Wednesday, October 1, 2008

What's so special about seven?

There are the Four-Figure Wonders (Nature, Science). There are the Go Ahead and Publish Your Last Eight Years of Work in 16 Figures Plus Three Tables (some society journals).

But the majority of mainstream biology papers I've seen contain seven figures. Why?

It isn't as though research just naturally falls into seven figures. Putting together a paper, we often struggle to rearrange the panels for maximum intellectual impact, of course, but we also aim to put it into the magic seven.

The last paper we sent out contained seven figures. One reviewer pointed out that two figures could be plausibly combined; and then suggested that we accumulate enough more data to produce another figure. We did, and we did. Reshuffle, and bingo--still seven figures.

(Speaking of bingo, I'm playing Palinbingo Thursday evening--who's in?)

It's clearly an accepted convention (and yes, there are 6 and 8 and 9 figure papers. But the mode in my field, for sure, is 7.) What are the pluses and minuses?

On the down side, from the author's perspective, it's silly to shoehorn your Great Ideas into a prefab structure. Maybe your paper would unfold best in 5 very dense figures, or 10 thinner ones. But because of the norms, I would hesitate to send out such a manuscript (of course, I'm a junior postdoc; maybe when I'm a big fancy pants person who gets mentioned on YouTube has her own grant I'll have the confidence to buck convention.)

As a reader, though, I rather appreciate it. Especially if I'm skimming a paper, I like to have a sense of how the scientific argument will build, and how many experiments will be synthesized into a picture.

It's kind of like reading a novel: you know that when your hero is trapped in a perilous situation, but you're still gripping three hundred pages in your right hand, that this will not be the final test. Conversely, when the lovers are separated by cruel fate and the pages have dwindled to a precious few, the resolution is not likely to be happy.

Not that any science papers I've read have involved lovers, or peril, but you get the drift.

Here's what I would like to know: how did we converge on seven? The old papers I've read are idiosyncratic--the four figures, or the sixteen, at authorial whim. Do we gravitate towards 7 because of its mystical qualities? (That'd be embarrassing, for us rational scientists.) Have our psyches been colored by the 7 Dwarves, or the 7 Deadly Sins (all depending on your childhood, I guess)?

I don't know the answers to these questions, but since 7 is so powerful, next time I write a manuscript I'm going to try for 7 subheadings, and 7 sentences in the abstract, and maybe 7 supplementary movies. If the voodoo works as expected, then I'm predicting reviews like, "I don't know why, but this paper exerted a strange and wonderful force over me."

20 comments:

Anonymous said...

Interestingly, at a departmental seminar run by one of our top professors on publishing papers, he talked about his strategy of matching the number of tables and figures in his ms. to the magic number generally published by that journal.

Professor in Training said...

I think you're missing the biggest 7 of them all ... the magical $7 billion???

Anonymous said...

I'm sooo playing bingo with you! I printed off my card (4). I can barely contain my excitement.

Anonymous said...

There may be a visual/design component. I used to edit a high school yearbook, and we learned that apparently the human eye is most pleased by an odd number of pictures on a double-page spread, and 7 tends to be the most pleasing of all. Now I know all your figures won't be on one page, but perhaps there is some sort of crossover; a golden proportion of figures.

Nat Blair said...

Great point DrJMrsH. In fact it's something I've been thinking about a lot lately, and will have to whip a post up about it.

Frankly I hate any artificial limitations on the total number of figures, because what I typically see happening is that people force their stuff into that number of figures. They do it either by increasing the number of panels in each figure, or by sticking things into Supplementary. Neither is ideal in my opinion.

As for 7 figures, I always assumed it was because most Cell and Neuron papers were supposed to be 7 figures and a model figure.

Dr. Jekyll and Mrs. Hyde said...

Anon at 4:43--Awesome. I'm not sure if I'm also turning it into a drinking game (a drink every time she says Alaska or Maverick? You'd need a fireman's lift home) or what. After the roe v wade clip with Couric, if she speaks in whole sentences it'll be amazing.

Anon 5:31...Maybe that'll be my next paper trick. In fact maybe I'll only put out trick papers in the future. That gives me a goal...

Welcome back Nat! Yeah, 7 + model is common, but so is 7 = 6 + model. Why? Why?

Anonymous said...

anon 443 again...

I'd be drunk with her first rambling attempt! 90 seconds is a long time to give me bingo easily and that's just one question. I think card 4 is an early shoe in with line 4 (Maverick, Obama, National Guard, Bailout, Heartland). They are all short words she can remember and string together with lipstick.

The Roe v Wade case... I think she thinks it is ROW or WADE over to Russia or where the bridge to nowhere was supposed to go. It's sick that I'm having fun with this.... I may be crying come November.

I'll log on at debate time. I'll put a bottle of wine on chill now. JC

Anonymoustache said...

---"Do we gravitate towards 7 because of its mystical qualities?"---

Mystical and wired into our brains, huh? Well, the Golden Ratio is approximately 1.6, and 1+6=7 so there!
Q-E-freakin-D!
Can you tell I'm getting geared up to be hit by blast after blast of for Palin logic (I know, that's an oxymoron) later?

Anonymous said...

Our most recently accepted manuscript has 12 figures + 4 supplemental figures.

Candid Engineer said...

The more figures, the less I want to read a paper. Seven is on the cusp of manageable to me. It just gets overwhelming after a while. Especially with a lot of these molecular biology papers I read nowadays where each figure has eight subpanels because of all of the required controls. Eeek.

Anonymous said...

Christ, i need a stiff drink already... betcha, darn, hockey mom... AHHHHHHHHHHHH

Anonymous said...

Card 4 BINGO!!!!!!!!!!!!!!!!!!!
Maverick, Obama, National Guard, Bailout, and Heartland!
WHOO HOO!

chall said...

Gee gosh, Im happy I didnt se the whole thing since these last 30 mins I am _craving_ a drink.. or three..

She really said "drill baby drill" on national tv?!? Not to mention the "special needs baby" and ya....

my my my... I am SO happy this is not my election. Sorry to say but seriously.... and I was kind of happy she was a woman but really??? "We are the best country in the world and need to b the only one with nuclear weapons since we can take care of them".

kicker; "Im a main streeter".

over and out, for now.

(Biden kicked ass about the single parent though. And about Cheney!! wow. Impressed me.)

ScienceMama said...

My grad paper totally had 7 figures.

Dr. Jekyll and Mrs. Hyde said...

Anon--nicely done! The demand was high so I invented a couple more cards....I did not bingo, sadly. Maybe counting on her to say "culture of life" wasn't a good idea :) Congratulations on a good pick!

I watched with the lab at the pub, and a photog from the local newspaper took a katrillion pics of me and Postdoc Advisor watching the debate. We'll see if they use any.

Comrade, 12? wow. My eyes are glazing just thinking about it. No, wait, that's the beer. Way to buck the trend!

Favorite Line of the Night: "Well, I've only been at this for about five weeks...." YEAH NO SHIT.

Nat Blair said...

Our most recently accepted manuscript has 12 figures + 4 supplemental figures.

Rock on CPP, I love it.

Can I ask why figures in supplemental? Did the journal have a limit, or were they peripheral points that still needed to be conveyed?

Dr. Jekyll and Mrs. Hyde said...

I just love "CPP." Put it on my cereal, even.

Anonymous said...

Can I ask why figures in supplemental? Did the journal have a limit, or were they peripheral points that still needed to be conveyed?

Actually, this journal has no word or figure limits. To be honest, the reason four of the figures are supplemental is that is what the first-author post-doc did, and we just left it that way.

Ms.PhD said...

I like the Palin drinking game better. Every time she dodges a question, you take a drink.

But I don't see why they couldn't be combined.

PiT, it's $700 billion. Not 7. But hey, we're talking billions, so who cares?

And I think the "what everyone else does" phenomenon is more important in publishing than some might realize.

But all this stuff came about originally for historical reasons.

Back in the day, publishing was slow, so you had super-short rush papers (e.g. BBRC, BBA) and then you had longer things for less competitive areas (but also a bigger accumulation of work and longer times between papers). It was more cost-effective that way.

I would be curious to see a mathematical model of whether we're more likely to continue to converge on, god forbid, more of a standardized format (you know to save time on reformatting bullshit?) now that the Internet is more involved, etc.

Seems to me that we're seeing more and more of a split between people who publish in Certain Journals and people who publish wherever (assuming it's all available online anyway).

Very curious to see how it all shakes out.

Anonymous said...

Cf http://en.wikipedia.org/wiki/The_Magical_Number_Seven,_Plus_or_Minus_Two