know hates at least some aspects of the source management system they use. But the key is we have a system. If you don't have one yet, adopt one your peers are using (or find one for you all to adopt). Pretty much any system will save you time and headaches at some stage in the process.