28 January 2016

The cost of selectivity

Scientific Reports and Nature Communications are both published by the same company, Nature Publishing Group. Both are online only, open access journals.

But Eigenfactor pointed out that Scientific Reports charges an article processing fee of US$1,495, while Nature Communications costs more than double that, US$5,200.

Why the price difference? Since they are both at the same publisher, it’s obviously not a simple infrastructure difference, like one journal having a physical print run, different manuscript submission systems, and so on. Both appear to offer the same services to authors.

There seem to be two factors that might explain the price difference: staff and selectivity.

Scientific Reports seems to be run by editors who are working scientists. Presumably they are not drawing most of their salary from the publisher, and are working mostly on a volunteer basis, which is common for scientific journals. The editors of Nature Communications are Nature Publishing Group staffers, who presumably are getting salary from the publisher. I wonder what the expected and actual difference in outcomes are between these two editorial schemes.

Neither journal seems to report what percent of manuscripts are ultimately published, but the criteria for Scientific Reports is that research be “scientifically valid and technically sound.” On the other hand, Nature Communications says it publishes “important advances of significance to specialists,” so clearly it is setting itself up as the more exclusive club.

Why be selective in a purely online journal? There is no limit to the number of pages, and I expect the cost of server storage per paper is fairly trivial. The selectivity is no doubt to increase journal Impact Factor, which in turn drives prestige and desirability. And at first glance, it seems to be working: the journals’ web pages report Scientific Reports Impact Factor is about 5, and Nature Communications is 11 and change.

But... the blog for Frontiers in journals (owned by Nature, incidentally) has a post claiming there is no relationship between Impact Factor and rejection rates. The problem that James Hardcastle and Anna Sharman pointed out is that while they archived data on Figshare, the data does no include journal names, so it’s not verifiable.

As far as I can tell, the only revenue stream for these journals is their article processing charges. As I mentioned before, this means that published papers are subsidizing the costs for the rejected ones. When I started this post, I though the comparison of these two journals might give a glimpse into just how big that subsidy is. But it’s hard to disentangle from the differences in editorial management.

I’m intrigued by all this because the open access “baby journals” that share the name of a paywalled glamour magazine (Science, Nature, Cell) seem to be able to charge prices that are well above the market for most open access journals. To reuse yesterday’s graph, they all break the axis:

I’m curious as to why this pricing scheme survives. Do people confuse Nature Communications with Nature? Is the reputation of the publisher just that strong that it commands a premium, even for a relatively new journal? Is there no competition on value or services to the authors? Do people really expect ten times the prestige because they paid ten times the cost?

Related posts

Fluctuating publication costs

External links

Selecting for impact: new data debunks old beliefs

No comments: