Thursday, September 8, 2016

Predatory versus low cost?


September 8, 2016

Predatory versus low cost?
 By David Wojick, Ph.D.


Synopsis: Low cost journals listed as predatory have taken off and are publishing a huge number of papers. The concept of "predatory journal" may incorrectly include a lot of legitimate low cost journals, masking a major change in scientific communication.

Last week we raised the issue of whether the the US Public Access agencies should screen the journals whose articles they post, as PubMed Central does. In discussing this issue with industry experts we discovered a recent report which presents some amazing, even revolutionary, numbers.

The report is Shen and Bjork, "‘Predatory’ open access: a longitudinal study of article volumes and market characteristics," BMC, 2015. What they did was study a sample of the roughly 11,000 journals on Beall's list of so-called predatory journals, and then they project the results to the entire list. This is crude, so the results are admittedly rough estimates and we will treat them that way by rounding them off a lot.

Here are the stunning numbers. First, the number of articles published in 2014 is over 400,000. This is an enormous number of articles. Second, the number published in 2010 was just over 50,000, so the growth has been explosive. So has the growth in the number of active journals over this short period, going from 2000 to 8000. And if the reported growth has continued there should be well over half a million articles published this year alone.

I think this is actually good news, masked by a colossal conceptual confusion.

Specifically, the so-called "predatory" net is actually capturing a lot of simple low cost journals. Note that they classify about 11,000 journals as predatory. Last I knew there were an estimated 30,000 indexed journals. So they are classifying roughly one third to one quarter that number of journals as predatory. Not likely. The total number of published articles may be as high as two million so the rough fraction is the same, one third to one quarter. Are we to believe that this many articles are somehow being published fraudulently? Surely not.

The key datum is the average APC of less than $200. Here is what I think is happening. The developing countries, especially China and India, are pouring a lot into research, hence generating a lot of articles. (Last I knew China was overtaking the US as the leading generator of scientific articles.) In pace with this we are seeing the rapid growth of the low budget APC journal, to serve the low budget researcher market. This makes economic sense and there is nothing predatory about it.

On the contrary, many OA advocates see the end state as one of very low APCs. Well here it is, in part anyway. The thing is that a $150 APC journal cannot look like a $1500 journal, which is very fancy. Back when Beall's list first gained prominence I studied it closely. My conclusion was that it was picking up low budget journals per se, the predatory ones being just a small fraction. My favorite example is a journal that seems to have been classified as predatory just because the mailing address was an apartment, not an office (in Montreal).

These numbers suggest that I was right. If so then what we are seeing is actually part of the globalization of science, which I consider a good thing. Poor researchers publishing in low cost journals.

My point is that if these journals are publishing on the order of half a million articles a year then they are not predatory. They are an extensive and fast growing new literature. And if they are not being indexed then that in itself is a major access problem.

To be clear, I am not claiming that there are no fraudulent journals. If fact I am sure there are. I just do not think that fraudulent journals can publish such a huge amount. My conjecture is that low cost journals have been wrongly classified as fraudulent.

As for peer review, it may be too expensive for this low cost business model. For that matter I have never been impressed by peer review. It is not a necessary condition for a scholarly literature. Perhaps it is a luxury.

What we seem to have here is a rapidly emerging new world of scholarship, which we know little about. I assume it is mostly science and thus it should be properly indexed and made accessible. This includes posting the relevant articles via the Public Access Program. These journals should not be screened out.


__________________________________________________________________

Inside Public Access is published weekly. For subscription information:  http://insidepublicaccess.com/. Single issues may be purchased separately. 

6 comments:

  1. http://www.sciencemag.org/news/2016/12/predatory-publishers-gain-foothold-indian-academia-s-upper-echelon?utm_campaign=news_weekly_2016-12-16&et_rid=17088840&et_cid=1058419

    ReplyDelete
  2. Hi David,
    As I suspected at the time and have since proved, the 400,000 number is very badly wrong. A 100% survey of those journals shows 255,000 for 2014--but only 114,000 in journals on Beall's lists when the study was done. See http://citesandinsights.info/civ17i1.pdf

    ReplyDelete
    Replies
    1. Thanks Walt. However, the difference between 400,000 and 255,000 is insignificant for my purposes.

      Delete
  3. This comment has been removed by the author.

    ReplyDelete
  4. This comment has been removed by the author.

    ReplyDelete
  5. Hi David,
    A few thoughts here:

    1- "Last I knew there were an estimated 30,000 indexed journals"
    Total number of scholarly journals are much more than 30,000 as much as I know. There are more than 54,000 journals registered with Crossref alone
    (https://data.crossref.org/reports/statusReport.html)

    2- "The total number of published articles may be as high as two million"
    The total number of published articles are much more than 2 million yearly as much as I know. You can see 4.2 million articles published in 2016, registered with Crossref here:
    http://api.crossref.org/works?filter=from-pub-date:2016,until-pub-date:2016

    ReplyDelete