We wish to thank Mark Guichard (CNRS-INIST) for his good offices with Thomson Reuters, which enabled us to obtain the indexes in Excel files, Elsevier, which made available on its website the general index of Scopus in an Excel file, Sonia Christon (documentalist, INTD trainee), and Gilles Liévin (computer scientist) for their collaboration
1In 1964, Eugène Garfield, founder of the Institute for Scientific Information in Philadelphia (ISI), created the well-known database, the Science Citation Index (SCI), and in addition few years later, two other databases of citations: the Arts and Humanities Citation Index (AHCI) and the Social Sciences Citation Index (SSCI). The SCI, at the outset intended strictly for the analysis and documentary information of the scientific articles of the sciences of matter and life (SMV), gradually evolved into bibliometry. In the 2000s, E. Garfield sold ISI to Thompson, the Canadian publisher, which in turn bought Reuters, in 2008, and became the Thomson Reuters group. The three databases of citations were consolidated by Thomson Reuters into the Web of Science (WOS).
2If the SCI is a bibliometric instrument that is widely used in the Sciences of matter and life (SMV), and can hardly be overlooked in medicine, in which, in order to be published, researchers must publish in journals inventoried by the SCI, this practice is on the other hand infrequent in the Social Sciences and the Humanities, especially because the AHCI and the SSCI have proved to be very incomplete. Aware of this situation and of the challenge it represents, another commercial publisher (Elsevier) launched the development of Scopus a database of citations that was supposed to have wider coverage. Alongside these private publishers there began to appear international or national lists of journals of reference, for instance in 2007 the ERIH lists produced by the ESF (European Foundation for Science),and in 2008 in France, the lists of the AERES (The Agency for the Evaluation of Research and Higher Education). In order to determine to what degree the different databases now available for the journals in the SSH are representative, in 2008, with the support of TGE Adonis, we undertook to create JournalBase, whose aim is to list the journals indexed in the international bibliometric lists and databases and to compare them. Numerous articles have already attempted comparisons between the WOS and Scopus (for example, Gavel Ylva. Iselid Lars, 2008; Bosman Jeroen. et al, 2006), but no study has yet analysed the collection of available data for the whole field of the SSH.
- 1 Pritchard A., 1969, “Statistical Bibliography or Biblionmetrics”.
3By bibliometry, we mean the application of mathematics and statistical methods to books, articles and other means of communication1. It is a question of measuring the quality of the work of researchers and the influence of the journals in which they publish, based on estimated statistical indicators produced on the basis of citation databases.
4Pressure from institutions involved with putting journals online has sped up the development of these bibliometric tools, and since 2004 has given rise to the elaboration of a whole series of indicators. Several online databases now exist that calculate the impact of a journal, an article or a researcher by means of different indicators based on the number of publications and citations (impact factor, h-index, g-index, hm-index, etc. …). The indicators derived from these databases would contribute to decision-making at the time of the scientific evaluation, along with the evaluation made by the peers. Indeed, a number of academic institutions have begun to use bibliometry for classifying or recruiting researchers.
5The question of the application of this model has come up frequently in the Social Sciences and the Humanities (SSH) throughout the international community over the past few years. An attempt to apply the model was made by the CNRS in France for the four-year reports (2006-2009), when a number of indicators were requested for each researcher and teacher-researcher of a research unit, from the databases of The Web of Science (WOS) or, if need be, from Publish or Perish (PoP – an open access software package that analyses citations from Google Scholar, developed and maintained by Will Ann Harzing since 2007). Indeed, if we take a closer look— unlike the SMV—the WOS with its two citation databases specifically related to the SSH (Social Science Citation Index[SSCI] and Arts and Humanities Citation Index [AHCI ]offers very incomplete coverage of the journals in these disciplines; this lack is particularly evident when it comes to the non-Anglophone journals.
- 2 The GFII/TGE-Adonis report on French scientific publishing in the Social Sciences and the Humaniti (...)
6Furthermore, some studies2 have emphasized the distinctive nature of the modes of distribution of knowledge in SSH, books being able to cover 50% of the scientific output of certain laboratories and certain disciplines. The diversity of research methods and subjects of study, the fact that these are inseparable from the social and linguistic context in which the research is carried out, render impossible any easy and objective international comparisons of the readership of the publications. The application of bibliometric tools as practiced in the hard sciences, based on journals alone, gives only a very incomplete representation of scientific production in SSH, and as a result produces erroneous results on the numbers of publications and citations. Doubtless this is why the CNRS has recommended resorting to the bibliometric tool Publish or Perish (PoP), which inventories a broader diversity of publications (journals and books) from Google Scholar (GS). But this tool still presents major disadvantages, in particular because of the opacity of its sources (there is no list of the journals or publishers inventoried by Google Scholar), and of the risky nature of the results that are entered in the calculation of the indexes.
- 3 Véronique Ginouvès (anthropology), Blandine Nouvel (archeology), Mireille Gely et Martine Chazelas (...)
7Aware of these problems, and anxious to bring concrete facts to the debate, we made a first attempt in 2007, with a group of ten scientific information professionals3 in France, to make a manual list of the journals of the databases of the WOS (AHCI and SSCI) and of Scopus, each one for its respective discipline. The first results (for 8 disciplines, entered in Excel files), revealed numerous problems, in particular a different reading of the sources, and gaps, which in fact indicated the difficulty in collecting precise information based on the WOS and Scopus databases, keeping to the classifications proposed by these two publishers. Even pushing the research very far, if not title by title (which cannot be achieved with several thousand titles), it was not possible to obtain an exhaustive list of the journals inventoried according to the different disciplines. This was due to several problems:
the journals could be listed in several databases of the same editor (for example AHCI and SSCI), either on purpose, or by mistake,
the same database included duplicates, according to the ISSN or the title, with a different ISSN or the same one,
the disciplinary categories or key words associated with the journals did not correspond with the discipline of the journal,
the method followed was only partially systematic, and relied upon a piece of subjectivity that was difficult to avoid because of the heterogeneousness of the indexing
8As a consequence, the results were difficult to utilize, and could not be exhaustive. This experience nonetheless served to expose the inconsistencies in the sources, by demonstrating that the empirical and individual approach by ‘discipline’ was not the best method. It could have been, if the databases had been well indexed, if among the different scientific publishers there had been a veritable consensus about the categories of disciplines, if the information provided had been understood in the same way, if all the ISSNs had been indicated, and presented in the same format, etc.
9Finally, to evaluate these bibliometric tools, it was indispensable to do an exhaustive list of their content, a catalogue that necessitated the establishment of a methodology that was systematic and adapted to the conditions, which handled the totality of the disciplines in the SSH and provided information of equal significance, in order to subject it to the assessment of experts and scientists.
- 4 “Innovating tools and digital processing for enhancing the value of data and for distributing data (...)
10In order to carry out this work, at the end of 2007, we answered the call for projects from the TGE Adonis4 of the CNRS. Our project was accepted, and received financing for 2008 and 2009. Its objective is to make a systematic comparison of the databases (Web of Science, Scopus), and unilateral lists of national and international journals that appeared for the SSH in 2007 and 2008 (ERIH, AERES), in order to evaluate how representative these sources for the journals are according to the disciplines, and to test the quality of their potential use as instruments of bibliometry. It is not a question of providing a validation for these instruments, but of presenting the largest possible number of elements, precisely in order to compare and evaluate them.
- 5 http://www.cybergeo.eu/index22492.html
11The first results5 that we put online in June 2009 in the journal Cybergeo are intended for the international scientific community in SSH, but equally for French, European and international scientific organizations, for experts in scientific Information, for producers of scientific information on a broad scale (commercial or institutional publishers, publishers of journals in open access, and for national and international evaluation authorities). This is why our final objective is to put online in 2010, in a bilingual English/French version, the database of JournalBase in interactive mode on a collaborative platform, so that the decision-makers, the scientists, the experts in scientific information, may have up-to-date information at their disposal, and may contribute to moving the thinking on these questions forward, by sharing their experiences, and by good user practices.
12Since their creation in 1964, the Citation Indexes have stood out as a unique source and reference for citation from articles. Our study is therefore concerned with two of the three citation databases of the Web of Science, which are specific to the SSH: the SSCI (2,012 journals) and the AHCI (1,166 journals), that is, a total of around 3,178 journals. Although in principle they are distinct, and produced by the same institution, these two bases in fact have 314 titles in common.
13The citations database Scopus, launched in 2004 by the Dutch publishing group Elsevier, shook the world monopoly held until then by Thomson Scientific. The ambition of the commercial publisher is to open the way for access to the whole of world institutional scientific literature. The number of journals posted on Scopus website in 2008 is impressive: for the SSH, it is about 6,200 titles (the initial general file inventories 22,688 journals, of which 15,863 active ones for the whole of the sciences). For Scopus as for the WOS, there is a fee for access to these private databases which is negotiated (at the full price) at the level of the management of the research organizations and university libraries. It is therefore important, from this perspective as well, to know its exact coverage.
- 6 HERA : Humanities in the European Research Area.
14In reaction to the hegemony of the WOS, and in the absence of much representation of databases for non-Anglophone SSH journals, the European Science Foundation (ESF), in collaboration with the consortium HERA6, for its part, threw itself into making inventories of the European journals in SSH, with the help of panels of experts, and published in 2007, and again in 2008, lists of journals according to the field of discipline (ERIH) for the Humanities, including around 5,200 different journal titles. Another list for the Social Sciences is under way.
15On the French side, in July 2008, it was the turn of the AERES to publish the lists of SSH journals, relying on the ERIH's lists, others established by the CNRS, and commissions of experts. Around 6,000 different titles are inventoried in this list.
16The ERIH lists, and those of the AERES, are as yet unaccompanied by bibliometric tools, but this does not rule out the possibility that they could be used to advantage in the future.
17It is therefore these five sources that we have proposed evaluating and comparing. We have not integrated the databases deriving from Google Scholar (GS) as for Publish or Perish. Indeed, these tools require a specific methodology of study.
- 7 Michèle Dassa et Christine Kosmopoulos, “JournalBase – a comparative international study of the da (...)
- 8 Classification of the journals in the PDF of JournalBase is not in alphabetical order. It is actua (...)
18On 25 June 2009, we published a first firm version of JournalBase from the end of 2008 in the open access electronic journal Cybergeo7. This file is in pdf, and can be downloaded from the site8. This first stage enabled us to measure the representation of the journals by discipline in each of the databases (AHCI, SSCI, Scopus, ERIH, AERES).
19In order to examine the proportion of journals in open access included in these sources, we made a first update of JournalBase on 17 July 2009 from the journals inventoried in the list of the Directory of Open Access Journals (DOAJ). A this time, the DOAJ inventories in all disciplines mixed together, more than 4,200 active open access scientific journals, of which around 44% are journals in SSH, that is, a little more than 1,600. It can be observed that only a very small percentage of the journals of the DOAJ for the SSH are present in JournalBase, a little over 300 journals, that is, only 0.33% of the total number of journals inventoried in JournalBase. It also indicates that about 78% of the SSH journals presented in the DOAJ are not indexed in anyone of the five studied sources.
20This is not the place to detail the objectives of each of the databases, their procedures for selecting journals, and their uses, which are discussed in other publications (Kosmopoulos et Pumain 2007, Durand-Barthez 2009), but it is evident that these are not without consequence in the results of our analysis.
21Once the whole collection of documents was assembled (WOS/AHCI, WOS/SSCI, Scopus/Social Sciences, ERIH, AERES), in addition to the numerous errors of data entry detected, the principal fact observed was the strong heterogeneity of the different sources, as much in the terms of presentation (the formats of the files were in Excel, in pdf, in html) as in the nature of the information inventoried in each of the databases (the ISSN, the country, publisher, etc., according to the particular case), and in their indexing.
22It is our duty to stress the very unequal quality of the sources. A large number of errors were observed in certain databases: duplicates, triplets, errors in the titles, in the indexing, and correcting them represented a task of several months.
23Most of the journals inventoried by the AERES had no ISSN, and the country where they were published is not specified.
24All told, the fusion of the five databases or lists brought to light more than 20,000 titles that at first sight would appear different, but after the elimination of the duplicates between the databases, there remain in total only 9,551 different journals in SSH that are inventoried in JournalBase.
25The collection of data sources actually included duplicates that we had spotted and removed from the general lists. The biggest surprise came with the examination of the Scopus database. While it posted 6,200 titles for the SSH, that is, twice the number of titles in the Web of Science, once the inactive journals, the duplicates and types of documents other than journals (colloquia, books), have been removed, and only the journals classified in SSH retained, we get 3,435 journals for this database, that is, hardly more than in the WOS. It is to be noted that for the Scopus database, the Humanities are a sub-category of the Social Sciences (see the 4th part).
26By adding the resources of the AHCI and the SSCI and eliminating the duplicates, therefore, the number of journals in SSH that WOS provides seems to be comparable to the number in the Scopus database: 2,864 titles in the first case (after the elimination of the 314 duplicates between the SSCI and the AHCI), 3,453 in the second. However, when each of these databases is analysed title by title, it becomes apparent that these numbers in fact correspond to noticeably different lists of journals.
- 9 Nevertheless, in 2009, Scopus announced the inclusion of several hundred journals in the Humanitie (...)
27Table 1 actually shows that only 15% of the journals inventoried by the AHCI (for the Humanities) are also inventoried in Scopus “Social Sciences”, which also includes almost 78% of those of the SSCI database. In fact, if we consider the whole of the Scopus database, and more particularly the “Social Sciences” section, almost all the journals of the WOS-SSCI (Social Sciences) are found in Scopus, because the overlap is about 97% (96.9%). On the other hand, the coverage is below 20% for the AHCI “Humanities” database (19.4%) This seems to indicate that Scopus has a better coverage in the Social Sciences than in the Humanities, or that there is a wider distribution of the journals in the Humanities9. Moreover, the rate of overlap with the ERIH list is greater for the AHCI than with the SSCI database (71% compared to 39.5%), which is logical, because ERIH is presented as a list of journals of reference in the Humanities. However, the overlap with the SSCI database is not negligible (39.5%). Thus we realize that the differentiation between the “Humanities” and the “Social Sciences” is not perfectly obvious, and that the borders are permeable (see Section 4).
Table 1: Comparison of the WOS journals and their overlap in Scopus and ERIH
28Table 2 indicates that only half of the journals contained in Scopus are also found in the WOS databases, and only roughly 30% in the ERIH lists. We can see clearly that these international databases present information that can hardly be compared to the collection of the journals of the SSH.
Table 2: Comparison of the Scopus journals and their overlap in the WOS and ERIH
29It is of course difficult to establish an idea of the coverage of the field by the databases, or a sense of how representative the bibliometric databases are for the SSH, as we do not have at our disposal a complete reference that would specify how many journals in the world and in each country publish scientific articles in this field of knowledge. What is missing is a denominator to the fraction! However, if we consider that bringing the available databases together already enables us to make a preliminary evaluation of the number of journals that have been indicated, we have no choice but to consider that each of the two big commercial bibliometric databases, WOS and Scopus, only covers a third of the field inventoried elsewhere. The lists of the ERIH and the AERES do a bit better, with each one covering half of the identified field, doing so when they have not yet even completed their plan to list the whole collection of disciplines.
30What remains to be done is to check whether the lists manage to correct one of the weaknesses for which the commercial bibliometric databases are often criticized, which is their over-representation of the Anglophone journals.
31The country where the journals are published is mentioned in the 4 databases of AHCI, SSCI, Scopus and ERIH. Table 3 presents for the AHCI, SSCI and Scopus databases respectively “Social Sciences” (in which are included the Humanities) the distribution in number and percentage of the journals listed by country of publication. The domination of the United States is crushing: half of the journals cited in the WOS are published there (42% for the AHCI, 54% for the SSCI), and the Scopus database does not correct this hegemony, with 42% of the journals being from the United States. In the four databases studied, the United Kingdom is in second place, with almost 20% of the titles in ERIH, 17% in AHCI, almost 25% in SSCI and 28% in Scopus.
Agrandir Original (png, 80k)
Table 3: The Top 10 for the countries publishing the journals in the 4 databases (ERIH, Scopus, AHCI, WOS-SSI)
32If we consider the group “United States and United Kingdom” (Table 4), we confirm that the commercial databases WOS and Scopus inventory a great majority of Anglophone journals, in any case more than 2/3 of the journals, and up to almost 80% for the WOS-SSCI database.
33ERIH, which positions itself as a European tool, does in fact only present 38.4% of journals published in the United Kingdom and the United States. Conversely, it inventories more than 50% of European journals outside the United Kingdom, while only 15% are found in the WOS-SCCI (Table 4).
34Geographic localization does not always make it possible to know the language of publication, which is only indicated for the journals of the ERIH classification. Doubtless the dominance of the publications in English explains the relatively good classification of the Low Countries, which may reach the 2nd position of the countries of Europe and the third position of the global classification in the SSCI and Scopus databases. This good Low Countries classification may also be due to the Dutch nationality of the publisher Elsevier. Thus certain journals published by Elsevier, even in a foreign language, are considered to be Dutch.
Table 4: Distribution of the journals in the 4 databases (ERIH, Scopus, WOS-AHCI, WOS-SSCI) according to the group of the publishing country.
The European Research Space is made up of the 27 countries of the European Union, plus Switzerland, Norway, Lichtenstein, Israël and Iceland.
- 10 Archambault É. Et Vignola-Gagné É., 2004. The Use of Bibliometry in the Social Sciences and the Hu (...)
35The portion of the journals published in France is a little less than 6% in the AHCI database (Table 3) – it is ahead of the Low countries there, but it reaches only 1.19% of the journals of the SSCI and 1.37% of those of Scopus, with 24 and 47 inventoried journals respectively. France is therefore 4th in Europe, after the United Kingdom, the Low Countries and Germany (3rd in the AHCI with 66 journals). The French journals are very clearly under-represented in the databases, since the total number of French scientific journals in SSH is estimated to be in the region of 2,00010. The countries of Southern Europe, Spain and especially Italy, are again much less well represented in each of the databases. On the other hand, a large number of European journals are found in the ERIH lists. For example the number of French journals is 319, as opposed to 47 in Scopus, 66 in WOS-AHCI and 24 in WOS-SSCI (Table 3)
36The ERIH lists, however, only correct imperfectly the bias of the Anglophone domination. In fact, it can be seen that English is the publication language of 45.4% of the journals inventoried in ERIH, that is, a little more than the totality of the journals published in the United States and the United Kingdom (38.4%). The second language is German (7.1%), then French (5.8%), while the German and French journals represent respectively 10.1% and 6.2% of the journals. A number of European journals therefore publish either totally in English, or in several languages including English.
37It is important for researchers to publish in journals that are recognized in their speciality, at once to maintain their research at the best level of the scientific debate, thanks to the systems of committees of readers and evaluation by peers, and also because the institutions that evaluate researchers often choose to set up committees of experts from the same field of discipline. Certainly the generalist journals, or journals placed at the interface between two or several disciplines, are not rare in the Social Sciences and the Humanities. These disciplines were in fact identified very gradually and institutionalized beginning at the end of the nineteenth century. The differences between them became accentuated as theoretical constructions advanced, but the borderlines between them are periodically brought into question, and new labels can emerge, constituted either beginning with an object or a specific methodology (for example, the History of Science), or at the interface between two disciplines (for example, Economic Sociology), or around a theme that carries great potential in society (for example, Information Science and Communication Sciences).
38The indexing of journals in the databases uses classifications and disciplinary designations to make it easier to spot the orientation of a journal, which is not always sensed from its title, and also in order to make various catalogues and counting lists. We will not detail here the very numerous problems raised by indexing, of a technical order, but also epistemological (including in it the cultural and linguistic aspect), well known to all those who are interested in these questions. International comparison adds other problems. Classifications by discipline do not cover the same notions, or the same perimeters, according to the country, thus according to the sources studies. Confusion can seem to diminish when the same term is used, but it is often an error: the same term does not carry the same meaning according to the language, and as a result, the current representation of a discipline in a country can be associated with different thematics from those of other countries. The epistemological difficulty can be multiplied by difficulties resulting from technical choices made by documentalists or publishers of databases. By way of example, Archaeology is listed separately in the WOS, Scopus and ERIH classifications, but it is amalgamated with History in the AERES list. History therefore does not in any way cover the same area in the 4 tools.
- 11 We met representatives from Elsevier in the spring of 2008, at the time of the Salon i-Expo, and i (...)
39The way indexing is organized differs considerably in the databases studied. The ERIH base is in fact a bringing together of 14 lists, one for each disciplinary field. The AERES database is made up of a single list, in which the disciplinary fields are pointed out with key words (up to 4 for the same journal), and lists pick up certain titles again for a given disciplinary field. Indexing is much more complex in the other databases, which use many more key words (up to 199 for Scopus in 200811), either at the same level, or by arranging them hierarchically.
40In the WOS, all the key words are united in the same field, and indexing is multivalued. We had to call a computer scientist to handle this field in order to be able to measure the share of each discipline in the journals of the AHCI and SSCI databases. The journals are indexed with 96 different key words in the AHCI database (2012 journals) and 190 key words for the SSCI (1,116 journals). Each journal can be indexed with 7 key words at most (Table 5).
41In Scopus, on the other hand, each key word occupies a different field, indexing is hierarchical, and there can be up to 14 levels of indexation. The Scopus database thus includes 5 Top Levels. The Top Level Social Sciences (3,453 journals) consists of 14 different levels of indexation and 199 key words. Certain inconsistencies exist in the indexing of the journals, and so some SSH journals are classified in other Top Levels. We have nevertheless decided to stick with the journals that were inventoried in the Social Sciences column (including the Humanities). Our inventorying of the Scopus base cannot therefore be considered perfectly exhaustive.
Table 5: The number of Key Words of the nomenclatures, and the maximum number for a single journal
42In the face of the difficulties of making a comparative presentation of this data, we have been led to develop an adapted methodology. A first stage was to select a common disciplinary nomenclature, avoiding producing a new one. For the SSH, there exists at the very least twenty lists of disciplines (SSCI, AHCI, Scopus, ERIH and AERES, but also those of the CNU, of the National Committee of the CNRS, of the HAL open archives, of the INIST, of the ESF, etc.). We had to choose… After a critical study, we finally opted for the European classification of the ESF, because it presents the advantage of going beyond the national level, while still being quite near to the French specializations. We translated into French and adapted this classification of the ESF to obtain 27 categories (Table 6). This nomenclature is nonetheless not without weaknesses. The categories it uses are not on the same level of scientific object or institution (for example, the category “International Relations” could be integrated into “Political Science”). It is not obvious that the distinction between the category “study of women” and “study of gender” is absolutely pertinent and necessary—and, just as in the previous case, the numerical significance, with fewer than two hundred journals, does not absolutely justify keeping this category in a nomenclature (the authors of the present article, owing to their gender, assume responsibility!) Music could also be grouped with the arts. On the other hand, other fields that come under the Social Sciences and the Humanities are forgotten, like Architecture, Urbanism and Landscaping. But without doubt the most shocking remains the Europeocentrism of the category called “Oriental and African Studies”: even if a patrimony exists in this field, we may doubt the usefulness of retaining such a column that involves fewer than fifty journals. This is why we do not claim that the classification of the journals into the 27 categories of JournalBase provides a ruling on the nature of the disciplines treated in these journals. This classification is to be taken above all as a methodological instrument. It offers the interest of producing a relatively manageable division into categories for the purpose of analyzing how representative the databases are by discipline, in a primarily European framework.
Table 6: the 27 categories of the ESF classification and their total number of journals in the 4 databases after the allocation of key-words to the categories of JournalBase
43We have then divided the 514 key words inventoried in the 5 sources (WOS-AHCI, WOS-SSCI, Scopus, ERIH, AERES) into these 27 categories. Table 7 gives several examples of these correspondences.
Table 7: Example of the allocation of key words to the disciplines
NB: The AERES’ list of journals in literature has not yet been published
44Furthermore, the dates when the tools are updated are not always indicated. This is the case in particular with the AERES, which group we asked to post the dates for updating on its web pages, without success. We still see the date 28/07/2008 (the date of the first online posting), even though we have noticed changes made well after this date.
45The number of journals per discipline is certainly highly dependent on the subtlety of the nomenclatures, but it is of some interest to note that the disciplines that are best represented in the group of 5 databases are in decreasing order: Psychology and Cognitive Sciences, History, Archaeology, Classical Studies, Arts and the History of Art and the Commercial and Administrative Sciences (Tables 6 and 8).
46In all these disciplines of the Humanities (present in ERIH), the AERES and ERIH lists have selected a greater number of scientific journals of reference, except in “Art and History of Art”, in which Scopus has counted more. In ‘Psychology and “Cognitive Sciences”, it is the AERES and Scopus lists that mention the greatest number of journals.
47In its 2009 version, the AERES “History” list contains journals of History and Archeology, but also of Art History. In the data presented in Table 8 it is the AERES lists distributed in 2008 that have been taken into account; the “Art History” category was therefore not indicated for the AERES lists.
Table 8: The Number of journals per discipline and per database
0 = absence of list
48The disciplines for which we note the almost double totals (percentage of named journals at least two times above 50%) are Linguistics, Philosophy and “Theology and Region” (Table 8). We can consider that each of these 3 disciplines presents a core of journals of reference greater than the other disciplines. It also happens that these disciplines are among those that have the largest proportion of journals that are common to all the 5 databases (Table 9). Nevertheless, these proportions remain very weak…Even for these more “consensual”, we observe again that there is a strong distribution of SSH journals.
49Conversely, the disciplines that appear the least homogeneous in terms of recognition and of choice of scientific journals of reference are “Art and History of Art”, and “History and Philosophy of Science”.
Table 9: Number of journals common to the 5 databases by discipline
50The data in Table 10 enable us to refine our analysis as to how geographically representative the databases studied are. We note that the predominance of Anglophone journals in the commercial databases, always greater than those in the ERIH and AERES lists, varies according to the disciplines.
51In the disciplines of the Humanities in the ERIH lists, the categories “Religion and Theology” and Linguistics are much more “European” (and above all “continental”) than in the WOS and Scopus databases. The ERIH list presents for these 2 disciplines more than 70% of the European journals except for the United Kingdom. In Philosophy, in Literature, in Linguistics, the ERIH lists correct markedly the geographical bias of the commercial databases. On the other hand, for Psychology, the Anglophone journals represent more than 80% of the totality of the journals, whatever the database being considered.
52For the Social Sciences (Economics, Geography, Sociology), we note less than a third of continental European journals in all the commercial databases. The discipline most “dominated” by the Anglophone journals, American and English, is Sociology, in which the proportion of “continental” European journals is below 18% in all the databases.
Table 10: the number of journals by geographical space and by discipline
EER : European Research Space
USA + UK : United Kingdom+United States
EER-UK : European Research Space minus UK
EER + USA : European Research Space plus United States
The gray boxes represent the European research space minus the United Kingdom
53As with all the scientific disciplines, the SSH are drawing on more and more digital tools, whether it be for publishing, archiving, documentary research, or the analysis and processing of information that they produce and use. The visibility of scientific output, like the intra- and inter-disciplinary exchanges, are certainly facilitated and amplified by the documentary databases. The temptation is great for certain institutions also to use these for purposes of evaluation, even if this practice remains highly open to criticism, particularly in the area of the SSH. But whether we wish it or not, the bibliometric databases exist. It is therefore important to know their contents well, to be able to clarify the contours and the limits, even evaluate their use for evaluation!
54What emerges from our study is that no tool reflects appropriately the totality of the important journals of reference in SSH. The broadest coverage in terms of the number of journals, and in particular European journals, is without doubt that of ERIH, but ERIH is only concerned with the “Humanities”. The AERES list also offers a much wider range than the big commercial databases. It is worth noting that Scopus has announced the inclusion of several hundred journals in the Humanities in 2009 that are not taken into account in the present study, which was stopped at the end of 2008. An updating of the totality of the data will be carried out in the course of 2010. Nevertheless, from now on, the list established in this article, which we call JournalBase, constitutes the broadest list, because it lists without double counts and with exact identifiers the journals mentioned in each of the other databases.
55For lack of an exhaustive list of existing journals, it is not possible to measure exactly the degree to which the bibliometric databases for the journals of the SSH are representative. We do however obtain very different results according to the disciplines. Obviously, whatever the tool under consideration, the degrees of representation calculated according to the language and the publishing country reveal very strong biases, which reflect not so much gradients of scientific quality as relations of domination in the present world, of a political and cultural, even economic, order.
56These results lead us, therefore, to stress the importance of demonstrating more transparence as to the procedures and criteria of selection of the journals in the different tools, whether or not they are commercial, especially to limit the geographical and disciplinary biases that we have been able to observe in the course of this catalogue of the contents of the databases.
57Our objective from now on is to go forward in updating the 2009 JournalBase, and in the development of a collaborative platform that will include dynamic access to the database, in order to make possible an international watchfulness over this information. The publication in an open access journal that allows interactivity should likewise make it possible to create a space for debate backed by information, and a bibliography validated by the scientific community.