View in EDS HTML Full Text PDF Full Text

The Prestige of Social Scientists in Spain and France: An Examination of Their h-Index Values Using Scopus and Google Scholar

Saved in:
Bibliographic Details
Title: The Prestige of Social Scientists in Spain and France: An Examination of Their h-Index Values Using Scopus and Google Scholar
Language: English
Authors: Dabós, Marcelo P., Gantman, Ernesto R., Fernández Rodríguez, Carlos J. (ORCID 0000-0002-2959-8195)
Source: Minerva: A Review of Science, Learning and Policy. Mar 2019 57(1):47-66.
Availability: Springer. Available from: Springer Nature. 233 Spring Street, New York, NY 10013. Tel: 800-777-4643; Tel: 212-460-1500; Fax: 212-348-4505; e-mail: customerservice@springernature.com; Web site: https://link.springer.com/
Peer Reviewed: Y
Page Count: 20
Publication Date: 2019
Document Type: Journal Articles
Reports - Research
Descriptors: Foreign Countries, Reputation, Social Science Research, Economics, Sociology, Administration, Citations (References), Publications, English, Researchers
Geographic Terms: Spain, France
DOI: 10.1007/s11024-018-9358-0
ISSN: 0026-4695
Abstract: We analyze the prestige of 1,500 scholars in economics, sociology, and management who have Spanish and French institutional affiliations operationalized by their h-index in Scopus and Google Scholar. We use a negative binomial count model to examine how some individual factors affect the h-index from both databases. The results show a non-monotonic relationship between the researchers' career length and their h-index. There is a positive and statistically significant relationship between total research output and the h-index. The share of publications in English over total publications has also a positive and statistically significant effect on the h-index, except in a single case, while the share of publications in other foreign languages does not have such effect. Finally, we found that the effects of the number of citations received by documents in English (international impact) and by those in the vernacular language (local or regional impact) on the h-index vary according to the database, the country, and the discipline in question.
Abstractor: As Provided
Entry Date: 2019
Accession Number: EJ1205281
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwG5XYJNNlPH5Wq-A7o8q8AcAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDHBOxzctJhxKVufvAAIBEICBm8op8neiQSJB8H-7siRUC6Len4fjePcytsDfcyufhggJtKcMfZzxWnTuMlm1dT_3KgeRWXdej3l_s9IMjcuUuZMYdXsfc04YxVt4ZCdQMU8XOkW9rZLZAMmvnM-aOGAf1eefUvZoaoD_VltYVhMaRS3oOeymDJk3KQN6Ubrp7zuPoD5RWN7zB_ziDRBQ59sRi5KPy7S6SkWBfd4V
Text:
  Availability: 1
  Value: <anid>AN0134622793;gr901mar.19;2019Feb12.09:14;v2.2.500</anid> <title id="AN0134622793-1">The Prestige of Social Scientists in Spain and France: An Examination of Their h-Index Values Using Scopus and Google Scholar </title> <p>We analyze the prestige of 1,500 scholars in economics, sociology, and management who have Spanish and French institutional affiliations operationalized by their h-index in Scopus and Google Scholar. We use a negative binomial count model to examine how some individual factors affect the h-index from both databases. The results show a non-monotonic relationship between the researchers' career length and their h-index. There is a positive and statistically significant relationship between total research output and the h-index. The share of publications in English over total publications has also a positive and statistically significant effect on the h-index, except in a single case, while the share of publications in other foreign languages does not have such effect. Finally, we found that the effects of the number of citations received by documents in English (international impact) and by those in the vernacular language (local or regional impact) on the h-index vary according to the database, the country, and the discipline in question.</p> <p>Keywords: Research output; Scholarly recognition; Citation analysis; Social sciences; Spain; France</p> <hd id="AN0134622793-2">Introduction</hd> <p>It has been argued that the social sciences have a national or regional character (Sztompka 2009). Therefore, the corpus of academic documentation that they produce can be largely decoupled from the "international" production indexed by the most widely used scientific databases. This is because the issues addressed by local researchers may not be of interest to an international audience or because the dynamics of intellectual recognition and validation of academic prestige in non-Anglophone countries may be to a large extent conveyed through a national or regional publication circuit in the vernacular language (Hicks 2005; Hanafi 2011; Beigel 2014; Beigel and Salatino 2015; Gantman and Fernández Rodríguez 2016). Moreover, the existence of different styles of "epistemological thought" (Abend 2016) may imply difficulties for translating content from the local circuit of social sciences knowledge to the Anglophone international circuit.</p> <p>Thus, there are indeed two publication circuits or segments for social scientists working in non-Anglophone countries (Gantman and Fernández Rodríguez 2016), the "international" one captured by bibliographic databases like Scopus and the SSCI, and a vernacular one, which is represented by the universe of publication in national languages and is underrepresented in these databases (Hicks 1999; Archambault et al. 2006; Frandsen and Nicolaisen 2008). In this paper, we explore the question of whether the predominance of participation in one of these segments (i.e., the vernacular vs. the "international") renders greater scholarly prestige to the individual researchers.</p> <p>To this end, we analyze the production of social scientists from two countries, Spain and France, which we selected because of our knowledge of their official languages and to the fact that they are leading nations in terms of scientific productivity in social sciences. According to data from SCIMAGOJR (2017), France and Spain occupy, respectively, the 7th and 8th places in research output in social sciences for the period 1996-2015 (and 3rd and 4th places behind Germany and China, if we only take non-Anglophone countries). Within the social sciences, we consider three disciplines: management, economics, and sociology.</p> <p>First, we will assess the correlation of several indicators of scholarly impact and prestige such as the Hirsch (2005) h-index estimated with data from Scopus and Google Scholar (GS). We do this in order to find out if a reduced sample of scientific production, such as that provided by Scopus, which captures the international segment of the scholarly literature and a limited part of the vernacular one, allows to obtain an indicator of the individual researchers' impact with a high level of correlation with the much larger sample provided by GS. The latter is the database that has a larger coverage of both segments and is considered by many authors as the most useful bibliometric tool for the analysis of the productivity and impact in the vernacular segment (Harzing and van der Wal 2008; Bornmann et al. 2016; Orduña-Malea et al. 2016; Halevi et al. 2017). Second, we will examine the level of the scholars' internationalization (i.e., share of published documents in foreign languages, particularly in English) in each discipline and country. Third, we will analyze some of the factors that are associated with the impact and prestige of the researchers in both Scopus and GS. Given the existence of a segmented market in the social sciences literature, based on publication language, we assess to what extent the differences of the individual researchers' prestige (h-index) are associated with their total productivity, the degree of internationalization of their production, and their differential impact in the local and international segments.</p> <p>Among the few antecedents related to these issues, we can mention the work of Delgado-López-Cózar et al. (2014), who used GS to estimate the h-index of a large number of Spanish scholars in the social sciences and humanities. In the French case, Courtault et al. (2010) have made an extensive use of GS to estimate the h-index of individual authors in management and economics. However, the analysis of the relationship between the researchers' impact and the degree of internationalization of their research output in non-Anglophone countries has not received much attention in the literature.</p> <p>The rest of this work is organized as follows. In the following section, we make some precisions about the measurement of scholarly prestige. Then, we describe the procedures used for data collection and the methods used in our analysis. Next, we present the results obtained. Finally, the conclusion summarizes the main findings.</p> <hd id="AN0134622793-3">Academic Prestige and the h-Index</hd> <p>Scholarly prestige and recognition is a construct that cannot be measured with certainty. In their classic study of stratification in science, Cole and Cole (1973) have argued that the number of citations is the most accurate predictor of scientific recognition, showing that this objective indicator, which they deem as tantamount to quality in scientific output, has a stronger association to subjective measures of prestige, which can be obtained from peer opinions, than quantity of scientific output (a measure of productivity).</p> <p>However, there is also evidence that prestige can be associated with organizational and institutional processes like the institutional affiliations of scholars whose prestige is being measured, their career pathways, their position within the scholarly community of their field, etc. (Blau 1976; Youtie et al. 2013)—aspects, like holding membership in journal editorial boards or their departmental affiliation, that in turn have been found to be associated with a larger number of received citations (Allison and Scott Long 1990; Lange and Frensch 2006). Yet the distribution of citations received by different researchers vary greatly, and even Cole and Cole (1971) admit that a scholar who has received <emph>n</emph> citations for a single work should be considered as different to a scholar who has received the same <emph>n</emph> number of citations but evenly distributed over a large number of published documents. Precisely to account for this difference, Hirsch introduced the h-index, which is defined as the h number of papers that have received at least h citations each. This-index can be easily calculated. For instance, if a researcher has five publications, which are ranked in descending order by number of citations (publication 1 has 4 citations, publication 2 has 4 citations, publication 3 has 3 citations, and the rest of the publications have 1 citation), then the corresponding h-index is 3 (the number of citations that is at least equal to the number of documents in descending order). With this procedure, a scholar who has produced five documents, but who has received 500 citations for only one of those documents will have an h-index of 1, regardless of the actual number of received citations. In contrast, a scholar who has also produced five documents and has received 5 citations for each of them (for a total of 25 citations) would have a higher h-index of 5, even when his total citation impact is lower than that of the first scholar. Therefore, by construction, this index has the property that extreme large values of citations are neutralized. The logic of the h-index is that it rewards not only quality (number of citations received) but also quantity (number of documents). Moreover, since both number of citations and number of documents are cumulative indicators, they are presumed to vary directly with career length (more experienced scholars are expected to have a higher h-index than less experienced ones). While the h-index has been criticized on several grounds (Costas and Bordons 2007; Waltman and van Eck 2012), and it cannot be used to compare the performance of scholars from different disciplines unless it has been subject to some form of normalization, its use has become widespread for the purpose of research evaluation, and we consider it as a reasonable proxy for an objective measure of the scientific quality of a scholar's research output.</p> <p>Moreover, the idea of using bibliometric databases to measure prestige may limit the type of prestige that is being measured. There is no objective measure of "international recognition", but rather the measure of prestige is dependent upon the bibliometric database used. If we use Scopus or the SSCI to measure international prestige, we will arrive at a measure that is heavily conditioned by the bias of these databases towards journals published in English. Thus, the production and impact of scholars from non-Anglophone countries may be underrepresented (Etxebarria and Gomez-Uranga 2010). Since we believe that Google Scholar (GS) indexes a large part of the research output in languages other than English, we think that by using this database we arrive at a better, although still imperfect, bibliometric measure of "overall prestige" that takes into account both international and local/regional publication circuits as well as documents that do not have a large presence in international bibliometric databases like books and book chapters.1 [<reflink idref="bib1" id="ref1">1</reflink>] This will allow us to analyze which type of citations (i.e., those to local and those to international publications) contribute the most to increase this index.</p> <hd id="AN0134622793-4">Data and Methods</hd> <p>We constructed our dataset on the individual productivity of social scientists in the following way. We collected data of an initial population from which we selected 1,500 individuals (750 per country and 250 per discipline) by proportional random sampling. Since there is no single open access database about individual researchers in the countries analyzed, we restricted our search to a group of institutions with high scientific productivity in the disciplines of interest (see the list in "Appendix"). For Spain, we selected the top ten universities by their scientific productivity in management, economics and sociology according to the I-UGR (2014) Rankings of Spanish universities. Since the discipline of management is composed of various subdisciplines such as accounting, finance, marketing, operational research, strategy, and others, and considering that some of the top Spanish institutions in this field are business schools not necessarily affiliated with universities, we added the three top Spanish business schools listed in the European ranking of business schools of the Financial Times (2015). Then, we obtained data for all professors in departments of economics, sociology, and management (or its subdisciplines) from the websites of the selected institutions. We excluded part-time or visiting professors, when such information was available. We thus collected a population of 3,728 researchers (1,473 in economics, 1,637 in management, and 618 in sociology).</p> <p>For the French case, lacking a ranking similar to that available for Spain and given that French universities' web pages do not always have complete individual data about faculty composition by department, we used information from the research laboratories of doctoral schools, which do provide the information that we needed and are the main locus for the development of social science research in France. We restricted our selection to those doctoral schools that are totally or partially devoted to management, economics, or sociology and that have been evaluated at least in category A by the AERES (Agence d 'Évaluation de la recherche et de l'enseignement supérieur) between 2009 and 2014.2 [<reflink idref="bib2" id="ref2">2</reflink>] Following this criterion, we selected 33 doctoral schools whose research labs are composed of members from one or more research or higher education institutions. From the research labs that were dedicated to our disciplines of interests, we collected the names of all researchers with rank of Maitre de Conference or higher and full-time dedication to teaching and research (we excluded visiting professors as well as doctoral and postdoctoral fellows). In addition, we included scholars from the three top business schools based in France according to the Financial Times (2015) ranking. We obtained with this procedure a population of 4,707 researchers (1,636 in economics, 2,120 in management, and 951 in sociology).</p> <p>From the populations thus constructed for each country, we selected a sample of researchers (250 for discipline in each country) by proportional random sampling according to the size in number of scientists of each institution in the population. Then, we searched the total research output of each of the 1,500 scholars in our sample using the Publish or Perish 4 program (Harzing 2010). In some cases, it was not possible to obtain reliable information for certain authors due to homonymy, so we randomly selected other researchers belonging to the same institution as replacements. This circumstance was more common in the case of Spanish scholars, who often present problems for the disambiguation of names in scientific databases (Jiménez-Contreras et al. 2002).3 [<reflink idref="bib3" id="ref3">3</reflink>]</p> <p>We collected the following information for each document obtained from the research output of the researchers:</p> <p></p> <ulist> <item> Type of document, which we classified as article, book, book chapter, doctoral thesis, book review, working paper, congress presentation, or other document.</item> <p></p> <item> Publication year.</item> <p></p> <item> Language of publication. In the case of Spain, we distinguished between Spanish or Castilian (the country's main language), dialect (a category that includes the regional languages such as Catalan, Basque, and Galician), English, and "others" as residual category. In France, we distinguished between French, English, and "others".</item> <p></p> <item> Number of citations received.</item> </ulist> <p>For each researcher, we collected the following data:</p> <p></p> <ulist> <item> Career duration, operationalized as the difference between the year of data collection and the year of the researcher's oldest document.</item> <p></p> <item> h-index, provided automatically by the Publish or Perish program based on the publications selected.</item> </ulist> <p>The literature has already pointed out some disadvantages of GS in terms of data quality (Jacsó 2005; Doğan et al. 2016), so an extensive manual work of data cleansing was necessary for the correct identification and allocation of documents to authors. To this end, we resorted to additional web sources to classify some documents by type using, among other sources, the database Dialnet for the Spanish case and the HAL repository for the French case. Among the most frequent problems, we have found numerous duplicates (some documents are repeated more than twice) that we consolidated with the Publish or Perish software. GS may occasionally index documents that are not valid research output (syllabi of courses, erroneous assignations of documents to authors, etc.). We have cleansed all detected problems to the best of our ability, especially regarding data on publication years because errors in this regard may lead to an incorrect estimation of the career length of the researchers. Moreover, it is quite usual, especially in economics, that some documents appear as both working papers and published articles. In these cases, we considered each one as a separate entry because working papers may have some differences with their published versions; but we eliminated duplicates in working papers, which we have found often.4 [<reflink idref="bib4" id="ref4">4</reflink>]</p> <p>In addition, we collected the research output of the researchers in our sample from the Scopus database and obtained the number of documents per author as well as the language, publication year, and number of received citations of each document. We also computed the researchers' Scopus h-index. The data of individual researchers from both Scopus and GS were collected between May 2016 and April 2017.</p> <p>We have used a negative binomial count model to analyze the determinants of the h-index in both Scopus and GS, which is a discrete dependent variable. The independent variables of the regression are the following: the length in years of each researcher's career; its square (because as we will see the relationship is non-monotonic); the degree of internationalization of the researcher's production with two variants: (a) internationalization in English, which we operationalized as the total number of publications (sum of books, articles, and book chapters) in English divided by total number of publications, both from taken from GS, and (b) internationalization of production in languages other than English and the vernacular one (total number of publications in foreign languages other than English divided by total number of publications); total productivity operationalized as the total number of documents obtained in GS; and the number of citations received by documents in English, on the one hand, and the number of citations received by documents in the vernacular language (in the case of Spain, we considered Spanish and the regional languages in this category), on the other one.</p> <hd id="AN0134622793-5">Results</hd> <p>First, we estimate the degree of association between the h-index, as an indicator of the authors' prestige or impact, in Scopus and the h-index obtained from GS data. To this end, we present the Spearman correlation coefficients of both indices in Table 1. In all cases, these impact indicators are positively correlated. However, it can be observed that in sociology, both for France and Spain, the values are not very strong; while the degree of association is stronger in management and economics for both countries. In these two disciplines, the values of the coefficient range from 0.77 to 0.83. We also include in Table 1 the means of the h-index from both GS and Scopus. In all cases, the GS h-index is higher than the Scopus one, due to the larger documentary coverage of the first database.</p> <p>Means of Scopus h-index and GS h-index and Spearman correlation between both indices (by discipline and country)</p> <p> <ephtml> <table frame="hsides" rules="groups"><thead><tr><th align="left">Discipline/Country</th><th align="left">GS h-index</th><th align="left">Scopus h-index</th><th align="left">Spearman correlation</th></tr></thead><tbody><tr><td align="left" colspan="4">Sociology</td></tr><tr><td align="left"> Spain</td><td align="char" char=".">6.49</td><td align="char" char=".">1.45</td><td align="char" char=".">0.601</td></tr><tr><td align="left"> France</td><td align="char" char=".">8.11</td><td align="char" char=".">1.57</td><td align="char" char=".">0.614</td></tr><tr><td align="left" colspan="4">Economics</td></tr><tr><td align="left"> Spain</td><td align="char" char=".">7.98</td><td align="char" char=".">2.24</td><td align="char" char=".">0.765</td></tr><tr><td align="left"> France</td><td align="char" char=".">9.21</td><td align="char" char=".">3.72</td><td align="char" char=".">0.815</td></tr><tr><td align="left" colspan="4">Management</td></tr><tr><td align="left"> Spain</td><td align="char" char=".">5.53</td><td align="char" char=".">2.48</td><td align="char" char=".">0.833</td></tr><tr><td align="left"> France</td><td align="char" char=".">5.96</td><td align="char" char=".">2.09</td><td align="char" char=".">0.776</td></tr></tbody></table> </ephtml> </p> <p>Note: n=250 for each discipline and country</p> <p>In Table 2, we present the bivariate associations of the indicators for total research output and number of citations received in both databases. In all cases, the correlations are positive. Consistently with the results obtained for the h-index, the correlations for total citations received in sociology are much lower (0.58 for France and 0.59 for Spain) than those in management and economics, which range from 0.81 to 0.85. Regarding total production, the correlations range from a minimum of 0.61 for economics in Spain to a maximum of 0.80 also for this discipline in France.</p> <p>Spearman correlations between total documents and total citations in Scopus and GS (by discipline and country)</p> <p> <ephtml> <table frame="hsides" rules="groups"><thead><tr><th align="left">Discipline/Country</th><th align="left">Docs. Scopus/Docs GS</th><th align="left">Cits. Scopus/Cits. GS</th></tr></thead><tbody><tr><td align="left" colspan="3">Sociology</td></tr><tr><td align="left"> Spain</td><td align="char" char=".">0.619</td><td align="char" char=".">0.593</td></tr><tr><td align="left"> France</td><td align="char" char=".">0.687</td><td align="char" char=".">0.575</td></tr><tr><td align="left" colspan="3">Economics</td></tr><tr><td align="left"> Spain</td><td align="char" char=".">0.608</td><td align="char" char=".">0.812</td></tr><tr><td align="left"> France</td><td align="char" char=".">0.803</td><td align="char" char=".">0.846</td></tr><tr><td align="left" colspan="3">Management</td></tr><tr><td align="left"> Spain</td><td align="char" char=".">0.767</td><td align="char" char=".">0.851</td></tr><tr><td align="left"> France</td><td align="char" char=".">0.712</td><td align="char" char=".">0.817</td></tr></tbody></table> </ephtml> </p> <p>Note: n=250 for each discipline and country</p> <p>An important dimension of analysis concerns the level of internationalization of the publications made by the researchers of each country. In Table 3, we provide descriptive data of the share of publications in English and the share of publications in other foreign languages.5 [<reflink idref="bib5" id="ref5">5</reflink>] It can be observed that the level of internationalization in English is more pronounced in economics and management (Spain has a higher level in economics, and France a higher level in management). In contrast, sociology seems a discipline with a low level of internationalization, and this value for Spain is higher than that of France. The level of internationalization in other foreign languages is very low in all fields and countries, suggesting that the more relevant circuits for publications are the vernacular one and the international circuit in English.</p> <p>Means of the internationalization of publications (by discipline and country)</p> <p> <ephtml> <table frame="hsides" rules="groups"><thead><tr><th align="left">Discipline/Country</th><th align="left">Share of publications in English</th><th align="left">Share of publications in other foreign languages</th></tr></thead><tbody><tr><td align="left" colspan="3">Sociology</td></tr><tr><td align="left"> Spain (1)</td><td align="char" char=".">0.164</td><td align="char" char=".">0.015</td></tr><tr><td align="left"> France (1)</td><td align="char" char=".">0.121</td><td align="char" char=".">0.036</td></tr><tr><td align="left" colspan="3">Economics</td></tr><tr><td align="left"> Spain (1)</td><td align="char" char=".">0.453</td><td align="char" char=".">0.004</td></tr><tr><td align="left"> France (2)</td><td align="char" char=".">0.443</td><td align="char" char=".">0.007</td></tr><tr><td align="left" colspan="3">Management</td></tr><tr><td align="left"> Spain (3)</td><td align="char" char=".">0.361</td><td align="char" char=".">0.002</td></tr><tr><td align="left"> France (4)</td><td align="char" char=".">0.397</td><td align="char" char=".">0.014</td></tr></tbody></table> </ephtml> </p> <p>Notes: (<reflink idref="bib1" id="ref6">1</reflink>) n=250, (<reflink idref="bib2" id="ref7">2</reflink>) n=246, (<reflink idref="bib3" id="ref8">3</reflink>) n=244, (<reflink idref="bib4" id="ref9">4</reflink>) n=241</p> <p>Perhaps some of the observed differences between both countries can be explained by the differential characteristics of the research evaluation system of both countries. In Spain, the policies adopted since the introduction of the University Law of 2001 (Ley de Ordenamiento Universitario) have established the importance of publishing articles in journals indexed in international bibliographic databases. This has introduced a "publish or perish" imperative for social scientists of all fields in order to obtain a professorial job or advance in their careers - the so-called ANECA effect (Masip 2011). Hence, the need for these scholars to increasingly publish in international journals in English (Suárez-Ortega et al. 2012; Gantman and Fernández Rodríguez 2013).6 [<reflink idref="bib6" id="ref10">6</reflink>] The pressure to publish journal articles is also present in France (Karpik 2012; Aggeri 2016), particularly after the enactment of the Law of Research 2006-450 (Loi de programme n° 2006-450 pour la recherche). The French legislation has incorporated incentives for rewarding those scholars that publish articles in a set of selected journals. However, Karpik (2012) points out that the evaluation system differs by scientific field. In economics and management (Pontille and Torny 2010), the list of journals contemplates a hierarchical classification in which international journals in English have typically a higher classification rank than those in French. In contrast, the journal list in sociology has no classification ranks; those journals that are considered relevant to the discipline by a group of expert peers are included in the list, regardless of their language, international indexation, and impact factor. So this scheme, which Karpik (2012: 11) denominates the "egalitarian list", does not discriminate against the local publishing circuit and may even facilitate its development by allowing sociologists in France to demonstrate their productivity by publishing articles in any local journal. This helps to explain that the value of internationalization in publications in English for this field in France is lower than that of Spain. Of course, these changes in evaluation systems in France and Spain do not affect the publications of scholars before their enactment - and many of the researchers in our dataset have begun their scholarly careers on an earlier date -, but they might have an impact upon the scholars' publication strategies subsequently to their implementation. In this regard, we can point out that 51% of all books, book chapters and articles in our dataset of sociologists in France were published after 2006, and 85% of the same document types were published by sociologists in Spain after 2001.</p> <p>In order to examine the determinants of the individual prestige of the researchers (h-index), we have used a negative binomial count model. Pearson bivariate correlations (see Online Supplementary Material) do not indicate the existence of multicollinearity problems between the independent variables. Aside from the trivial nearly perfect multicollinearity of career length in years and its squared value, the only high correlation between two independent variables occurs between the indicator of total research output and the number of citations received by documents in the vernacular language (with a maximum of 0.69 for sociology in France and 0.72 for management in Spain), which shows that with data from GS there is a strong association between research output and impact in the vernacular language, an association that is lower when total research output is correlated with the number of citations received by documents in English, with a maximum of 0.59 for economics in France and 0.52 for management in Spain.</p> <p>When the dependent variable is the GS h-index, the results for the non-monotonic pattern of the duration of the researchers' career are qualitatively similar for all disciplines and countries (Tables 4 and 5). In all cases, career length in years has a positive sign and a statistically significant effect, but its square has a negative sign and is also statistically significant. This finding is consistent with the results obtained by Courtault et al. (2010) with a sample of French researchers. The h-index is of a cumulative type and increases with age, provided that the scholar whose index is being measured receives more citations as time goes by; but this logic does necesssarily imply that comparatively younger scholars must have a lower h-index than their senior counterparts do. Our results suggest that the variable h-index increases with career seniority up to a point from which a longer career span decreases the value of this variable (we provide data of the distributions of scholars by career length in the Online Supplementary Material). This finding may indicate that older researchers who have already achieved some level of stability in their academic positions may not be affected by the publish-or-perish imperative.7 [<reflink idref="bib7" id="ref11">7</reflink>] In the Spanish case, the ANECA effect is consistent with this finding, as it has led to an increase in the individual productivity of those scholars who have not yet gained stability in their jobs. However, access to research funding is increasingly becoming contingent on making publications in internationally indexed journals, which suggests that the imperative to publish will also extend to all Spanish faculty ranks (Fernández Esquinas 2016). Moreover, as we have noted earlier, the imperative to publish is also gaining importance in France and may especially affect the younger cohorts of social scientists (Jedlicki and Pudal 2018).</p> <p>Negative binomial regression of GS h-index (Spain)</p> <p> <ephtml> <table frame="hsides" rules="groups"><thead><tr><th align="left" rowspan="2">Variable</th><th align="left" colspan="2">Sociology</th><th align="left" colspan="2">Economics</th><th align="left" colspan="2">Management</th></tr><tr><th align="left">Model 1</th><th align="left">Model 2</th><th align="left">Model 1</th><th align="left">Model 2</th><th align="left">Model 1</th><th align="left">Model 2</th></tr></thead><tbody><tr><td align="left">Constant</td><td align="left">0.16607 (0.17365)</td><td align="left">0.17255 (0.17356)</td><td align="left">0.21990 (0.16448)</td><td align="left">0.22497 (0.16418)</td><td align="left">−0.16125 (0.20073)</td><td align="left">−0.15458 (0.20120)</td></tr><tr><td align="left">Career length</td><td align="left">0.08076**** (0.01405)</td><td align="left">0.07966**** (0.01406)</td><td align="left">0.07328**** (0.01244)</td><td align="left">0.07260 (0.01244)</td><td align="left">0.07403**** (0.01513)</td><td align="left">0.07339**** (0.01520)</td></tr><tr><td align="left">Career length sq.</td><td align="left">−0.00137**** (0.00028)</td><td align="left">−0.00135**** (0.00028)</td><td align="left">−0.00126**** (0.00025)</td><td align="left">−0.00125**** (0.00025)</td><td align="left">−0.00135*** (0.00030)</td><td align="left">−0.00134*** (0.00030)</td></tr><tr><td align="left">Tot Docs.</td><td align="left">0.00734**** (0.00087)</td><td align="left">0.00725**** (0.00087)</td><td align="left">0.00722**** (0.00094)</td><td align="left">0.00714**** (0.00093)</td><td align="left">0.00894**** (0.00136)</td><td align="left">0.00891**** (0.00136)</td></tr><tr><td align="left">Pubs in Eng. (share)</td><td align="left">1.24086**** (0.15781)</td><td align="left">1.21514**** (0.15902)</td><td align="left">0.84637**** (0.09904)</td><td align="left">0.84636**** (0.09878)</td><td align="left">1.14270**** (0.13208)</td><td align="left">1.13928**** (0.13227)</td></tr><tr><td align="left">Pubs other lang. (share)</td><td align="left" /><td align="left">0.88086 (0.72415)</td><td align="left" /><td align="left">1.23372 (1.53968)</td><td align="left" /><td align="left">0.78671 (1.93856)</td></tr><tr><td align="left">Cits. docs. Eng.</td><td align="left">0.00016 (0.00010)</td><td align="left">0.00016 (0.00010)</td><td align="left">0.00011**** (0.00003)</td><td align="left">0.00011**** (0.00003)</td><td align="left">0.00014*** (0.00005)</td><td align="left">0.00015*** (0.00005)</td></tr><tr><td align="left">Cits. docs. vern.</td><td align="left">0.00020*** (0.00006)</td><td align="left">0.00021**** (0.00006)</td><td align="left">0.00047**** (0.00013)</td><td align="left">0.00048**** (0.00013)</td><td align="left">0.00124**** (0.00032)</td><td align="left">0.00125**** (0.00032)</td></tr><tr><td align="left">Alpha</td><td align="left">0.02504 (0.01369)</td><td align="left">0.02469 (0.01362)</td><td align="left">0.03273 (0.01246)</td><td align="left">0.03169 (0.01243)</td><td align="left">0.08123 (0.02131)</td><td align="left">0.08099 (0.02130)</td></tr><tr><td align="left">Pseudo R<sup>2</sup> (Cragg & Uhler)</td><td align="left">0.690</td><td align="left">0.692</td><td align="left">0.710</td><td align="left">0.710</td><td align="left">0.691</td><td align="left">0.691</td></tr><tr><td align="left">Log likelihood</td><td align="left">−564.14832</td><td align="left">−563.42335</td><td align="left">−598.53959</td><td align="left">−598.22625</td><td align="left">−539.11684</td><td align="left">−539.03607</td></tr><tr><td align="left">Obs.</td><td align="left">250</td><td align="left">250</td><td align="left">250</td><td align="left">250</td><td align="left">244</td><td align="left">244</td></tr></tbody></table> </ephtml> </p> <p>Notes: **** p<0.0001 ***p<0.001 ** p<0.01 *p<0.05, standard errors in parentheses</p> <p>Negative binomial regression of GS h-index (France)</p> <p> <ephtml> <table frame="hsides" rules="groups"><thead><tr><th align="left" rowspan="2">Variable</th><th align="left" colspan="2">Sociology</th><th align="left" colspan="2">Economics</th><th align="left" colspan="2">Management</th></tr><tr><th align="left">Model 1</th><th align="left">Model 2</th><th align="left">Model 1</th><th align="left" /><th align="left">Model 1</th><th align="left">Model 2</th></tr></thead><tbody><tr><td align="left">Constant</td><td align="left">0.53690 (0.16124)</td><td align="left">0.53960 (0.16142)</td><td align="left">0.79546 (0.17307)</td><td align="left">0.79344 (0.17326)</td><td align="left">0.10118 (0.16139)</td><td align="left">0.09850 (0.16108)</td></tr><tr><td align="left">Career length</td><td align="left">0.07298**** (.01290)</td><td align="left">0.07252**** (0.01298)</td><td align="left">0.07414**** (0.01270)</td><td align="left">0.07453**** (0.01279)</td><td align="left">0.0897**** (0.01312)</td><td align="left">0.09017**** (0.01309)</td></tr><tr><td align="left">Career length sq.</td><td align="left">−0.0009**** (0.00023)</td><td align="left">−0.00093**** (0.00023)</td><td align="left">−0.00120**** (0.00023)</td><td align="left">−0.00121**** (0.00023)</td><td align="left">−0.00148**** (0.00026)</td><td align="left">−0.00148**** (0.00025)</td></tr><tr><td align="left">Tot Docs.</td><td align="left">0.00305**** (0.00056)</td><td align="left">0.00306**** (0.00057)</td><td align="left">0.00468**** (0.00073)</td><td align="left">0.00462**** (0.00076)</td><td align="left">0.00556**** (0.00119)</td><td align="left">0.00555**** (0.00118)</td></tr><tr><td align="left">Pubs in Eng. (share)</td><td align="left">0.56077**** (0.20322)</td><td align="left">0.56064*** (0.20320)</td><td align="left">0.02302 (0.10156)</td><td align="left">0.01932 (0.10251)</td><td align="left">0.58150**** (0.11184)</td><td align="left">0.58705**** (0.11153)</td></tr><tr><td align="left">Pubs other lang. (share)</td><td align="left" /><td align="left">0.11805 (0.36967)</td><td align="left" /><td align="left">0.30196 (1.13095)</td><td align="left" /><td align="left">−0.48623 (0.60262)</td></tr><tr><td align="left">Cits. docs. Eng.</td><td align="left">−0.00005 (0.00005)</td><td align="left">−0.00005 (0.00005)</td><td align="left">0.00018**** (0.00003)</td><td align="left">0.00018**** (0.00003)</td><td align="left">0.00052**** (0.00012)</td><td align="left">0.00051**** (0.00012)</td></tr><tr><td align="left">Cits. docs. vern.</td><td align="left">0.00026**** (0.00004)</td><td align="left">0.00026**** (0.00004)</td><td align="left">0.00005 (0.00004)</td><td align="left">0.00005 (0.00004)</td><td align="left">0.00016**** (0.00003)</td><td align="left">0.00016**** (0.00003)</td></tr><tr><td align="left">Alpha</td><td align="left">0.01946 (0.0119)</td><td align="left">0.01942 (0.01192)</td><td align="left">0.10125 (0.01814)</td><td align="left">0.10117 (0.01813)</td><td align="left">0.04878 (0.01737)</td><td align="left">0.04719 (0.01731)</td></tr><tr><td align="left">Pseudo R<sup>2</sup> (Cragg & Uhler)</td><td align="left">0.728</td><td align="left">0.728</td><td align="left">0.664</td><td align="left">0.664</td><td align="left">0.697</td><td align="left">0.698</td></tr><tr><td align="left">Log likelihood</td><td align="left">−596.70496</td><td align="left">−596.65432</td><td align="left">−651.90774</td><td align="left">−651.87214</td><td align="left">−536.70833</td><td align="left">−536.36795</td></tr><tr><td align="left">Obs.</td><td align="left">250</td><td align="left">250</td><td align="left">246</td><td align="left">246</td><td align="left">241</td><td align="left">241</td></tr></tbody></table> </ephtml> </p> <p>Notes: **** p<0.0001 ***p<0.001 ** p<0.01 *p<0.05, standard errors in parentheses</p> <p>It is also interesting to point out that there is a moderate negative correlation between career length and the degree of internationalization in English in all cases (see Online Supplementary Material). More senior researchers, then, may have less interest in publishing in English than their less experienced colleagues. In addition, in all cases, the positive correlation between career duration and impact in terms of received citations is greater for citations received by documents in the vernacular language than for those received by documents in English, which suggests that more senior researchers may have a larger impact with a local audience rather than at an international level.</p> <p>Another regularity detected is that total research output in GS is associated with a positive and statistically significant effect to the h-index, which is quite logical since this index is assumed to be positively associated to individual productivity. Moreover, the degree of internationalization in English has also a positively and statistically significant impact on the dependent variable, except in the case of economics in France. In order to assess whether the international opening of the researchers' published output to non-Anglophone countries also implies a higher level of prestige for the authors, we have included a second model that incorporates the indicator of the share of publications in foreign languages other than English over the total number of publications. In all cases, this indicator lacks statistical significance.</p> <p>Finally, we also included two impact indicators to determine if they differentially affect the h-index, citations received by documents in English and citations received by documents in the vernacular language. In the case of Spain, both indicators are associated positively and in a statistically significant way with the GS h-index, except in sociology for which citations in English lacks a statiscally significant effect. In management and economics, the indicator of citations received by documents in the vernacular language is larger than that of citations received by documents in English language. This finding indicates that in Spain, even in more internationalized disciplines like economics and management, citations in the domestic publication circuit seem more important for overall prestige than those obtained in the international literature segment.</p> <p>In the case of France, citations to documents in English are positively associated with a statistically significant effect to the h-index in the fields of management and economics, while citations to documents in French have a positive and statististically significant effect only in sociology and management. In economics, only citations to documents in English have this effect in the prestige variable, which is not surprising because economics is a more internationally-oriented discipline. In contrast, in a more locally-oriented field like sociology, only citations in the vernacular language show a positive, statistically significant effect.</p> <p>Tables 6 and 7 present the results of similar regressions with the Scopus h-index as dependent variable. The non-monotonic pattern for career length is verified in all disciplines and countries. Total research output also affects positively and in a statistically significantly way the dependent variable in all cases. The degree of internationalization of production has in all cases a positive and statistically significant association with the Scopus h-index, which is expected since this database collects mostly articles in English. As in the case of the GS h-index, the level of internationalization of publications in other foreign languages lacks statistical significance, except for sociology for France, but in this case the indicator has a negative effect on the h-index. The indicator of citations to documents/publications in English is positively and statistically significant associated with the Scopus h-index in all cases, except sociology in France, possibly because Scopus covers mostly articles in French for this discipline (72% of articles in Scopus are in French, against 26% of articles in English in our dataset). In Spain, the indicator for citations to documents in the vernacular language only has a positive and statistically significant effect in management. In France, this indicator has a positive and statistically significant effect in management and sociology, which again may be explained by the fact that most of the sociology articles in Scopus are published in French. In contrast, citations to documents in the vernacular language have a statistically significant but negative effect on the h-index for economics in France. The cases of both Spain and France suggest that in a highly internationalized social science like economics, the citations received by publications in the vernacular language (an indicator of local prestige) do not have a positive influence for the level of international prestige, as measured by the h-index obtained with data from an international database like Scopus.</p> <p>Negative binomial regression of Scopus h-index (Spain)</p> <p> <ephtml> <table frame="hsides" rules="groups"><thead><tr><th align="left" rowspan="2">Variable</th><th align="left" colspan="2">Sociology</th><th align="left" colspan="2">Economics</th><th align="left" colspan="2">Management</th></tr><tr><th align="left">Model 1</th><th align="left">Model 2</th><th align="left">Model 1</th><th align="left">Model 2</th><th align="left">Model 1</th><th align="left">Model 2</th></tr></thead><tbody><tr><td align="left">Constant</td><td align="left">−1.32661 (0.32999)</td><td align="left">−1.32561 (0.32994)</td><td align="left">−1.81885 (0.27356)</td><td align="left">−1.81617 (0.27319)</td><td align="left">−1.54418 (0.35307)</td><td align="left">−1.56897 (0.3548)</td></tr><tr><td align="left">Career length</td><td align="left">0.06350* (0.02958)</td><td align="left">0.06305* (0.02960)</td><td align="left">0.10717**** (0.02048)</td><td align="left">0.10641**** (0.02050)</td><td align="left">0.06121* (0.02580)</td><td align="left">0.06367* (0.02596)</td></tr><tr><td align="left">Career length sq.</td><td align="left">−0.00170* (0.00066)</td><td align="left">−0.00170* (0.00066)</td><td align="left">−0.00217**** (0.00042)</td><td align="left">−0.00217**** (0.00042)</td><td align="left">−0.00123* (0.00052)</td><td align="left">−0.00127* (0.00052)</td></tr><tr><td align="left">Tot Docs.</td><td align="left">0.00752**** (0.00176)</td><td align="left">0.00747**** (0.00175)</td><td align="left">0.01281**** (0.00152)</td><td align="left">0.01270**** (0.00153)</td><td align="left">0.01272**** (0.00265)</td><td align="left">0.01286**** (0.00265)</td></tr><tr><td align="left">Pubs in Eng. (share)</td><td align="left">2.90481**** (0.25721)</td><td align="left">2.89134**** (0.25830)</td><td align="left">2.00102**** (0.16986)</td><td align="left">2.00413**** (0.16975)</td><td align="left">2.19577**** (0.24733)</td><td align="left">2.20940**** (0.24806)</td></tr><tr><td align="left">Pubs other lang. (share)</td><td align="left" /><td align="left">0.54697 (1.32284)</td><td align="left" /><td align="left">1.30760 (2.36843)</td><td align="left" /><td align="left">−4.12482 (4.69185)</td></tr><tr><td align="left">Cits. docs. Eng.</td><td align="left">0.00044*** (0.00014)</td><td align="left">0.00044*** (0.00014)</td><td align="left">0.00007<sup>+</sup> (0.00004)</td><td align="left">0.00007<sup>+</sup> (0.00004)</td><td align="left">0.00026* (0.00011)</td><td align="left">0.00025* (0.00011)</td></tr><tr><td align="left">Cits. docs. vern.</td><td align="left">0.00017 (0.00013)</td><td align="left">0.00017 (0.00013)</td><td align="left">−0.00018 (0.00024)</td><td align="left">−0.00017 (0.00024)</td><td align="left">0.00118<sup>+</sup> (0.00061)</td><td align="left">0.00115<sup>+</sup> (0.00061)</td></tr><tr><td align="left">Alpha</td><td align="left">0.00859 (0.04424)</td><td align="left">0.00733 (0.04448)</td><td align="left">0.06303 (0.03127)</td><td align="left">0.06162 (0.03127)</td><td align="left">0.30196 (0.06684)</td><td align="left">0.30105 (0.06663)</td></tr><tr><td align="left">Pseudo R<sup>2</sup> (Cragg & Uhler)</td><td align="left">0.578</td><td align="left">0.578</td><td align="left">0.674</td><td align="left">0.675</td><td align="left">0.588</td><td align="left">0.590</td></tr><tr><td align="left">Log likelihood</td><td align="left">−312.15713</td><td align="left">−312.07335</td><td align="left">−440.82468</td><td align="left">−440.67586</td><td align="left">−403.72277</td><td align="left">−400.42675</td></tr><tr><td align="left">Obs.</td><td align="left">250</td><td align="left">250</td><td align="left">250</td><td align="left">250</td><td align="left">244</td><td align="left">244</td></tr></tbody></table> </ephtml> </p> <p>Notes: **** p<0.0001 ***p<0.001 ** p<0.01 *p<0.05 <sups>+</sups>p<0.10, standard errors in parentheses</p> <p>Negative binomial regression of Scopus h-index (France)</p> <p> <ephtml> <table frame="hsides" rules="groups"><thead><tr><th align="left" rowspan="2">Variable</th><th align="left" colspan="2">Sociology</th><th align="left" colspan="2">Economics</th><th align="left" colspan="2">Management</th></tr><tr><th align="left">Model 1</th><th align="left">Model 2</th><th align="left">Model 1</th><th align="left">Model 2</th><th align="left">Model 1</th><th align="left">Model 2</th></tr></thead><tbody><tr><td align="left">Constant</td><td align="left">−1.14633 (0.34005)</td><td align="left">−1.18388 (0.33997)</td><td align="left">−0.72384 (0.25905)</td><td align="left">−0.71432 (0.25885)</td><td align="left">−1.60919 (0.30420)</td><td align="left">−1.60912 (0.30416)</td></tr><tr><td align="left">Career length</td><td align="left">0.08064*** (0.02800)</td><td align="left">0.08749*** (0.02812)</td><td align="left">0.12140**** (0.01960)</td><td align="left">0.12009**** (0.01966)</td><td align="left">0.07489*** (0.02379)</td><td align="left">0.07495*** (0.02380)</td></tr><tr><td align="left">Career length sq.</td><td align="left">−0.00155*** (0.00053)</td><td align="left">−0.00164*** (0.00053)</td><td align="left">−0.00236**** (0.00037)</td><td align="left">−0.00234**** (0.00038)</td><td align="left">−0.00131*** (0.00047)</td><td align="left">−0.00131*** (0.00046)</td></tr><tr><td align="left">Tot Docs.</td><td align="left">0.00557**** (0.00106)</td><td align="left">0.00537**** (0.00105)</td><td align="left">0.00559**** (0.00105)</td><td align="left">0.00576**** (0.00108)</td><td align="left">0.00859**** (0.00243)</td><td align="left">0.00859**** (0.00243)</td></tr><tr><td align="left">Pubs in Eng. (share)</td><td align="left">1.96246**** (0.34157)</td><td align="left">1.97017**** (0.33807)</td><td align="left">0.42357*** (0.13979)</td><td align="left">0.43601*** (0.14074)</td><td align="left">1.94019**** (0.20499)</td><td align="left">1.94049*** (0.20497)</td></tr><tr><td align="left">Pubs other lang. (share)</td><td align="left" /><td align="left">−2.16428* (1.06509)</td><td align="left" /><td align="left">−1.12752 (1.67407)</td><td align="left" /><td align="left">−0.07001 (1.07593)</td></tr><tr><td align="left">Cits. docs. Eng.</td><td align="left">−0.00007 (0.00011)</td><td align="left">−0.00007 (0.00011)</td><td align="left">0.00027**** (0.00004)</td><td align="left">0.00027**** (0.00004)</td><td align="left">0.00017*** (0.00005)</td><td align="left">0.00017*** (0.00006)</td></tr><tr><td align="left">Cits. docs. vern.</td><td align="left">0.00015* (0.00008)</td><td align="left">0.00019* (0.00008)</td><td align="left">−0.00016* (0.00007)</td><td align="left">−0.00017* (0.00007)</td><td align="left">0.00042<sup>+</sup> (0.00023)</td><td align="left">0.00042<sup>+</sup> (0.00023)</td></tr><tr><td align="left">Alpha</td><td align="left">0.03735 (0.05459)</td><td align="left">0.03105 (0.05345)</td><td align="left">0.14286 (0.03450)</td><td align="left">0.14204 (0.03444)</td><td align="left">0.16245 (0.06550)</td><td align="left">0.16214 (0.06563)</td></tr><tr><td align="left">Pseudo R<sup>2</sup> (Cragg & Uhler)</td><td align="left">0.383</td><td align="left">0.396</td><td align="left">0.547</td><td align="left">0.547</td><td align="left">0.571</td><td align="left">0.571</td></tr><tr><td align="left">Log likelihood</td><td align="left">−363.99698</td><td align="left">−361.61738</td><td align="left">−500.64534</td><td align="left">−500.41463</td><td align="left">−374.15824</td><td align="left">−374.1561</td></tr><tr><td align="left">Obs.</td><td align="left">250</td><td align="left">250</td><td align="left">246</td><td align="left">246</td><td align="left">241</td><td align="left">241</td></tr></tbody></table> </ephtml> </p> <p>Notes: **** p<0.0001 ***p<0.001 ** p<0.01 *p<0.05 <sups>+</sups>p<0.10, standard errors in parentheses</p> <p>Finally, it should be noted that all models presented have a good fit to data, judging from the Cragg and Uhler pseudo R<sups>2</sups> values, except in the case of sociology in France with the Scopus h-index for which this coefficient is just 0.38 in model 1. Moreover, and with the aforementioned exception, model 1 has in all cases better fit than model 2.</p> <p>To sum up, in Spain, and considering the GS h-index, which we take as a more accurate measure of a scholars' overall prestige in the social sciences, both citations in English and in the vernacular languages have a statistically significant impact on the h-index, except citations in English in sociology; but the coefficient for this latter variable is lower in all disciplines. In France, citations in the vernacular language have a positive impact on the h-index in sociology and management, while citations in English have a positive statistically significant impact on the h-index in management and economics. So, Hirsch's hypothesis that citations in English have a larger impact than citations in other languages (Hirsch and Buela-Casal 2014) does not hold in sociology in both countries and in management and economics in Spain, possibly because GS is used to estimate the h-index. In contrast, when the prestige of the scholars is measured with the h-index elaborated from Scopus, we obtain different results. For Spain, in sociology and economics only citations in English have a positive and statistically significant impact on the h-index; while in management both types of citations have a positive and significant impact, although the coefficient is higher for citations in the vernacular languages. In the case of France, in economics, only citations in English have a positive, statistically significant impact; in management, both types of citations have such impact; while in sociology, only citations in the vernacular language achieve this effect, which is not surprising because in our dataset Scopus has a large coverage of French articles in sociology.</p> <hd id="AN0134622793-6">Conclusions</hd> <p>Our findings show a positive association between the Scopus h-index and the GS h-index, which is much larger in economics and management than in sociology. Identical results are obtained with the Spearman correlations of total citations received in both databases. We also find that in terms of the level of internationalization of each discipline (average share of publications in English per researcher), sociology is a more locally-oriented field than management and economics in both countries. Regarding the determinants of individual prestige, measured by the h-index from Scopus and GS data, the results show a non-monotonic pattern for the effect of the authors' career length. The h-index first increases with seniority up to a certain level, and then begins to decrease with further increments in career duration. This could be explained by the existence of different patterns for the researchers' productivity. Researchers with lower seniority may be subject to more institutional pressures to publish than those who have already gained some stability in their academic positions. Moreover, total research output in GS is in all cases associated positively with their h-index. Another regularity detected is that the degree of internationalization for publications in English language is positively associated with the h-index, except in the case of economics in France with the GS h-index. In contrast, the degree of internationalization of publications in other foreign languages does not have a positive and statistically significant effect on the h-index.</p> <p>The patterns of association between the h-index and the number of citations received by documents in English (international impact) and by documents in the vernacular language (local or regional impact) vary according to the database, the country, and the discipline in question. So, even considering disciplines with similar objects of study, there is no homogeneity in the differential effect of local and international impact on the individual prestige as measured by the Scopus and GS h-index. However, if having a better bibliometric measure of individual prestige is the desired goal, then the GS h-index should be preferred. Since GS has a quite extensive coverage of both citable and citing documents in English and especially in vernacular languages, including book chapters and books, which in fields like sociology may report more citations than articles (Clemens et al. 1995), the GS h-index is higher than that obtained from Scopus data and reflects more accurately the scholarly impact of social scientists in non-Anglophone countries. The fact that, with the exception of management and economics in France, citations in the vernacular languages do have a higher impact upon the GS h-index than those in English shows the marked relevance of the local/regional publication circuit for the scholars in the fields under study.</p> <hd id="AN0134622793-7">Electronic supplementary material</hd> <p>The online version of this article (10.1007/s11024-018-9358-0) contains supplementary material, which is available to authorized users.</p> <hd id="AN0134622793-8">References</hd> <ref id="AN0134622793-9"> <title> References </title> <blist> <bibl id="bib1" idref="ref1" type="bt">1</bibl> <bibtext> Abend Gabriel, Styles of sociological thought: Sociologies, epistemologies, and the Mexican and U.S. quests for truth, Sociological Theory, 2016, 24, 1, 1, 41</bibtext> </blist> <blist> <bibl id="bib2" idref="ref2" type="bt">2</bibl> <bibtext> Aggeri Franck, L'obsession de la productivité et la fabrique du chercheur publiant, Le Libellio d'Aegis, 2016, 12, 2, 21, 32</bibtext> </blist> <blist> <bibl id="bib3" idref="ref3" type="bt">3</bibl> <bibtext> Allison Paul, Scott Long J, Departmental effects on scientific productivity, American Sociological Review, 1990, 55, 469, 478</bibtext> </blist> <blist> <bibl id="bib4" idref="ref4" type="bt">4</bibl> <bibtext> Archambault Eric, Vignola-Gagne Etienne, Cote Grégoire, Lariviere Vincent, Gingras Yves, Benchmarking scientific output in the social sciences and humanities: The limits of existing databases, Scientometrics, 2006, 68, 3, 329, 342</bibtext> </blist> <blist> <bibl id="bib5" idref="ref5" type="bt">5</bibl> <bibtext> Beigel Fernanda, Publishing from the periphery: Structural heterogeneity and segmented circuits. The evaluation of scientific publications for tenure in Argentina's CONICET, Current Sociology, 2014, 62, 5, 743, 765</bibtext> </blist> <blist> <bibl id="bib6" idref="ref10" type="bt">6</bibl> <bibtext> Beigel Fernanda, Salatino Maximiliano, Circuitos segmentados de consagración académica: las revistas de Ciencias Sociales y Humanas en la Argentina, Información, cultura y sociedad, 2015, 32, 11, 36</bibtext> </blist> <blist> <bibl id="bib7" idref="ref11" type="bt">7</bibl> <bibtext> Bianco Mariela, Gras Natalia, Sutz Judith, Academic Evaluation: Universal Instrument? Tool for Development?, Minerva, 2016, 54, 4, 399, 421</bibtext> </blist> <blist> <bibl id="bib8" type="bt">8</bibl> <bibtext> Blau Judith, Scientific Recognition: Academic Context and Professional Role, Social Studies of Science, 1976, 6, 3/4, 533, 545</bibtext> </blist> <blist> <bibl id="bib9" type="bt">9</bibl> <bibtext> Bornmann Lutz, Thor Andreas, Marx Werner, Schier Hermann, The application of bibliometrics to research evaluation in the humanities and social sciences: An exploratory study using normalized Google Scholar data for the publications of a research institute, Journal of the Association for Information Science and Technology, 2016, 67, 11, 2778, 2789</bibtext> </blist> <blist> <bibtext> Clemens Elisabeth S, Powell Walter W, McIlwaine Kris, Okamoto Dina, Careers in print: Books, journals, and scholarly reputations, American Journal of Sociology, 1995, 101, 2, 433, 494</bibtext> </blist> <blist> <bibtext> Cole Jonathan, Cole Stephen, Measuring the quality of sociological research: Problems in the use of the Science Citation Index, The American Sociologist, 1971, 6, 1, 23, 29</bibtext> </blist> <blist> <bibtext> Cole Jonathan, Cole Stephen, Social Stratification In Science, 1973, Chicago, University of Chicago Press</bibtext> </blist> <blist> <bibtext> Courtault Jean-Michel, Hayek Naila, Rimbaux Eric, Zhu Tong, Research in economics and management in France: A bibliometric study using the h-index, Journal of Socio-Economics, 2010, 39, 2, 329, 337</bibtext> </blist> <blist> <bibtext> Costas Rodrigo, Bordons María, The h-index: Advantages, limitations and its relation with other bibliometric indicators at the micro level, Journal of Informetrics, 2007, 1, 3, 193, 203</bibtext> </blist> <blist> <bibtext> Delgado-López-Cózar Emilio, Orduña-Malea Enrique, Jiménez-Contreras Evaristo, Ruiz-Pérez Rafael, H Index Scholar: el índice h de los profesores de las universidades públicas españolas en humanidades y ciencias sociales, El profesional de la información, 2014, 23, 1, 87, 94</bibtext> </blist> <blist> <bibtext> Doğan G Güleda, Şencan İpec, Tonta Yaşar, Does dirty data affect Google Scholar citations?, Proceedings of the Association for Information Science and Technology, 2016, 53, 1, 4</bibtext> </blist> <blist> <bibtext> Etxebarria Goio, Gomez-Uranga Mikel, Use of Scopus and Google Scholar to measure social sciences production in four major Spanish universities, Scientometrics, 2010, 82, 2, 333, 349</bibtext> </blist> <blist> <bibtext> Fernández Esquinas Manuel, Las revistas de ciencias sociales en los sistemas de I+D, Revista Española de Sociología, 2016, 25, 3, 427, 442</bibtext> </blist> <blist> <bibtext> Financial Times. 2015. European Business School Rankings 2015., <ulink href="http://rankings.ft.com/pdf/european-business-school-rankings-2015.pdf">http://rankings.ft.com/pdf/european-business-school-rankings-2015.pdf</ulink>. Accessed 5 Apr 2016.</bibtext> </blist> <blist> <bibtext> Frandsen Tobbe F, Nicolaisen Jeppe, Intradisciplinary differences in database coverage and the consequences for bibliometric research, Journal of the American Society for Information Science and Technology, 2008, 59, 10, 1570, 1581</bibtext> </blist> <blist> <bibtext> Gantman Ernesto R, Fernández Rodríguez. Carlos J, La profesión académica en Argentina y España y su productividad científica en ciencias sociales, Revista de Ciencias Sociales, 2013, 19, 3, 500, 510</bibtext> </blist> <blist> <bibtext> Gantman Ernesto R, Fernández Rodríguez Carlos J, Literature segmentation in management and organization studies: The case of Spanish-speaking countries (2000-10), Research Evaluation, 2016, 25, 4, 461, 471</bibtext> </blist> <blist> <bibtext> García Pérez Miguel A, Accuracy and completeness of publication and citation records in the Web of Science, PsycINFO, and Google Scholar: A case study for the computation of h indices in Psychology, Journal of the American Society for Information Science and Technology, 2010, 61, 10, 2070, 2085</bibtext> </blist> <blist> <bibtext> Halevi Gali, Moed Henk, Bar-Ilan Judit, Suitability of Google Scholar as a source of scientific information and as a source of data for scientific evaluation—Review of the literature, Journal of Informetrics, 2017, 11, 3, 823, 834</bibtext> </blist> <blist> <bibtext> Hanafi Sari, University systems in the Arab East: Publish globally and perish locally vs. publish locally and perish globally, Current Sociology, 2011, 59, 3, 291, 309</bibtext> </blist> <blist> <bibtext> Harzing Anne-Wil, The Publish or Perish Book: Your guide to effective and responsible citation analysis, 2010, Melbourne, Tarma Software Research</bibtext> </blist> <blist> <bibtext> Harzing Anne-Wil, van der Wal R, Google Scholar as a new source for citation analysis, Ethics in Science and Environmental Politics, 2008, 8, 1, 61, 73</bibtext> </blist> <blist> <bibtext> Hicks Diana, The difficulty of achieving full coverage of international social science literature and the bibliometric consequences, Scientometrics, 1999, 44, 2, 193, 215</bibtext> </blist> <blist> <bibtext> Hicks Diana, Moed Henk F, Glänzel Wolfgang, Schmoch Ulrich, The four literatures of social science, Handbook of quantitative science and technology research, 2005, Dordrecht, Springer, 473, 493</bibtext> </blist> <blist> <bibtext> Hirsch Jorge, An index to quantify an individual's scientific research output, Proceedings of the National Academy of Sciences of the United States of America, 2005, 102, 46, 16569, 16572</bibtext> </blist> <blist> <bibtext> Hirsch Jorge, Buela-Casal Gualberto, The meaning of the h-index, International Journal of Clinical and Health Psychology, 2014, 14, 2, 161, 164</bibtext> </blist> <blist> <bibtext> I-UGR. 2014. Ranking of Spanish Universities (5th. edition), <ulink href="http://www.rankinguniversidades.es">http://www.rankinguniversidades.es</ulink>. Accessed 5 Apr 2016.</bibtext> </blist> <blist> <bibtext> Jacsó Peter, Google Scholar: The pros and the cons, Online Information Review, 2005, 29, 2, 208, 214</bibtext> </blist> <blist> <bibtext> Jedlicki, Fanny, and Romain Pudal. 2018. Introduction. Droit d'entrée dans la carrière académique: à quel prix?, Socio-logos 13., https://journals.openedition.org/socio-logos/3194. Accessed 20 May 2018.</bibtext> </blist> <blist> <bibtext> Jiménez-Contreras E, Ruiz-Pérez Rafael, López-Cózar Emilio Delgado, Spanish personal name variations in national and international biomedical databases: Implications for information retrieval and bibliometric studies, Journal of the Medical Library Association, 2002, 90, 4, 411, 430</bibtext> </blist> <blist> <bibtext> Karpik Lucien, "Performance", "excellence" et création scientifique, Revue Française de Socio-Économie, 2012, 10, 113, 135</bibtext> </blist> <blist> <bibtext> Lange Lydia, Frensch PA, Gaining scientific recognition by position: Does editorship increase citation rates?, Scientometrics, 2006, 44, 3, 459, 486</bibtext> </blist> <blist> <bibtext> Masip Pere, Efecto Aneca: producción española en comunicaciones en el Social Science Citation Index, Anuario ThinkEPI, 2011, 5, 206, 210</bibtext> </blist> <blist> <bibtext> Orduña-Malea Enrique, Martín-Martín Alberto, Ayllón Juan M, López-Cózar Emilio Delgado, La revolución Google Scholar, 2016, Destapando la caja de Pandora académica. Granada, Editorial Universidad de Granada</bibtext> </blist> <blist> <bibtext> Pontille, David, and Didier Torny. 2010. Revues qui comptent, revues qu'on compte: produire des classements en économie et gestion., Revue de la régulation (8)., https://journals.openedition.org/regulation/8881. Accessed 22 May 2018.</bibtext> </blist> <blist> <bibtext> Scimago. 2017. Scimago Journal and Country Rank., <ulink href="http://www.scimagojr.com/countryrank.php?area=3300">http://www.scimagojr.com/countryrank.php?area=3300</ulink>. Accessed 2 June 2017.</bibtext> </blist> <blist> <bibtext> Suárez-Ortega Magdalena, García-Mingo Elisa, Román José Ruiz San, When Español is not enough: research, write, translate and publish or ... perish, International Journal of Leadership in Education, 2012, 15, 4, 463, 482</bibtext> </blist> <blist> <bibtext> Sztompka Piotr, Patel Sujata, One Sociology or Many?, The ISA handbook of diverse sociological traditions, 2009, Los Angeles, CA, Sage, 21, 39</bibtext> </blist> <blist> <bibtext> Waltman Ludo, van Eck Nees J, The inconsistency of the h-index, Journal of the Association for Information Science and Technology, 2012, 63, 2, 406, 415</bibtext> </blist> <blist> <bibtext> Youtie Jan, Rogers Juan, Heinze Thomas, Shapirad Philip, Tange Li, Career-based influences on scientific recognition in the United States and Europe: Longitudinal evidence from curriculum vitae data, Research Policy, 2013, 42, 8, 1341, 1355</bibtext> </blist> </ref> <ref id="AN0134622793-10"> <title> Footnotes </title> <blist> <bibtext> While GS data for individual scientists may not be as comprehensive as the list of publications in a CV, in most cases the coverage is almost complete as shown by García Pérez (2010) for the case of four Spanish psychologists.</bibtext> </blist> <blist> <bibtext> The AERES has been replaced by the HCERES (Haut Conseil de l'évaluation de la recherche et de l'enseignement supérieur) from whose website (<ulink href="http://www.hceres.fr">http://www.hceres.fr</ulink>) we obtained the reports for classifying the French doctoral schools.</bibtext> </blist> <blist> <bibtext> For instance, in some cases, the same researcher publishes with a double surname in Spanish journals and with a single surname in foreign journals.</bibtext> </blist> <blist> <bibtext> Besides, in some local journals, the titles of published articles appear both in English and in the vernacular language; and both could be retrieved by GS as individual entries. In these cases, we consolidated the entries, assigning the language of full-text publication to the article. In contrast, when complete versions of the same article were published in English and in the vernacular language, we took them as separate entries.</bibtext> </blist> <blist> <bibtext> The number of observations is slightly less than 250 in some disciplines because a few social scientists do not have publications (articles, books or chapters) in their GS research output.</bibtext> </blist> <blist> <bibtext> The increasing importance of the "publish or perish" imperative, particularly in journals published in English, is considered by some authors as leading to changes in research agendas, especially against those of a more contextualized (local or regional) nature that are deemed relevant for scientists in non-Anglophone and developing countries (Bianco et al. 2016).</bibtext> </blist> <blist> <bibtext> It could also be contended that more senior researchers have made many publications not detected by GS. Nevertheless, books and chapter with even minimum impact may indeed be captured by GS. If a book or chapter is cited in the full text of a GS-indexed document, then GS retrieves it as a "citation", and the document thus appears as an entry in the author's research output generated by the Publish or Perish software.</bibtext> </blist> </ref> <aug> <p>By Marcelo P. Dabós; Ernesto R. Gantman and Carlos J. Fernández Rodríguez</p> </aug>
Header DbId: eric
DbLabel: ERIC
An: EJ1205281
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: The Prestige of Social Scientists in Spain and France: An Examination of Their h-Index Values Using Scopus and Google Scholar
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Dabós%2C+Marcelo+P%2E%22">Dabós, Marcelo P.</searchLink><br /><searchLink fieldCode="AR" term="%22Gantman%2C+Ernesto+R%2E%22">Gantman, Ernesto R.</searchLink><br /><searchLink fieldCode="AR" term="%22Fernández+Rodríguez%2C+Carlos+J%2E%22">Fernández Rodríguez, Carlos J.</searchLink> (ORCID <externalLink term="http://orcid.org/0000-0002-2959-8195">0000-0002-2959-8195</externalLink>)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Minerva%3A+A+Review+of+Science%2C+Learning+and+Policy%22"><i>Minerva: A Review of Science, Learning and Policy</i></searchLink>. Mar 2019 57(1):47-66.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Springer. Available from: Springer Nature. 233 Spring Street, New York, NY 10013. Tel: 800-777-4643; Tel: 212-460-1500; Fax: 212-348-4505; e-mail: customerservice@springernature.com; Web site: https://link.springer.com/
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 20
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2019
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Foreign+Countries%22">Foreign Countries</searchLink><br /><searchLink fieldCode="DE" term="%22Reputation%22">Reputation</searchLink><br /><searchLink fieldCode="DE" term="%22Social+Science+Research%22">Social Science Research</searchLink><br /><searchLink fieldCode="DE" term="%22Economics%22">Economics</searchLink><br /><searchLink fieldCode="DE" term="%22Sociology%22">Sociology</searchLink><br /><searchLink fieldCode="DE" term="%22Administration%22">Administration</searchLink><br /><searchLink fieldCode="DE" term="%22Citations+%28References%29%22">Citations (References)</searchLink><br /><searchLink fieldCode="DE" term="%22Publications%22">Publications</searchLink><br /><searchLink fieldCode="DE" term="%22English%22">English</searchLink><br /><searchLink fieldCode="DE" term="%22Researchers%22">Researchers</searchLink>
– Name: Subject
  Label: Geographic Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Spain%22">Spain</searchLink><br /><searchLink fieldCode="DE" term="%22France%22">France</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1007/s11024-018-9358-0
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0026-4695
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: We analyze the prestige of 1,500 scholars in economics, sociology, and management who have Spanish and French institutional affiliations operationalized by their h-index in Scopus and Google Scholar. We use a negative binomial count model to examine how some individual factors affect the h-index from both databases. The results show a non-monotonic relationship between the researchers' career length and their h-index. There is a positive and statistically significant relationship between total research output and the h-index. The share of publications in English over total publications has also a positive and statistically significant effect on the h-index, except in a single case, while the share of publications in other foreign languages does not have such effect. Finally, we found that the effects of the number of citations received by documents in English (international impact) and by those in the vernacular language (local or regional impact) on the h-index vary according to the database, the country, and the discipline in question.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2019
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1205281
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1205281
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s11024-018-9358-0
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 20
        StartPage: 47
    Subjects:
      – SubjectFull: Foreign Countries
        Type: general
      – SubjectFull: Reputation
        Type: general
      – SubjectFull: Social Science Research
        Type: general
      – SubjectFull: Economics
        Type: general
      – SubjectFull: Sociology
        Type: general
      – SubjectFull: Administration
        Type: general
      – SubjectFull: Citations (References)
        Type: general
      – SubjectFull: Publications
        Type: general
      – SubjectFull: English
        Type: general
      – SubjectFull: Researchers
        Type: general
      – SubjectFull: Spain
        Type: general
      – SubjectFull: France
        Type: general
    Titles:
      – TitleFull: The Prestige of Social Scientists in Spain and France: An Examination of Their h-Index Values Using Scopus and Google Scholar
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Dabós, Marcelo P.
      – PersonEntity:
          Name:
            NameFull: Gantman, Ernesto R.
      – PersonEntity:
          Name:
            NameFull: Fernández Rodríguez, Carlos J.
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 03
              Type: published
              Y: 2019
          Identifiers:
            – Type: issn-print
              Value: 0026-4695
          Numbering:
            – Type: volume
              Value: 57
            – Type: issue
              Value: 1
          Titles:
            – TitleFull: Minerva: A Review of Science, Learning and Policy
              Type: main
ResultId 1