February 16, 2018

Measuring the Value of Journals | Periodicals Price Survey 2014

Journal price data is important for budget management processes, but price alone is not the sole factor determining value. Some metrics, like Impact Factor, have become important in assessing value, and similar value metrics will only increase in importance in the future. The implementation of the Counter 4 during 2014 will expand the availability of usage data from journals, databases, ebooks, and multimedia to support better decision-making. Building upon COUNTER (Counting Online Usage of Networked Electronic Resources) and working with the digital object identifier (DOI) and ORCID (open researcher and contributor ID) identifier, the PIRUS (Publisher and Institutional Repository Usage Statistics) Code of Practice is designed to provide usage data at the individual article level, consolidating usage across platforms.

Many groups are exploring different sets of data and altmetrics that may be potential descriptors of the impact of journals, especially as the immediacy of Twitter and the social web affect scholarly communication. Indicators of the growing importance of altmetrics include the EBSCO purchase of Plum Analytics and the partnership between HighWire Press and Altmetric LLP. Efforts to create altmetrics are still being developed, but better tools to help assess the import of scholarly works would be welcomed by the community. One effort that will speed development is the National Information Standards Organization (NISO) Alternative Assessment Metrics (Altmetrics) Project. The project will study, propose, and develop community-based standards and recommend practices. Clear standards are necessary for altmetrics to move into the mainstream.

Table 4: Comparison of Average Price of Titles
in ISI Indexes by Price to Impact Factor

Price Band No of
% Price
of Latest
Less than $410 2,712 $158 1.9 1.76 0.0053 0.99
Between $410 and 760 1,406 574 6.7 1.47 0.0043 0.83
Between $760 and 1,455 1,515 1,070 6.5 2.21 0.0120 1.04
Between $1,455 and 2,475 1,016 1,898 6.9 2.64 0.0159 1.07
Greater than $2,475 1,258 5,188 5.4 3.33 0.0296 1.30

This year we continue to use the title and publisher data collected for this article to explore the relationship between prices and metrics used to assess journals like Impact Factor, Eigenfactor, and the Article Influence Score. This year, the relationship between serial costs and numbers of citations was also explored, with interesting results.

  • The EIGENFACTOR rates journals according to the number of incoming citations, with citations from highly ranked journals weighted to make a larger contribution to the score than citations from poorly ranked journals. Journals are considered to be influential if they are cited often by other influential journals.
  • The IMPACT FACTOR of a journal is the average number of citations received per paper published in that journal during the two preceding years. The Impact Factor of a journal is calculated by dividing the number of current-year citations to the source items published in that journal during the previous two years.
  • The ARTICLE INFLUENCE SCORE is determined by the average influence of a journal’s articles over the first five years after publication. It is calculated by dividing a journal’s Eigenfactor Score by the number of articles in the journal, normalized as a fraction of all articles in all publications. The mean Article Influence Score is 1.00. A score greater than 1.00 indicates that each article in the journal has above-average influence. A score less than 1.00 indicates that each article in the journal has below-average influence.


The pricing data titles in the merged ISI indexes for 2014 was divided into five price bands; journals priced at less than $410, between $410 and $760, $760–$1,455, $1,455–$2,475, and at more than $2,475. These bands were selected only to be sure that the number of titles in each area is reasonably comparable. The average for Impact Factor, Eigenfactor, and Article Influence Score for all titles in a price range was compared to the averages in the other price bands (Table 4). The Impact Factor and the Eigenfactor tended to show a fairly strong increase with the rise in price. The Article Influence Score did not show a significant increase, with the average for titles in the less than $410 price band showing an average of 1.00 and the most expensive titles showing an average of 1.2.

Higher priced titles do have higher Impact Factors and Eigenfactors, but the increase in the metrics is small when compared to the increase in costs, since the average price ($5,188) for the most expensive journals was 30 times higher than the average price ($158) for the least expensive journals. The increase in prices for the lower cost titles was lower than for the more expensive titles. Article Influence Score did not show a strong correlation between higher scores and prices.

Table 5: Merged ISI Indexes Average Cost per Citation by LC Subject

LJ Article Subject Terms No of Titles Total Cost Total Citations Cost per Cite
Library Science 43 $33,844 26,314 $4.55
Political Science 105 $77,804 66,387 $3.68
Physics 227 $943,220 1,917,182 $3.68
Geography 110 $157,072 279,521 $3.43
Business & Economics 586 $725,461 1,018,691 $2.35
Language & Literature 137 $79,514 52,976 $1.32
Arts & Architecture 19 $15,093 14,601 $1.20
Social Sciences 68 $55,746 42,444 $1.08
History 125 $66,581 53,044 $0.96
General Works 15 $10,432 8,477 $0.90
Agriculture 175 $262,357 802,468 $0.78
Music 9 $4,679 3,638 $0.56
Sociology 328 $299,034 422,543 $0.41
Psychology 197 $176,104 575,429 $0.40
Heath Sciences 1502 $2,276,457 10,534,322 $0.39
General Science 107 $155,235 1,936,701 $0.38
Military & Naval Science 12 $11,097 8,643 $0.38
Biology 581 $1,601,861 6,253,448 $0.37
Zoology 123 $240,456 476,633 $0.37
Recreation 35 $22,875 30,404 $0.35
Education 169 $150,339 151,409 $0.35
Technology 88 $157,058 398,036 $0.28
Anthropology 36 $28,283 57,285 $0.23
Engineering 433 $1,242,351 3,148,792 $0.21
Astronomy 26 $67,367 885,887 $0.21
Philosophy & Religion 30 $18,689 11,040 $0.19
Botany 61 $131,996 397,237 $0.11
Law 110 $51,245 61,852 $0.10
Geology 98 $201,160 600,175 $0.08
Food Science 21 $52,819 111,864 $0.06
Math & Computer Science 229 $398,200 583,454 $0.02
Chemistry 226 $1,080,506 3,662,003 $0.02

The ratio of citations to serials costs is reviewed in Table 5. For STM journals, the average prices tend to be high in comparison to other subjects. This scenario changes if the costs are divided by the numbers of citations for the journals. Chemistry has the highest average price for journals but the lowest cost per citation. Journals in chemistry are heavily cited. If citations are considered an indicator of value, then chemistry journals, despite high average prices, are extremely high-value journals. Conversely, journals in the area of library science are relatively cheap but are infrequently cited, so journals in library science show the highest cost per citation. Based upon this set of data, if cost per citation is reviewed by type of publisher, it is not surprising that commercial publishers have higher per citation costs than other types of publishers. Commercial publishers showed a cost per citation of 48¢, while university presses showed 18¢ and societal publishers showed 9¢.

<< Read “Steps Down the Evolutionary Road”

Stephen Bosch is Materials Budget, Procurement, and Licensing Librarian, University of Arizona Library, Tucson, and Kittie Henderson is Vice President for Academic, Law and Public Library Markets, EBSCO Information Services, Birmingham, AL


This article was published in Library Journal. Subscribe today and save up to 35% off the regular subscription rate.

The Latest Trends in Library Design
Hosted in partnership with Salt Lake County Library and The City Library—at SLCo’s Viridian Center—the newest installment of our library building and design event will let you dig deep with architects, librarians, and vendors to explore building, renovating, and retrofitting spaces to better engage your community.
Diversity and Cultural Competency Training: Collections & RA

Do you want to ensure that your library’s collections are diverse, equitable, inclusive, and well-read?

Do you want to become a more culturally literate librarian and a more effective advocate for your community?

We've developed a foundational online course—with live sessions on February 28 & March 14—that will explore key concepts essential to cultivating and promoting inclusive and equitable collections.


  1. This study is a much-needed start into what I’ve been banging on about for years: altmetrics may be useful in collection development and other library work.

    There are a number of things about this article that don’t quite add up, though–likely due in part to editorial omissions that are critical to understanding.

    How can this information be used?

    So what if Chemistry journals have a lower cost-per-cite. Should that factor into selection decisions? I’d argue no, since citation norms differ from field to field. Comparing citation costs in Poli Sci to Chemistry is misleading; a better way to do it would be to compare costs for fields that have relatively similar cultures/norms around citations.

    >> costs are divided by the numbers of citations for the journals <> Commercial publishers showed a cost per citation of 48¢, while university presses showed 18¢ and societal publishers showed 9¢. <<

    What are the implications of this?

  2. “Measuring the Value of Journals | Periodicals Price Survey 2014” raises numerous relevant questions for librarians as well as those pursuing the grail of universal bibliometrics. The article did not, however, make a compelling argument on any of the fronts identified: journal influence metrics, periodical usage data, or trends in costs and budgets, all of which are subject to the increasing speed of technological change and the uncertainties of the economics of academic and STM publishing.

    The tabular data referred to in the article is puzzling in its presentation. It would seem that “price per use (per Journal or per article)” would be the practical arbiter, not “price per citation,” since it is presumptuous to suggest that library patrons are as a group more interested in the citations within a journal (“highly cited”) than they are with its content (the articles). Both of course are relevant, but to differing degrees depending on the constituent.

    That we now have a surfeit of data and myriad ways of analyzing it is a double-edged sword. It seems logical that the richer datasets so immediately available to libraries and researchers alike would support effective decision making around spending on serial publications, and perhaps that is the case. But to arbitrarily divide a sample into “price bands” of equal size regardless of the subject areas, frequency of publication, reputation, or type of subscriber would seem to limit the utility of the result by trimming the context to such an extreme.

    The Journal Impact Factor is defined too loosely as “the average number of citations received per paper published in that journal during the preceding two years.” No author publishing in any ISI ranked journal would expect to garner the number of citations represented by its Impact Factor. The Impact Factor reflects the journal’s capacity as a whole to generate citations for an “average” article published in it—it is based on the publication, not on the article; it is not an average; it is an indicator of potential based on past performance in the aggregate. Also, because it is temporal, and because there are several other factors at play in its calculation (how “total items” and “source items” are defined; editorial policy; number of articles actually published each year, etc.) it is truly a moving target that can fluctuate or remain relatively stable.

    The complexity of the prevailing economic climate for journals is daunting, and will continue to be of interest so long as research and education employ them. Each of the aspects considered in “Measuring the Value of Journals | Periodicals Price Survey 2014” could be usefully expanded in relation to the possible connections between them, and this is as good a jumping off place as any.