Abstract
Top-tier journals were examined in terms of the financial contributions of authors, including page charges. Such fees began as early as the 1930s and continue to become more common. A comparison within and between the journals in the Sciences and Social Sciences sections of the Journal Citation Reports indicated that Sciences journals more commonly had page charges. A quantitative examination of page fees may provide some insight into the ongoing debate about author-pays Open Access.
High citation rates and publication in journals with high impact factors are viewed as commendable accomplishments for individual researchers (Schumm, 2010). Another factor likely to affect scholarly publication is author charges. The purpose of this study is to compare page and other charges within and between journals in the Sciences and Social Sciences categories of the Journal Citation Reports. This study is a companion piece to Schumm 's case study of citation rates (2010), in this case characterizing top-tier journals with respect to author-paid fees.
Most scholarly journals are financed by subscriptions—corporate, institutional, and individual. Society journals provide “free” subscriptions to members of professional organizations, but these are actually financed in whole, or part, by the organization's membership dues. Beyond these sources of funding, it is common for journals to apply author-paid page fees and publication fees (submission fees, fees for use of color, figures, exceeding a page limit, or a combination of these). The purpose of this study is to provide some quantitative data on such author-paid fees, as a contribution to the current debate on journal prices and author-paid fees.
The American Physical Society (APS) devised page charges in the 1930s to help finance journals, one of them being Physical Review. The reasoning was that the research published in that journal was not only of interest to those in academia, but also in the broader research community (Barton, 1963; Scheiding, 2009). In the 1960s, scientific research flourished and exceeded the allotted publication space in journals; many not-for-profit professional publications emerged during this decade. In 1965, the United States government authorized researchers with federal grants to pay page charges, in order to support these not-for-profit organizations (Adkinson, 1963). A 1983 National Institutes of Health (NIH) guide allowed payment of page charges from research grants to both profit and not-for-profit organizations (Klingfield, 2005). In 1994, revised guidelines allowed page charges to be paid not only by government-sponsored contributors, but also by non-funded contributors, a guide still followed in the latest (2003) NIH Grants Policy Statement. National Science Foundation grants have a line item for payment of page charges, as well. At the present time, page charges are common in high prestige journals, for example, the journals of the American Geophysical Union, the American Meteorological Society, and the American Statistical Association.
Open Access is a hotly-debated concept for improvement of access to publicly funded research, in which the author pays a one-time base publication cost and then the research report is posted online with no access or subscription fee. The Open Access movement was accelerated by the Public Library of Science (PLOS) launching the journal PLOS One in 2003. Since institutions typically receive some portion of grants for payment of overhead, it seems plausible that Open Access publication costs, like page fees and other author-paid fees, might be straightforwardly borne this way. Such a shift would greatly benefit libraries, since not only would subscriptions be fewer, or even eliminated eventually, but also research articles would be freely available on the Internet, reducing high technology costs for maintaining security of subscription-based access.
There is no question that publishers provide standardized information systems and platforms, consistent backup and information technology (IT) upgrades (IEEE, 2010), and contracted services for subscribers. Industry-wide databases and access management have increased the storage and dissemination of scholarly materials dramatically in recent years. These new, ongoing costs, which provide rich, online access to published materials, have been borne mostly by publishers. In addition, production costs include peer review management, copyediting, proofreading, layout, printing, binding, shipping, and customer service (Benos, Navar, & Reich, 2000). Most publishers also provide online versions of their journals and a permanent, third-party public repository (Oxford, 2010) by contracted services such as Portico and CLOCKSS.
Income from subscriptions varies widely among journals: some topic areas of science are quite specialized, and these journals would be expected to have small circulation numbers, i.e., less revenue from subscriptions, but operating costs are mostly not scalable for smaller publishers. The Bergstrom research on ecology journals showed that in 2005, the average combined print and electronic subscription was $0.29 per page for non-profit journals,1 $0.92 for jointly published journals, and $1.42 for for-profit journals (Bergstrom & Bergstrom, 2006). Among small publishers, and particularly society publishers, author-paid fees have been used to avoid very high subscription costs. Donald O'Shea, long time editor of Optical Engineering, stated: “… why should authors pay page charges? [It] is a professional obligation … page charges allow [the Society] to keep its journal subscription prices low and provide its journal to a large audience” (p. 30101).
Despite these realities of the publication industry, critics (Elliott, 2004; Liesegang, Schachat, & Albert, 2005) claim bias in the author-pays system. Liyange and MacIntrye (2006) claim that “journals are more inclined to publish authors who can afford the article processing charges over those who cannot, thereby neglecting to consider the actual standard of the work being published” (p. 216). The organizational structures (e.g., PLoS One) and charitable activities of publishers do not bear this out, however. For example, most Science, Technology, and Medicine (STM) publishers do not charge fees to authors in developing countries, reduce or waive fees for these underprivileged researchers, and actively pursue initiatives to make research information available at greatly reduced prices or at no cost in developing nations. Some of these initiatives include the International Network for the Availability of Scientific Publications (INASP), Consultative Group on International Agricultural Research (CIGAR), and Health InterNetwork Access to Research Initiative (HINARI; Springer, 2011).
The goal of the present research was to assess page charges and hidden charges and compare them across the disciplines of Natural Sciences, Formal Sciences (including subdivisions of computer science and mathematics), Applied Sciences, and Social Sciences. As a sample, the 10 journals with the highest impact factors in each of 224 categories of the Journal Citation Reports were examined. Since other fees are usually not listed with page charges, the discovery process required navigation through web sites, telephone calls, and e-mails for more information.
Method
Materials
Journal Citation Reports (2006) and the cited journals' online sites accessed through Google were the data sources. The top-10 ranked journals (or if fewer than 10 were in a category, all were used) according to impact factor were selected for each of 169 of the 172 categories of the Sciences section (omitting only three categories of psychology, psychiatry, and behavioral sciences) because the majority of these same journals (18/30) were represented in the Social Sciences section. Likewise, the top 10 ranked journals according to impact factor from all 55 categories of the Social Sciences section were also selected from the Journal Citation Reports.
Procedure
Of the 1,702 journals in the Sciences category, 1,640 journals were examined. Sixty-two journals were excluded: 10 from each of the categories of psychology, psychiatry, and behavioral science (for a total of 30) since the majority of these areas were included in the Social Sciences area; seven journals had changed their titles, four ceased publication, 17 were not in English, two were invitation-only and no information was given about charges, and two were online only. Of the 542 journals in the Social Sciences area, 524 were examined. Eighteen journals were excluded because eight had changed titles, three were not in English, four gave no information about charges, one was online only, and two provided no author instructions.
Obtaining information about page charges often required searching the journals' web sites. Only a few publishers had tables of contents featuring page charges. For most, it was necessary to scroll through pages of author instructions. Fee structures and their descriptors were not consistent across journals. For example, the phrase “requested but not required” in regard to page charges was easy to interpret, but requested or expected was not clear in the context that followed since the wording made it appear they had to be paid by the author. Some stated a fee waiver or partial fee waiver might be available, but the author had to apply; others stated these were available, but only on a limited basis. For some journals, there were discounts for society members; some of these were explicit, but others required logging in as a member to view the discount.
Determining whether or not a journal charged the author for color figures or color pages was even more difficult. Some examples of vague wording were, “a limited number of free color pages” and “editor's discretion for use of color” (Expert Opinion on Therapeutic Patients, 2010) and “if color figures are necessary the Archives may pay part of the expense” (Archives of General Psychiatry, 2010); or “if color is deemed scientifically necessary, there is no charge” (Multiscale Modeling & Simulation, 2006). Further, some journals listed color charge per page (Meteorites & Planetary Science, 2010), while others did so per figure (Endocrine Reviews, 2010), and yet others merely listed an amount that did not distinguish between the two. The charges frequently were not listed in U.S. dollars (Genome Biology: Biology for the Post-Genomic Era), although the instructions to authors were in English. Currency value conversions were done in July 2011.
Journal web sites usually stated that charges for color fees would be provided upon receipt of the manuscript (e.g., Applied Animal Behaviour Science, 2010), so in most cases investigation was required, and frequently only vague information was available. Color charges, thus, were classified as follows. If a color charge was stated, the classification “color charge” was used. Because the price typically varied for the number of pages of color, the amount recorded was the fee for the first page of color. When the publisher said there was a color charge but did not list the amount (e.g., Journal of Biomedical Materials Research Part A), the classification was “color charge, but unstated.” Other publishers said there was no charge for online color (e.g., Biological Cybernetics, 2010), but in these cases there might be a charge for print color; therefore the charges were classified as “color hinted.”
Analysis
Once the data were gathered, they were analyzed within and between the Sciences and Social Sciences categories. Due to the larger number of journals in the Science category, the journals were divided into three areas: natural sciences (n = 552), formal sciences (n = 20), and applied sciences (n = 516). Within the natural sciences, the subdivisions were physical sciences (astronomy, chemistry, earth sciences, environmental sciences, and physics) and life sciences/biology. The formal sciences contained subdivisions of computer sciences and mathematics. The applied sciences included agronomy, engineering, health sciences, interdisciplinary sciences, and materials sciences.
Results
The results show that of the journals in the Sciences category, 83.1% had some type of author-paid charge compared to 38.4% of those in the Social Sciences category. Journals in the Science category more frequently had page charges than those in the Social Sciences category among these “top-tier” journals (Table 1). Twenty-nine percent of Sciences journals had page charges, compared to less than 1% of the Social Sciences journals—among which only 1% had charges for “exceeding the page limit.” Of the categories in the Science area with 10% or more of their journals levying page charges, there tended to be a three-tier grouping. Disciplines with the highest percentage of journals having page charges were computer science and astronomy. 2 Next there was a cluster of three disciplines with intermediate frequency of page fee use: agronomy, engineering, and health sciences. Three more disciplines formed a cluster with the lowest frequency of page fees: biology 12%, mathematics 10% and interdisciplinary 10%. “Exceeds page limit” charges were less common; in the Sciences area, two disciplines had more than 10% of their journals with these charges, computer science and engineering. In the Social Sciences area, less than 1% of journals had page charges, none had a one-time charge, and less than 1% had an “exceeds page limit” charge.
Comparison of page and other charges in science journals and social science journals
Both the Sciences and Social Sciences journals charged more frequently for color (Table 1), probably due to greater printing expenses. Summing the percentages for specific stated color, color but amount not stated, and color hinted, 53% of the Science journals indicated some color charge, compared to 33% of the Social Sciences journals. There were hidden color charges as well, because in both the Science and Social Sciences journals, it was more common to find a color charge with the amount unstated rather than stated. In the Sciences journals, 21% had a color charge while 28% said there was a color charge but did not state it, and another 4% implied a color charge. In the Social Sciences journals, 11% had a color charge while 18% said there was a color charge but did not state the amount, and another 4% implied a color charge. 3
Comparison of page charges in science journals and social science journals
Science categories with 50% or more levying color charges—explicit and implicit
Some journals charged submission fees. The Social Science journals levied more of these than the Science journals. In the Social Sciences journals, about 4% compared to about 1% of the Sciences journals had submission fees. Of the 18 journals in the Social Sciences area with submission charges, eight charges were in the business/finance area. The average submission fee was $143.44 (mode = $25.00; median = $67.50). There was no such trend of being heavily concentrated in any one discipline in the Sciences journals. Submission fees might become more common with online submission of articles, to offset the use of web platforms for manuscript handling and to discourage submissions of questionable provenance. 4
The third division was the Formal Sciences, which had the math and computer science areas. The charges did not follow the same pattern as those journals in the Natural Sciences and Applied Sciences. Fewer than 50% of the mathematics journals had any types of color charges. Of the 20 computer science journals, 5 (25%) hinted color charges, 3 (15%) had color page charges, 2 (10%) had color amount not stated and 1 (.05%) had a color figure charge.
Discussion
This study of the top-tier journals shows that Sciences journals have more frequent page charges than Social Sciences journals. This situation will change with the advent of Open Access digital-only journals. As an example, the PLoS One publication fee is $1350.00, but PLoS has added more online journals with fees up to $2900.00 for PLoS Medicine and PLoS Biology. The Directory of Open Access & Hybrid Journals (DOAJ) listed 7,383 journals with Open Access fees ranging from $650.00 to several thousand dollars. In the first category (agriculture and food sciences), for example, after excluding 26 journals not in English and the 14 missing information, 52% (127/245) of the journals had an Open Access fee. Others may be using society memberships to offset publication costs; the effects of such financial models require further study.
Open Access publication makes articles readily available to anyone with an internet connection, potentially greatly stimulating use of results. Although current legislation does not support specific changes in government granting institutions' policies to give preference to Open Access publication, as they did in the 1960s with journal page fees, such options are being intensely studied. Researchers in the medical and closely related fields may not be adversely affected because they more likely receive grants that have provisions for publication fees. To finance their publications less expensively, authors may turn to less expensive digital-only Open Access journals, despite the fact that there are usually no print versions of such journals.
In conclusion, results show that Sciences journals had more page charges than Social Sciences journals and some charges are hidden. Since all journals could not be examined, the focus was on those journals with top-10 impact factors in each category instead of their citation rates. However, as Schumm (2010) and Seglen (1997) have noted, publication in a top-tier journal does not mean an article will be cited nor does it indicate that an article in a low-tier journal has less impact. Perhaps, the journal impact factor should be replaced by the eigen-factor and specific measures of article impact. More research is certainly needed to clarify how the subscription fees and author-paid charges are related to the financial models of various publishers. This research should raise awareness of author-charges as the trend is for more articles to become available through open access.
Footnotes
1
This may be confounded by the fact that such costs usually are subsidized by society membership dues.
2
Of the 10 top-tier journals in each of the categories of Social Sciences as listed in Journal Citation Reports, 52% of the categories had more journals with for-profit status. Twenty-eight percent (15/54) of the categories had equal numbers of profit and non-profit journals, and 20% (11/54) of the categories had journals that were mostly for-profit. The data suggest that page fees are connected to financial models of society versus non-society or for-profit versus non-profit journals, but more research is needed to establish this point. The tier with the most frequent page fees had the largest number of not-for-profit journals: computer science journals 57% (45/79) and astronomy 90% (9/10). In the tier with medium frequency of page charges, agronomy and engineering, there was an almost equal number of not-for-profit and for-profit journals, and in the health sciences, slightly more had for-profit status. In the group with lowest frequency of page charges, there were slightly more for-profit than nonprofit journals: biology 54%, math 58% (11/19); and interdisciplinary at 56% (97/174). Second, notice the “exceeds page limit” charges. In the Science area, two disciplines had more than 10% of their journals charging for “exceeds page limit.” These were computer science with 27% (21/79) and engineering with 15% (22/143). As for the Social Sciences area, less than 1% (2/524) had page charges, none had a one-time charge, and less than 1% (5/524) had an “exceeds page limit” charge.
3
In the Social Science area, those in which 50% or more journals have color charges of some type are: anthropology, environmental, ergonomics, geography, linguistics, psychology (biological, developmental, education, and multidisciplinary), rehabilitation, and substance abuse. Most of the charges were charge indicated but amount not stated, followed by stated color charges, then color figures, and finally color charge implied. A more detailed analysis was done in the Science area for those disciplines with 50% or more journals levying some type of color charge (
). To do this, the Science journals were assigned to one of three categories. The first category was the Natural Sciences. Of the 552 journals that had a practice of levying some type of color charge, 190 (34%) had color but amount not stated charges, 74 (13%) had color figure charges, 65 (12%) had color page charges to cover the cost of the paper, and 26 (5%) implied color charges. The physics journals had the most color charges with 102 (18%), followed by earth sciences with 90 (16%); life science/biology with 82 (15%), chemistry 42 (8%), environmental 33 (6%), and astronomy 6 (1%). The trend was the same for the second category, Applied Sciences; of the 516 journals that levied some type of color charge, 199 (39%) had color charges but amount not stated, 69 (13%) had color figure charges, 67 (13%) had color page charges, followed by 20 (4%) that implied color charges. The health science journals had the most color charges with 151 (29%) followed by engineering with 83 (16%), materials science with 55 (11%), interdisciplinary with 50 (10%) and agronomy with 16 (3%).
4
E.g., queries submitted by hackers or for nuisance purposes.