It would be great to have 35% more citations and over 5 times as many downloads, but I’d rather have 2,740 Jamaican beef patties.
P.S. Here’s the article in question: The ladder of abstraction in statistical graphics. I absolutely love this paper. It’s based on an idea I’ve had for awhile that I spoke on a few years ago in Ron Yurko’s statistical graphics class at CMU. Then I wrote it up and submitted it to the journal, which gave some useful comments, and I enlisted Kaiser Fung to get it over the finish line. Enjoy.
You don’t want them to pay them an exorbitant amount of money to do something that you could do anyway for free???
Have you ever been contacted by a journal for posting a greyware PDF? And you are so far past tenure, nobody cares whether Gelman et al. 2026c is in a fancy journal. So what on earth do they think they can offer in exchange for the price of an overseas vacation?
Sean:
One could ask why I publish in journals at all. Here are some answers:
1. A feeling of closure. When a paper is unpublished, it feels like it’s just hanging there, and I to close the loop.
2. A published paper can be cited and can have more legitimacy, so I feel it will get more readers and influence.
3. If I’m coauthoring with students, publications can help their careers (not an issue in this case).
4. It’s good to publish some minimum number of papers per year for grant reports etc. (not an issue in this case, as I’m not crediting any funders).
5. I have this silly goal of publishing in 200 different journals (not an issue in this case, as I’ve published in the American Statistician many times already).
Agreed that an overseas vacation could be better than 2,740 Jamaican beef patties. I could even go to actual Jamaica!
I think journals still have some value for curation and archiving and community formation, but anyone charging $4,000 to share a fourteen-page PDF is nuts and senior research academics should definitely push back
Quote from above: “1. A feeling of closure. When a paper is unpublished, it feels like it’s just hanging there, and I to close the loop.”
I noticed the “close the loop” words, which migth be a fitting description for what I have noticed in my mind when it comes to writing things. Whether it be poetry or lyrics or scientific manuscripts, they have to be “out there somewhere” to feel like I completed the task or cycle or closed the loop.
For me though, I merely need a pre-print server to “publish” my manuscripts to make me feel that the thing is done and the loop is closed. In fact, I have noticed and I reason that journals and peer-review might even cause things to not be done, because they likely have different or other criteria, suggestions, recommendations, etc. that might all influence the final product in ways I do not necessarily intend or want.
It is a good listing, although (2) is something I’m no longer sure about.
As for the APC business model, I’d argue that diamond OA is probably the way to go, including with respect to public sector money, funding bodies and foundations, and the like. And as long as the beneficiary is a non-profit or a public sector entity.
As there has been a lot of (partially misguided) discussion here about “computer science conferences”, I’d also note that ACM went full-OA recently. So CS certainly does something right too.
But, in general, I agree with with the “No” conclusion; I wouldn’t pay either (and including from grant budgets).
Some more consideration is useful: Andrew’s paper isn’t a research article and doesn’t result from funded research. It’s a write up of an idea, and no doubt Andrew could produce one of these 2 or 3 times a year. It would indeed be dumb to pay $8-12000 per year on this – where would that money come from?
But science publishing is a little different. Andrew’s example is: “have an idea – write the paper”. Normally the scenario is “have an idea – write and submit a grant – obtain funding – do the research – write the paper”. The grant writing and funding will likely take at least a year (no guarantee of success) and the research project is likely to be funded for, say, three years.
An NIH grant might be funded at $250,000 per year (direct cost) over three years to support a postdoc or two, a couple of PhD students and a contribution for technical support (it funds equipment and equipment access, laboratory consumables and so on). Over the course of the grant 4-6 papers might be produced. Government and charity funded research must be published Open Access (quite reasonably – the public that ultimate funded the research should get to see it!). Perfectly good journals charge around $3000 for Open Access publishing (Scientific Reports – $2850; eLife $3300), although there are reasons why one might wish to aim for a higher impact more costly journal.
So from one year of the grant 1-1.5 papers might be published – that’s $3000-4500 out of $250,000; i.e. publishing costs are a very small percentage of the total award.
What does the journal do to justify this $3000 charge? eLife ($3300) pays editors, staff involved in article processing including hosting the published paper and supplementary data, support staff, and does some marketing, feature writing and journal development. It has to set aside funds for the fee-waiver system whereby research from low-income countries, or unfunded research can be published with reduced or zero charge.
I’m not trying to justify journal publication fees – just presenting a more realistic view of scientific publishing – it’s both a large amount of money and a small amount of money depending on one’s perspective!
Chris, I publish one or two papers a year while working a day job that pays less than anyone on your NIH grant is paid. Academic publishers are trying to get rid of the paid editors, and the data and image wranglers often wreck things instead of helping because they rush them and refuse to hire competent people. Nobody in ancient world studies has a meaningful marketing budget or paid “feature writing.” And I am not fond of rent-seekers. The fact that lab sciences have a lot of money does not mean that anyone is entitled to drain off a few percent.
I don’t disagree Sean; as I said I’m not trying to justify journal publication fees. You’re probably right though that the willingness of funders to pay Open Access fees for the research they fund makes researchers somewhat blasé about the costs. But I like that publicly funded (govrnmnt/charities) research is accessible to everyone.
Anyway, for scientific research the Open Access publication costs should more realistically be in the $2000 ballpark as assessed by journals like F1000Research (Open Access publication costs $1758, though an odd publication process) or society journals (e.g. Journal of General Physiology $2000). The near $4000 OA charge for Andrew’s paper in American Statistician seems like excessive profiteering – it would be interesting to know how they justify this.
Here’s a paper that indicates that the actual raw cost of publishing a research article should be less than $1000: Grossmann A and Brembs B. Current market rates for scholarly publishing services [version 2; peer review: 2 approved]. F1000Research 2021, 10:20.
BTW I’m not funded by the NIH – I was using that as an example likely to be more relevant to US readers. My research is/has been funded by UK medical charities that not unreasonably fund at much lower levels. I’m guessing that ancient world studies may have journals with more volunteer support or might pair up with anthropology studies – do you have to find significant publication fees for your papers?
I’m curious if this is pay-for-performance, or just pay-me. A lot of marketing contracts do not offer guarantees.
I enjoyed working on this paper. There is more work that should be done along these lines.
Kaiser:
I think the system is: pay $3940 for open access or pay $0 and it’s under a paywall at the journal website. I’d be willing to pay $100 or something to support the American Statistical Association, but $3940 is ridiculous, it’s pure rent-seeking by the publisher and they’re not even trying to hide it.
I obviously agree, but I wanted to say my reason to like open-access, which is not one of those they listed: I much prefer reading from a nice, one-column, reflowable HTML than from the usual PDF, and I also enjoy the convenience of having references and supporting materials one click away. Of course that’s still no reason to give up the beef patties, and some preprint sites let you host the article this way (let me know your favourites!). The other worry, that it’s harder to find the preprint, is a minor concern now that the scholar browser plugin is very good at finding them with one click.
Then we also have diamond open-access, which seems like the best of both worlds.
Wait, “Rigorous peer review for every open access article”?
So if I don’t pay for open access, I don’t get rigorous peer review?
Sounds like a win-win.
On closing the loop: Allen Ginsberg said of Jack Kerouac, who had written ten or so books without being able to find a publisher, that he had been “published in heaven.” Eventually though, the loop was closed on earth as well.
Just keep me out of the loop
I don’t want, nor need, to know
I need courage, strength, and wisdom
While I ignore the show
I’ll focus on what I can do
If need be, I’ll step out of the queue
I’ll try and find what’s good and true
I’ll hear my soul and look for a clue
Then maybe there will be a day
When all is quiet and serene
That I will hear, think, and say
I just closed the loop