This document contains a detailed description of the way that we calculated statistics reported on journalprices.com. The database may be updated at any time with revisions or corrections. Please contact the website owner with any questions.
Our database provides the Title, ISSN (International Standard Serial Number), Publisher, Subject(s), Price Per Article, Price Per Citation, Composite Price Index, the Year First Published, the Profit Status, Relative Price Index, and a Value category for approximately 7000 academic journals. We use the data available in publishers' price lists and the ISI Journal Citation Reports database to calculate these value statistics. The calculation and/or origin of these fields for each journal in the database is described below under "Fields". All journals in the JCR database are potentially including in this database. Journal without a listed ISSN number are excluded. Occasionally journals change names and/or ISSN numbers. For such journals, citation counts may be underestimated. We try to make corrections for such changes if they are called to our attention, but we will not have caught every instance.
Title: The journal title is retrieved from the publisher's price list and from the JCR database.
ISSN: The International Standard Serial Number is retrieved from the JCR database.
Publisher: The journal publisher is retrieved from the JCR database.
Subject: The subject of the journal is one or more of the following list: Agriculture, Biology, Chemistry, Computer Science, Economics, Education, Engineering, Geology, History, Humanities, Law, Mathematics, Medicine, Physics, Psychology, Social Science, and Miscellaneous (Nocat). This categorization represents a blurring of the categorization provided by the JCR database with the intent of making it easy to search for journals in a particular field. Additionally, some journals labeled Nocat in the JCR database were re-categorized by hand when it was clear where they belonged.
Year First Published: This is the year in which the journal was first published. It is retrieved from the Ulrich directory. If Ulrich does not provide this information for a journal, the string "STARTDATEUNKNOWN" is displayed.
Price Per Article: The number of articles published by each journal in the five years 2002-2006 (the most recent years with data available) is retrieved from the JCR database. The price per article is simply the price of this journal for a year's subscription to an academic library (see below under "Calculation of Price" for details) divided by the average number of articles published per year.
Price Per Citation: From the JCR database, we obtain a "recent citation rate", for each journal in 2006. This is the number of times that volumes of a journal published between 2002 and 2006 were cited in 2006, divided by 5. The price per citation is the price of this journal for a year's subscription to an academic library (see below under "Calculation of Price" for details) divided by the recent citation rate.
Composite Price Index: The Composite Price Index (CPI) is the geometric mean of the Price Per Article and the Price Per Citation.
Profit Status: The profit status of the journal. This was determined by hand using various internet resources. Errors should be submitted to the website manager. A few have unknown status, and these default to for-profit for the purposes of calculations since only non-profits contribute to the average non-profit price (see below). They are labeled "unknown" in this category.
Relative Price Index: The relative price index (RPI) is the CPI divided by the average CPI of non-profit journals in the same subject category. Journals that have multiple subject listings are factored into the average CPI for each field it belongs to, and its RPI is its CPI divided by the average of the average CPIs for each field.
Value: The value category is a broad categorization of a journal as "high value" "low value" or intermediate. A journal with an RPI less than 1.25 is classified as "good value", more than 2.0 as "bad value" and everything else as "medium". The cutoff for bad value was reduced from 2.5 to 2.0 in 2009.
Calculation of Price
In most cases, the price of the journal is retrieved from the publisher's price list or the publisher's web site. We use the price quoted for academic libraries located in the United States. Where there are different prices for such options such as a paper edition, online edition, or combined paper and online, we use the price of the cheapest of these options.
The price, if necessary, is converted to United States Dollars using the Currency Converter.
This is the fourth edition of journalprices.com. The first edition was posted in November 2005. It used prices for the year 2004 and article and citation information for the years 1998-2002. The second edition was posted in 2007 and uses prices for the year 2006 and citation information for the years 2000-2004. The third edition was posted in December 2008 and uses prices for 2008 and citation information for the years 2002-2006. The current edition was posted in September 2009 and uses prices for 2009 and citation information for the years 2003-2007.
For the first edition, we used an alternative ISI-published database, "Journal Performance Indicators" rather than "Journal Citations Reports" which we have used in later editions. The JPI, which we used in the first edition, reports recent citations in a different way from the JCR. The "recent citations" data from the JPI data includes all citations regardless of the year in which the citation occurred to articles written in the interval 1998-2002. "Recent citations" as reported by the JCR include only those citations to articles published in the most recent 5 years and cited in the current year. The JCR-based count of recent citations is therefore considerably smaller than the count of recent citations calculated by the JPI method. Consequently, our costs per citation are systematically higher in later editions than in the first edition.
Unfortunately, while there is a JCR for Science and for Social Science, there is not one for the Humanities. Thus we are forced to exclude some humanities journals that were included in our JPI based report.