May 14, 2009

Science is a Gaussian distribution away from truth.

Today I had a very interesting conversation with YoungEnthusiasticScientist (YES) over lunch. I was asking him some statistical questions because I was having trouble getting my head around how to do certain things. And then our conversation veered to the dreaded question, had my paper gone out for review? Have I submitted TheMonkey yet?

I sheepishly said 'no.' But then explained I didn't want to submit something that I didn't think wasn't truthful. How could I submit something that wasn't accurate to my best ability, wouldn't that be lying?

His reply, "Then you'll never be able to submit it." Gesturing with his hands, he drew a line up in the air, "Truth is up here." Then he drew an imaginary Gaussian distribution below this imaginary line. "And all of us that do science, we're down here. Science is an approximation of the truth. I have papers that I submitted years ago that I totally disagree with the conclusions now. That's how science progresses. You do shit - find out its wrong - change how you do shit and do some more shit - then find out that it's wrong, etc. Science is recursive."

I answered, "But won't the reviewers reject it if they find problems? And my PhD supervisor, GeneralSolutionGuru says you should be your own worst critic."

YES responded with, "Reviewers will often reject something when they can feel clever. If they read your paper and find something that they can go, 'aha, now I feel clever' your paper will be rejected. So don't let them feel clever, just put in all the caveats and then send the thing out!"

Okay, YoungEnthusiasticScientist. I will. I know I have a problem called perfectionism, but I can do this. Despite the personal shit that is currently happening, I will have this paper to GSG by the end of next week. I can let go.

4 comments:

Ms.PhD said...

This person is right. Truth is relative. All we can do is try to aim for it. We know we'll usually be off, sometimes by a little, sometimes by a lot. When you submit a paper, you take your best shot: and then the reviewers use their egos to try to block you. Being your own worst critic can't protect you from that. There are always going to be a million little holes. This is one of the hardest things to live with when you want your work to stand the test of time. Nothing is perfect forever, but it can be really good right now, and that's about as good as it gets.

Successful Researcher: How to Become One said...

The observation in the penultimate paragraph is just great!

Anonymous said...

Send out the paper. Let the reviewer's have their say - some comments will be "too clever" and obnoxious, others may actually be insightful and help strengthen your work.

It's a lot harder to address imagined concerns vs. concrete ones.

Phagenista said...

The people who are telling you to err on the side of submitting more quickly are right. Do you have a group of scientists (your former labmates, current fellow postdocs) that can critique your paper this week before you send it to a journal? You'll get some feedback for revisions and you'll feel more confident about the submission.

The liability of a brown voice.

 It's 2am in the morning and I can't sleep.  I'm unable to let go of the ruminations rolling around in my brain, I'm thinkin...