Insight into what they cite: a citation analysis of publications at the School of Business, Economics and Law at the University of Gothenburg

Abstract

The purpose of this citation analysis is to understand what kind of publications are cited by scholars at the School of Business, Economics and Law at the University of Gothenburg. The study uses publication data from 2015 to 2018, and analyses the school’s four departments and their research sections. Four questions are examined: type of documents cited, most-cited journals, age of cited materials, and to what extent the library provides access to these materials. The analysis seeks to provide a better understanding of the similarities and differences across citation cultures, to respond to faculty research needs and to facilitate collection management decisions.

Keywords:

• Citation analysis
• business schools
• collection management
• collection development
• law schools
• citation studies

Introduction

Academic libraries have many methods at their disposal for evaluation of their collections. Analyzing usage data and researchers’ publications are common examples which are now well-integrated aspects of library work, with influence on acquisitions. Another way to evaluate a library collection is to conduct citation analyses. A citation analysis aims to study what is being cited in various publications. Libraries have used citation analysis in different ways. They commonly try to assess to what extent a given library collection supports scholars’ need for literature in relation to research.

The authors of this article, librarians of the media team at the Economics Library at the University of Gothenburg,¹ Sweden, conducted a citation analysis in 2019. The citation analysis used publication data from the School of Business, Economics and Law, which is one of eight faculties at the University of Gothenburg. The faculty consists of four departments: Economy & Society, Law, Business Administration, and Economics. Each department is structured according to its particular research interests. The sources used were published between 2015 and 2018.

The aim of this study has been to understand the faculty’s research needs and to find what kinds of publications are being cited by the faculty’s scholars. This type of analysis would highlight patterns for how the departments differ in terms of what kinds of materials they need and whether the library provides access to them. Awareness of these aspects makes it possible to better develop the library’s acquisition practices, and to make informed decisions for the purpose of collection management.

The authors will answer the following research questions:

1. Document type analysis. What kinds of materials are used by the scholars at the School of Business, Economics and Law at the University of Gothenburg, and how do citation patterns differ throughout the departments and the research sections?

2. Journal frequency analysis. Which are the most-cited journals? Is there a particular concentration of specific journals in the respective departments, and does the 80/20 rule, aka the Pareto principle,² apply?

3. Age analysis. What is the median and average age of cited materials?

4. Holdings analysis. To what extent does the Gothenburg University Library provide access to the publications cited?

Literature review

The literature consulted for this study can be divided into three main areas: a first one dealing with citation analysis as a research method; a second, consisting of citation analysis studies; and a third selection of sources that focus on the analysis of journals’ impact and relevance within a given field.

The first selection includes Hoffmann and Doucette (2012) which have served as a point of reference since their review of thirty-four citation analysis studies, and provided many useful recommendations and guidelines. Similarly, Nisonger provided a theoretical framework for understanding the role of core lists within library work (2007) as well as how the 20/80 rule can be understood—and criticized—in relation to core journals, citation data, and more (2008).

For the second group of sources, a diverse selection of citation analysis studies was used. In Vaaler (2018), the local Business school and its five departments were studied by focusing on common research questions of citation analysis, such as document type, age, subscription and access, as well as journals’ concentration and the 20/80 rule. The work of Dewland (2011) and Dees (2016) further dealt with the same Business school, and similar to Vaaler, sought to better understand citation culture and the faculty’s needs, in order to improve library support. Here, journals were at the core of the analysis, with all other document types grouped as non-journal sources. Librarians Currie and Monroe-Gulick (2013) conducted an interdisciplinary analysis that focused on three broad disciplines: humanities, social sciences, and science. The authors’ main goal was to assess library adequacy in providing support for scholars. Another multidisciplinary study was that of Tucker (2013), who assessed nine different colleges within a university, sampled a large number of citations data, performed an in-depth investigation of collections ownership, and provided a detailed account of the top-ten cited journals for each college. Gao’s (2015) citation analysis concerned the field of communication. Even though the author highlighted seven different document types, the main focus here was on Articles and Books, their age distribution, and their local availability. Ke and Bronicki (2015), likewise, focused their efforts on a single discipline: psychology. In their study, the data covered a sampling interval of ten years of citations, making it one of the largest among our cited sources. Interestingly, Ke and Bronicki (2015) compared citation data with usage data over the course of one year, but their result showd no correlation between the two. A similar conclusion was drawn by Eva (2016), who sought to study the relationship between usage data and citation data, and concluded that the two sets of statistics should be seen not as interchangeable, but as complementary. Tavernaro and Salisbury (2015) sampled publication and citation data for the faculty of marketing from a thirteen-year time span. The authors provided a long list of most-cited journals, which they enriched by using WorldCat subject headings, in order to better understand topics cited by the faculty. For the field of Law, the work of Cln and Dike (2016) showed how undergraduate students cited sources in their research projects. It also highlighted the value of the document types Law Reports and Statutes for this user group. Burman (2011) focused instead on PhD students’ dissertations, whose citation patterns made plain the importance of Reports, Statutes, and other Governmental Documents (also called Gov. Docs). At the same time, the study emphasized both journal Articles and Books as major resources.

Also part of the second group were studies that concerned themselves with student theses (both graduate and undergraduate) and doctoral dissertations, Becker and Chiware (2015) and Datig (2016) served as a methodological point of reference. Datig’s study (2016) distinguished itself from others by using focus groups to better investigate the behavior of students. With Widmark (2014) as the only exception, no other citation analysis conducted in Swedish institutions was found.

The third group of selected literature concerned the analysis of journal ranking and citation patterns. These studies sought either to establish which journals were the most impactful within a given discipline, or to study a certain topic by looking at a group of relevant journals within the field. Swanson, Wolfe, and Zardkoohi (2007) looked into article concentration of highly ranked academic journals within the fields of Accounting, Finance, Management and Marketing. Wei and Zhang (2020) provided an in-depth analysis of the evolution of intellectual structure within the discipline of Business, by looking at a list of 24 internationally recognized Business Journals compiled by the University of Texas (UTD24). Lim et al. (2009) observed citation and impact for the field of Management and Information Systems (MIS), among others. Finally, Summers and Wood (2017), and Danielson and Heck (2014) evaluated, respectively, in which journals Finance researchers published themselves, as well as the relationship between general and specialist journals within Accounting. For the field of Economics, fewer studies that dealt with journal ranking and analysis were found, but the work of García-Romero et al. (2016) proved to be an important source for assessing a core journals list for the field.

Literature concerning bibliometric and journal analysis within the fields of Economic History and Human geography was limited. The study by Fernández-de-Pinedo and Muñoz (2019) provided, however, a solid base of relevant journals with which to compare the results of the citation analysis within Economic History.

Finally, Eisenberg and Wells (2014) crucially pointed out how the field of Law was poorly represented in the journal ranking published by Journal Citations Reports (JCR), a resource widely used to assess research quality. Awareness of this underrepresentation was an important aspect that allowed the authors to interpret the results of their department more correctly.³

Method

The first step in this study was to create a Microsoft Access database of all the scholars at the Faculty of Business, Economics and Law. The database was organized so that each scholar was assigned an affiliation on both department and research section level. This information, and the structure of it, were developed according to data from the university’s webpage, where the research activity of each faculty and department is thoroughly described. Four years of publications, and their respective citations from 2015 to 2018, were collected from Scopus for all scholars whose work could be retrieved. Those whose publications and corresponding citations were absent from Scopus (52% of all scholars) were not included in the study. As for most other studies like Gao (2015), Vaaler (2018), and Dees (2016), Excel was the main tool for analysis. Additionally, the spreadsheets containing publications and citations were later imported into the Access database, following the example of Dewland (2011), and linking each scholar with their publications and citations. This made it possible to create lists of citations belonging to each research section, according to the correct scholar affiliation taken from the university’s home page; see Table 1. Once the data was collected, the first analysis was conducted in order to assess and answer the research question concerning the format of materials.

Table 1. Research structure at the Faculty of Business, Economics and Law.

Dept. of Business Administration Accounting; Marketing; Industrial & Financial Management & Logistics (IFEL); Management & Organizations	Dept. of Law: Education-Driven Research; Societal Challenge-Driven Research
Dept. of Economy & Society Economic History; Human Geography; Innovation & Entrepreneurship	Dept. of Economics Behavioral & Experimental Economics; Development Economics; Environmental Economics; Financial Economics; Health Economics; Labor Economics; Public Economics

The Scopus database proved a problem, since 43% of all citations were poorly described and lacked the Document Type field. It is interesting to observe how the citation analyses, that depended on Scopus for the mining of citations (Gao, 2015; Tucker, 2013; Ke & Bronicki, 2015; Vaaler, 2018), do not mention (and do not seem to have encountered) this serious deficiency. To solve the issue, the authors manually coded each citation lacking this important field, meanwhile also correcting minor errors, such as lack of publication year, incomplete titles and so on. This turned out to be a crucial effort, since a high number of older monographs and non-article publications were found among the references with insufficient metadata. Ignoring these would certainly have yielded misleading results. The results from Dewland (2011) concerning non-journal citations within Business schools show how these sometimes constitute a considerable share of the total number of references. This, again, shows the necessity of manual codification to provide sufficient metadata.

To manually code citations in order to include more diverse formats is a practice recurring in many other studies (Becker & Chiware, 2015; Datig, 2016; Tucker, 2013; Vaaler, 2018). It allows for a broader understanding of citation cultures in the departments, and in the case of this study, across the research sections. Every citation received a specific document type categorization, from Books to Company Info or Datasets, yielding more than a hundred formats. To make the data more readable, both Scopus types and manually coded format types were grouped into eight general categories, plus one for Unknown sources (meaning the irretrievable ones, or those with very poor metadata). This meant that a general category such as Papers included both Conference Papers from Scopus as well as Working Papers and Discussion Papers, which were manually coded for the aim of this study.

The general categories are the following: Articles, Books, Papers, Reports, Other, Gov. Docs, Online Resources, and Jurisprudence.

Through the linking of citations to their native publication in the Access database, the citation age, “the difference between the date of citation and the date of the publication in which it was cited” (Hoffmann & Doucette, 2012, p. 325), could also be determined. Finally, the spreadsheet documents were used to review holdings in the library catalogue (Koha), and in the OpenUrl link server (SFX).

In presenting its results, this study has a focus on the faculty and department levels, as the vast amount of research sections would make the presentation of data much too bulky. However, all noteworthy results from the research sections will be highlighted in turn, and compared with their respective department, as well as with other studies.

Limitations

Like any study involving the analysis of large amounts of data, and so many options for narrowing the research scope, this study has had to limit itself and prioritize some questions before others. As mentioned above, one limitation that has been taken into account is that not all publications presented in the GUP, the publication repository of the University of Gothenburg, can be retrieved through Scopus and/or through other citation databases. This is both due to resource format and language. Monographs and chapters, as well as minor papers and reports published in Swedish are not retrievable through Scopus. Obviously, human error must also be factored in, and some publications may have been overlooked during mining. The absence of such a large amount of Swedish publications from the data, even though it does not affect the vast majority of peer reviewed sources used, certainly has an impact on the understanding of citation cultures. During publication mining, the authors observed how much more common it is for the department of Law to publish in Swedish monographs and journals absent in Scopus. The poor metadata quality of Scopus proved a considerable obstacle, since it penalizes monographs and non-journal publications, whose lower bibliographic quality limits the possibilities for analysis.

Document type analysis

The total amount of publications and citations mined from Scopus were 652 and 38,451 respectively (Table 2), and as mentioned earlier, the publications themselves dated from 2015 to 2018.

Table 2. Number of publications, scholars and citations.

Departments	Publications	Scholars	Total citations
Business administration	277	59	18627
Economics	186	41	8347
Economy & Society	141	44	9072
Law	48	22	2405
Total	652	166	38,451

The general results of the analysis across the departments and at faculty level (Tables 3–5) reveal, surprisingly, the high incidence of book citations. With a total of 24.24%, a peak of 40.62% for Law, and 34.53% for the Economy & Society departments, monographs are commonly used. Also noteworthy is the fact that the original data from Scopus amounted to a much smaller share of Books used for citations, resulting exclusively in single-digit percentages.

Table 3. Document type per department and faculty (by percentage).

Document type	Business Administration	Economics	Economy & Society	Law	Faculty
Articles	65.77%	71.15%	49.45%	31.85%	60.96%
Books	23.47%	10.05%	34.53%	40.62%	24.24%
Books/dissertations	0.75%	0.24%	1.68%	2.04%	0.94%
Papers	3.40%	9.65%	3.47%	3.28%	4.76%
Papers/conference	2.01%	1.53%	1.63%	1.08%	1.76%
Papers/working papers	0.56%	5.16%	1.09%	0.67%	1.69%
Reports	2.50%	3.93%	3.00%	6.03%	3.15%
Other	2.28%	2.95%	4.64%	3.99%	3.09%
Newspapers/magazines	0.74%	0.34%	0.87%	1.33%	0.72%
Case study/study/survey	0.13%	0.62%	0.24%	0.08%	0.26%
Archive materials/books	0.01%	0.00%	0.67%	0.12%	0.17%
Conference/meeting/consortium/seminar/workshop	0.06%	0.08%	0.17%	0.75%	0.14%
Unknown	1.33%	1.07%	2.55%	8.15%	1.98%
Gov. docs	0.53%	0.78%	1.08%	3.41%	0.89%
Online resources	0.72%	0.37%	1.21%	1.58%	0.81%
Jurisprudence	0.01%	0.06%	0.07%	1.08%	0.10%

Table 4. Number of total citations.

Document type: general categories	Citations
Articles	23,441
Books	9320
Papers	1832
Reports	1211
Other	1187
Unknown	763
Gov. Docs	344
Online Resources	313
Jurisprudence	39

Table 5. Document type, focus on Articles and Books (by percentage).

These results differ significantly from those of Dewland (2011), who found that the sum of all non-journal citations⁴ amounted to 22% (ibid., p. 149), while in the case of Business Administration at the University of Gothenburg, all non-journal and Books citations account for 39.9% and 23.47%, respectively, of the department’s citations. In similar fashion, Vaaler (2018) reports a smaller number of books at the Business school, with a total of 11.78%, but a higher amount at the Management department. Tucker’s results (2013) further indicate how the Gothenburg department of Business Administration differs from the Business college of the University of Nevada, with a 10% higher incidence of books citations. Comparatively, the results of the Education and Liberal Arts colleges in Tucker’s study are closer to those of the Gothenburg Business department. Currie and Monroe-Gulick’s (2013) results, too, are interesting in this regard, since they show that 26% of citations at the Social Sciences faculty at the University of Kansas were Books, and 68% were Articles. As shown in Table 3, this result is similar to that of the Gothenburg department of Business Administration. In Gao’s study (2015), communication research scholars cited 59% Articles and 29% Books. This affinity with the citation culture of the field of social sciences undoubtedly needs more in-depth analysis, but nonetheless it is an interesting result, which expands the understanding of the department of Business. Monographs play thus an important role in the research practices of the department’s scholars. Besides Books and Articles, other important documents are Reports and Papers, whose relevance is stressed by Vaaler (2018), as can indeed be observed at the Business Administration in Gothenburg as well, where Reports and Papers amount to 2.50% and 3.40% of all citations respectively.

The Economics department of the Gothenburg University has the highest number of Articles and the least number of Books, which constitute 71.15% and 10.05% of all citations respectively. Meanwhile, Economics is the department with the highest amount of Papers cited (9.65%), including Working Papers and Conference Papers. This suggests that Books in general seem to be used less at this department, and it could be advisable to be more conservative in the purchase of monographic resources for the Economics department. With these results, the department is more similar in citation trends to the Business departments examined by Vaaler (2018), Dewland (2011), and Tucker (2013).

The Economy & Society department has much in common with the international humanities departments and faculties in terms of number of Articles (49.45%), and Books (34.53%) citations. This can be observed, for example, in both Tucker (2013) and Currie and Monroe-Gulick (2013). In the latter study, the proportions between Articles and Books for the humanities are reversed compared to the social sciences (2013, p. 476), with 68% Books. Tucker’s results for the college of Education and Liberal Arts show instead 51% and 53% Articles, and 40% and 38% Books respectively (2013, pp. 23–24). This clearly puts the departments of Economy & Society closer to the fields of Social Sciences and Humanities, in terms of what kinds of material the scholars within the discipline are using in their publications.

Interestingly, the Economy & Society and Law departments display a higher interest in Dissertations (1.68% and 2.04%); a less common format at the Economics department, but a slightly more common one for Business Administration (see Table 3).

The department of Law displays the highest incidence of resources almost completely lacking metadata, or being irretrievable; what the authors call Unknown, as well as the format type Other, which includes a vast variety of resources featuring in low numbers. With only 31.85% of Articles, Law appears highly interested in other types of sources to support its publications; Gov. Docs at 3.41%, Reports at 6.03% and Other at 3.99%, thus making it the department with the highest amount of sources that are neither Articles nor Books: 27.53%. The results show similarity to other citation analysis studies concerned with Law departments. Burman (2011) observed that Constitution/Statutes and Acts amounted to 16% of all citations, while Reports amounted to 9%. The former document type by Burman can be compared to this study’s Gov. Docs, thus confirming that both the Gov. Docs and Reports categories are indeed relevant to law scholars. This is also confirmed by Cln and Dike (2016), whose analysis of undergraduate law projects stressed not only the importance of Books, but also highlighted the role that Law Reports, Statutes, Acts, Conventions and other unique resources uphold among citations. The specificity and uniqueness of Law departments’ citation interests call for a more dedicated analysis, where increased variety in the classification and identification of document types would provide a better understanding of citation culture. For the library, these results imply that even more open purchase policies should be taken into consideration, regarding, for instance, databases containing other types of documents, especially for the benefit of the Law department.

Overall, the citation patterns of the department of Business Administration coincide most with those of the overall faculty, but it is important to remember that the very high numbers of publications and citations sampled for this department alone have a clear impact on the overall document type results of the whole faculty. However, when looking at all the departments individually, they display more idiosyncratic qualities compared to the faculty as a whole. For instance, the results clearly show how Economy & Society and Law bear affinities with other disciplines, such as Humanities, Education, and Liberal Arts.

The Business Administration department proved to be fairly unique compared to similar colleges and schools, since it showed citations patterns bringing it closer to the field of Social Science. Finally, the Economics department, too, has shown its uniqueness, while at the same time being consistent with how the field’s citations patterns are described in the literature.

Noteworthy cases from the research sections

Given the large number of research sections within each department, the authors have decided to concentrate on some noteworthy examples within these, which delineate differences in citation praxis.

Business Administration

Within the Business Administration department, Accounting and IFEL are the two sections that differentiate themselves the most from the department as a whole. Comparatively, both sections are closer to the results of the department of Economics than to the department of Business, and they both account for the least number of Books with 11.97% and 11.33% respectively. Meanwhile, the sections of Management & Organization Research and Marketing account for the highest occurrence of Books citations: 29.45% and 25.45% respectively. The IFEL section has further elements that make it more similar to Economics than Management, since it relies more on Papers (7.39%), Reports (5.22%), Articles (64.10%) and less on Books (11.33%). Interestingly, Newspapers/Magazines (2.37%) is also a well-consulted source type among scholars from this section. These findings are similar to those of Vaaler (2018) and Dewland (2011), who could ascertain how their Accounting and Finance departments relied less on books. Similarly, Vaaler (2018) points out that the Management department at the Texas University Business Faculty had the highest number of cited Books; this is also true for the Gothenburg Management and Marketing sections within the Business Administration department, even if the numbers of non-journal citations at the University of Gothenburg are higher than those of Vaaler (2018), Dewland (2011), and Tavernaro and Salisbury (2015).

Overall, it can be observed that the Management and Marketing research sections have citation patterns that reflect the department, while Accounting and IFEL divert slightly.

Economics

The research sections within the department of Economics show a consistent internal citation culture, since the amount of citations per document types is similar across the seven fields. The Public Economics section is the only one with a higher incidence of Books citations (15.46%), compared to the average of 9.14%. Papers⁵ and Reports are well-consulted across the board, with a peak of 12.85% and 6.41% at the Development Economics section. Interestingly, the Environmental Economics section shows the highest amount of Gov. Docs sources cited (1.15%) within the department.

Law

The Law department is divided into two research sections: Education-driven Research and Societal Challenge-driven Research. Given the low amounts of publications downloaded from Scopus (48), it would have been too dispersive to divide the research sections into any further subcategories. The most noteworthy difference between the two sections lies in the incidence of other formats: Reports amount to almost 15% for the Societal section, and 7% for the Education section. Gov. Docs are also more frequently used by the former (9.12% and 4.37% respectively). It might be inferred that the study of themes such as Governance and Sustainability, which are common topics for the Societal section, needs these kinds of sources. Reports and documents from large institutions and governments could be considered major sources of information for assessing policies and jurisdiction. This, however, needs further inquiry, since both Cln and Dike (2016) and Burman (2011) point out the relevance of these document types exactly: from Law Reports to foreign Statutes and Acts. The education-driven section has a high incidence of Books: almost 40%. It can be argued that this result is coherent with the topics of Law that distinguish the section—such as Public Law, Criminal Law etc—and the fact that the focus lies on education-driven research. Volumes on Swedish and international jurisdiction are certainly at the core of this section’s research. This is a central argument of Cln and Dike (2016); namely that “the study and practice of the legal profession is book intensive” (ibid. p. 478). As this research section is focused on education, it comes as no surprise that Books play such a vital role. The low amount of publications retrieved, and the high numbers of Unknown sources are clearly an issue. Despite this, the results are still significant in their uniqueness compared to those of the other departments, and can be used by the library to better acknowledge departmental needs, and as an incentive to conduct further and more detailed analysis of the citations culture within Law studies.

Economy & Society

Alongside Law, Economy & Society is another department with a high incidence of Books citations. The research section of Economic History with 42.20% Books citations, is the section with the highest amount of use of this document type. Both Human Geography and Innovation & Entrepreneurship instead maintain a proportion between Articles and Books that is closer to that of the Faculty as a whole. The Economic History section distinguishes itself further because of the high number of Book Dissertations occurrences within Books (2.66%), but also because the authors have found in this section a certain amount of sources classified as Archive Materials (1.94%): particularly older primary sources, which are commonly used by historians. Even if the number of Books does not bring the section to the same level as the humanities analyzed by Currie and Monroe-Gulick (2013), it leans in that direction, and demonstrates the diversity and interdisciplinary quality of the research at the Faculty of Business, Economics and Law.

Journal frequency analysis

At the faculty of Business, Economics and Law, 5009 unique journals were cited 23,440 times. In the context of this analysis, the Pareto Principle will be examined in order to determine whether a certain concentration of citations exists for a particular group of journals. Despite not being a strictly bibliometric tool (Nisonger, 2008), the Pareto Principle is often used to identify core journal lists, and to support acquisition and evaluation practices “when found in reasonable approximation” (ibid. p. 72).

Alongside the statistical results produced by the Pareto 20/80 ratio, this analysis also seeks to highlight different types of core lists which can be useful to libraries for collection management purposes. Being aware of all deficiencies and problems regarding the concept of “core” (Nisonger, 2007, 2008), the authors believe these results are nonetheless important in order to understand which journals are relevant to the faculty’s scholars. Only top-ten lists at the department level (with the exception of the Economy & Society research sections) will be included for the sake of presentation, comparison and analysis. However, in cases where the Pareto Principle is reasonably respected at about 20%, it is this 20% amount—always higher than ten—that will be considered core, and taken into account for collection management purposes. Further, these core lists focus on numbers of citations, and even though core lists can also be produced by looking at rankings, as Nisonger suggests, this is secondary for the goals of this study, which focuses instead on what has been used and cited by faculty scholars (Table 6).

Table 6. Top ten most-cited journals at the Business Administration department with SCImago ranking (SJR), Norwegian register (NSD), and Dewey classification (DDC).

Title	Citations	SJR	NSD	DDC
American Journal of Sociology	251	4.33	2	301
Accounting, Organizations and Society	227	2.04	2	657
Administrative Science Quarterly	225	13.52	2	351
Journal of Finance	222	17.13	2	332
Research Policy	216	3.41	2	658
Academy of Management Journal	211	10.76	2	658
Journal of Financial Economics	194	13.63	2
Organization Studies	193	2.36	2	300
American Sociological Review	191	5.86	2	301
Journal of Consumer Research	184	6.59	2	658.8

At the department of Business Administration, 403 journals (20%) contributed 77.43% (11,807) of the total number of citations (15,249). Thus, the department adhered almost completely to the Pareto pattern, and is consequently close to the results of Vaaler (2018) and Dewland (2011), whose Business school results display a good 20/80 ratio. One major difference from the two aforementioned studies is the very low concentration of citations for the top ten journals, which contribute only 13.86% of all citations. The fact that the Business Administration department and its research sections use large numbers of journals is an interesting result, and even if the 20/80 rule is reasonably respected, these numbers (403 titles at department level) do question the idea of “core,” since many journals are cited, but with a lower amount of citations per journal.

At the research sections level, the differences compared to the departments examined by Vaaler and Dewland are more striking. These results can be summarized by noting the following amounts, corresponding to 20% of top cited journals: Accounting 68.84%; Marketing 72.94%; Management 77.48%; IFEL 72.08%. Accounting provides the most noteworthy anomaly, since for both Vaaler and Dewland the number of journals necessary to cover 80% of citations is less than 20%. Almost all of Vaaler’s departments use a high concentration of citations, needing few journals to reach 80%. Three out of five—Accounting, Finance, and Management—needed 9%, 15% and 10% respectively to reach 80% of citations (Vaaler, 2018, pp. 7–8). As can be observed above, the results for the Swedish Business Administration department indicate the opposite: all sections are below 80%, with Management closest to that threshold.

The list of most-cited journals within Business Administration shows some unique results. A comparison with Wei and Zhang (2020) shows how only five titles out of ten on the most-cited journals list of the Gothenburg Business Administration department could be found through the UTD24. However, when looking at Gothenburg’s own top-24 journals list, and at each section’s own top lists, the UTD24 is well represented throughout the department in general.

Interestingly, the ranking of journals according to the SJR is also inverted, since the highest-ranked titles appear further down the list: in third, fourth, sixth and seventh place. However, all ten titles are among those ranked highest according to the well-accredited Norwegian Register for Scientific Journals Series and Publishers (NSD). The most interesting and unique result is the presence of the American Journal of Sociology as the most cited journal, the journal Research Policy in fifth place, and the American Sociological Review in ninth. This highlights the closeness of the Management research section (from where most citations of these journals are taken) to a multidisciplinary citation culture, and to the social sciences, as was hinted earlier during the analysis of document types. Despite being classified as Management (658), Research Policy distinguishes itself as a “multi-disciplinary journal devoted to analysing, understanding and effectively responding to the economic, policy, management, organizational, environmental and other challenges posed by innovation, technology, R&D and science” (Research Policy, n.d.).

By comparing the lists of most-cited journals at both department and research section levels to Wei and Zhang (2020), Swanson et al. (2007), Lim et al. (2009), Danielson and Heck (2014), and Summers and Wood (2017), it can be observed that many of the most important journals in the field of Business recur in the citation patterns of the Gothenburg Business Administration department. An interesting divergence is the lack of prominent journals within the field of Management Information Systems (MIS) and Computing & Operations Management. Especially MIS seems thus to be less prominent at the Gothenburg department. Overall, the department of Business Administration displays some unique characteristics compared to its international sister departments, such as a pronounced interest in social sciences and interdisciplinary topics, a lower concentration of journals, and dependency on a larger number of titles.

The most-cited 20% of journals (228) at the department of Economics accounted for 77.85% (4623) of all citations (5938), making the department adhere closest of all four to the 20/80 Pareto ratio. Compared to Business, Economics relies on fewer journals, and thus on a higher concentration of sources. 26% of all citations are from the ten most cited journals in Table 7. This indicates that the department is using a smaller group of relevant journals for their research. The most cited journal, the American Economic Review, has 376 citations: 125 more than the Business Administration department’s most cited journal, the American Journal of Sociology. This says a lot about the importance of journals for the department, since the total number of citations accounted for in the analysis is much lower than the amount sampled for Business.

Table 7. Top ten most cited journals at the Economics department with SCImago ranking (SJR), Norwegian register (NSD) and Dewey classification (DCC).

Title	Citations	SJR	NSD	DDC
American Economic Review	376	11.88	2	330.9
Quarterly Journal of Economics	178	30.49	2	330.1
Journal of Public Economics	173	3.69	2	330.01
Econometrica	157	17.63	2	330
Journal of Economic Behavior and Organization	156	1.71	2	330
Economic Journal	126	5.00	2	330
Journal of Political Economy	108	22.90	2	320
Experimental Economics	96	2.97	2	330.1
Journal of Environmental Economics and Management	90	2.19	2	363.7
World Development	84	2.25	2	942

On the research sections level, three fields of study, namely Behavioral, Development and Environmental Economics have the largest amount of citations, adhering fairly close to the 20/80 rule with 20% of journals providing 81%, 73%, and 79% of all citations respectively. Financial, Health, Labor, and Public Economics deviate more from the ratio, showing around 60% of citations corresponding to 20% of journals cited. This, however, has little effect on the departments’ overall citation figures, given the lower number of citations sampled for them.

Among the journals cited, fourteen titles from the list of twenty journals sampled by García-Romero et al. (2016) are represented both in the Gothenburg lists of twenty most cited (for the department) and ten (for the sections) journals. In Table 7, five titles from García-Romero et al. can be found. The aforementioned ranking focuses on the more general topic of Economics, with the exception of the elite financial journals. In comparison, the Economics department at the University of Gothenburg shows, through its most cited journals, the influence of the specialized interests of its research sections, such as Behavioral, Experimental, Development, Public, and Environmental studies. This indicates that even though a smaller group of journals needs to be provided by the library in order to sustain the department’s citation culture, awareness about the specific topics within the field is equally important.

The department of Economy & Society distinguishes itself through its heterogeneity. For this reason, the results of the Pareto analysis and the lists of most cited journals will be presented for each research section as seen in Table 8. The results of the analysis of the 20/80 rule for the departments of Economy & Society are very different from those of the department of Economics and Business Administration. Here, 20% of journals (110) account for 63.92% (1084) of all citations (1669). This ratio is more or less the same at the research section level, with Economic History at 63.56%, Human Geography at 66.94%, and Innovation at 72.50%. The Innovation section is closer to the 20/80 rule, and corresponds to the patterns of the Business Administration department, with which it shares many of its research interests. Clearly, the department of Economy & Society and two of its sections depend on a larger group of journals. This makes it more difficult to identify and define lists of core journals and, as observed by Nisonger (2008), other criteria might need to be taken into consideration in order to approach a possible central body of journals.⁶ In terms of collection management, serials acquisition, and renewal, the results for Economy & Society indicate that larger collections of journals within these disciplines might need more attention, in order to match scholars’ needs for variegated sources.

Table 8. Top ten most-cited journals for the research sections within the department of Economy & Society with SCImago ranking (SJR), Norwegian register (NSD) and Dewey classification (DCC).

Title	Citations	SJR	NSD	DDC
Economic History
Economic History Review	74	1.24	2	330
The Journal of Economic History	73	2.11	2	330
Scandinavian Economic History Review	63	0.56	2	330
Explorations in Economic History	49	1.15	2	330
European Review of Economic History	38	0.62	2	330.1
The Journal of African History	35	0.54	2	960
History in Africa	30	0.1	1	960
American Economic Review	29	11.889	2	330.9
Quarterly Journal of Economics	24	30.49	2	330.1
History of the Family	23	0.71	1	306.8
Human Geography
Journal of Transport Geography	82	1.66	1	910.02
Transport Policy	65	1.52	1	380.5
Progress in Human Geography	64	4.36	2	910
Journal of Rural Studies	59	1.62	2	333.72
Annals of Tourism Research	58	2.18	1	910.91
Regional Studies	40	2.32	1	354.35
Urban Studies	39	2.11	2	307.1416
New Media and Society	39	2.84	2	302.23
Information Communication and Society	37	3.21	1	003
Population, Space and Place	34	1.72	1	304.6
Innovation and entrepreneurship
Research Policy	226	3.41	2	658
Industrial and Corporate Change	63	1.5	1	330.1
Scientometrics	48	1.11	1	500
Academy of Management Review	46	9.31	2	658
Strategic Management Journal	46	8.83	2	658
Small Business Economics	45	1.91	1	338.642
Journal of International Business Studies	40	5.54	2	658
Industry and Innovation	39	1.61	1	338.005
Organization Science	31	6.54	2	658
Ny Teknik	28	N/A	N/A	N/A

Table 8, showing the ten most-cited journals for each research section, also displays new and unique patterns compared to the two previous departments. The SJR rankings are generally lower, with the exception of some Economics and Business journals. The publications are smaller and more specialized, and it should come as no surprise that their SJR ranking is lower than those of more established journals. This can also be observed in the NSD, since all three sections have a combination of 2 and 1 scores.

Two journals dealing exclusively with Economics, which rank highest in that department (see Table 7), appear in the Economic History top ten list, together with five of the thirteen titles chosen by Fernández-de-Pinedo and Muñoz (2019). Among these thirteen distinguished titles, the authors have also chosen journals with a clear geographic scope, such as Australia, India, or other. In Gothenburg, interests seem instead to rely more heavily on African history, with two titles having this focus in its most-cited list.

Titles for Human Geography display a more homogenous ranking, with no titles under 1.0 SJR points, while including fewer 2 s from NSD compared to Economic History. The research topics within Human Geography are many, and cover broad areas of interest, which is made clear by the Dewey classification and the journal titles: from geography to urban and rural studies, transport, and information sciences. Almost every journal covers its own area, whereas the classifications of Economic History almost exclusively feature 330 as their Dewey code. The complexity of Human geography, as well as the lack of citation and ranking analysis literature, also make it harder to establish what positions these journals occupy within their respective fields.

It follows that these results call for a more pronounced, multidisciplinary awareness throughout the library’s collection management, in order to ensure access to a broad range of titles (given the results of the Pareto principle analysis) from different disciplines, not necessarily exclusive to the Economics Library.

Finally, the Innovation & Entrepreneurship research section shares five titles with the Business top twenty list, and is generally management-focused. The Research Policy journal stands out, with a very high number of citations compared to the second position, as well as the largest gap between two journals across all research sections. As previously mentioned, the Research Policy focuses on many topics, one of which is innovation. Interestingly, a Swedish newspaper also appears in the last position, and even if Ny Teknik is not a peer-reviewed publication, it seems to be an important source for scholars wanting to survey technological news from a Swedish perspective (Table 9).

Table 9. Top ten most-cited journals at the Law department with SCImago (SJR) and Norwegian ranking (NSD), and Dewey classification (DDC).

Title	Citations	NSD	SRJ	DDC
Marine Pollution Bulletin	19	1	1.22	363.7
Common Market Law Review	14	2	1.61	341
Review of International Studies	11	2	1.61	327
Security Dialogue	9	1	1.96	327
Svensk Juristtidning	9	1	N/A	340
Journal of Political Power	9	1	0.31	300
Ambio	8	1	1.69	363.7
European Review of Private Law	8	2	N/A	341
Signs	8	2	0.27	305
Science	7	2	13.25	500

With 20% of journals (86) corresponding to 48.24% (361) of all citations (765), Law is the department that diverges most significantly from the 20/80 rule. Even more prominently than the department of Economy & Society, Law’s journal citations cannot be reduced to a clear-cut core list. Further, it is important to point out that the document type Articles, from which the lists have been generated, is specifically connected to serial publications such as journals and similar periodicals, but as shown by Cln and Dike (2016), the series of Law Reports are also significant to Law practitioners, and could have been excluded from the Articles category.

The list of ten most-cited journals shows the influence of the research section of Societal Challenge-driven research in journal citations, with the Marine Pollution Bulletin being the most cited.⁷ The presence of Svensk Juristtidning reflects the Educational-driven research section and its focus on local national conditions. As demonstrated by Cln and Dike (2016), and Burman (2011), the concerns of law practitioners, in terms of sources and citations, rely on other document types compared to article-heavy disciplines such as STEM fields, business and economics. Thus, this top-ten list should not be analyzed strictly in terms of ranking and output. Further, as showed by Ramsay and Stapledon (1997) in their review of Australian law journals, the local and national concerns of the law scholar are central to the field. This implies that more highly ranked international Law elite journals such as Journal of Law and Economics, Supreme Court Review and American Law and Economics (Eisenberg & Wells, 2014) do not necessarily have the same relevance to Swedish Law scholars. The journal Svensk Juristtidning is not ranked on SRJ, but is impactful and relevant nonetheless, an argument that Eisenberg and Wells (2014) have generally held to be true for many other examples of journals in the field of law which are underrepresented in these ranking databases. Additionally, the fact that the aforementioned journals and other similar Swedish publications are absent from the databases makes it harder to download citations lists, and to study the citation culture in greater detail.

One similarity across all four departments is that even for the two departments closest to the 20/80 rule, the 20% of most-cited journals still account for very large numbers of citations. This makes it difficult to establish a concentrated core, which, according to Nisonger (2007, 2008), is one of the main goals of Pareto analysis in a collection management context. Further, the vast number of research sections and their uniqueness also disrupt the idea of core lists, since the concept itself loses utility the more cores there are (2007). One alternative way to grasp more clearly the extension of which journals can be considered core is to set specified values corresponding to a top-ten list, or a defined percentage of citations that journals need to cover, as done by, for example, Hoffmann and Doucette (2012). In the case of this study, a value of between 20% and 40% would result in more readily grasped lists, which are also more comprehensive than a limited account of only the ten most-cited journals, and could be used in collection management decisions and evaluations.

Age analysis

The average age of citations for the overall Faculty of Business, Economics and Law is 14 years, with a median of 10. The most noteworthy differences occur when comparing the department of Economy & Society to the rest of the Faculty, with the average age of the citations of the former being 17 years, with a median of 10. Predictably, the research section for Economic History is the one increasing the average age of citations. It has a 28-year average and a 17-year median. Another interesting result is the average age of 12 and median age of 7 years of the department of Law, making its citations the youngest. Only the Economics department has an average age as young as 12 years, with a median of 9 years. Within Law, the research section of Societal challenge-driven research employs the youngest sources, which might be due to the section’s interests in current topics, such as Ocean Governance, Sustainability, and Gender, among others (Tables 10 and 11).

Table 10. Average age of citations from 2015 to 2018 per department.

Table 11. Median age of citation from 2015 to 2018 per department.

Finally, the Business Administration Department has a 13-year average and 9-year median, showing very similar results with other Business Schools such as Dewland’s 12-year average and 9-year median (2011), and Vaaler’s 12.45 years average and 9.6 years median.

Both Dewland (2011) and Tucker (2013) point out the relevance of awareness about scholars’ citation of older materials, and the need to sustain these through the purchase of back-filling electronic collections (Dewland, 2011, p. 152). This is equally true for the Gothenburg Faculty of Business, Economics and Law, and in particular for the research section of Economic History. Despite this section’s more accentuated need for older literature, the other departments and sections also cite material older than the authors expected to find. Undoubtedly, this is valuable knowledge for future work with collection management.

Holdings analysis

To determine journal holdings within the library catalogue, both printed and digital, the authors compared a list of 3280 unique journals from the citation analysis to an Excel export from the library’s SFX. Titles matching the SFX document were manually retrieved. Results show that the library subscribes to or provides access to 2938 titles, or 89.57% of all journals cited (Table 12).

Table 12. Access to cited journals and books at the Gothenburg University Library.

For Books citations, the authors chose to focus on a random sample of approximately 20% (1272) of all unique Books citations (6428), since the sheer numbers would have made the process of holdings analysis too time-consuming. Within this group of 1272 unique Books citations, the library provides access to 1044 of the titles, or 82.08% of the faculty’s Books citations. This percentage includes titles that are either published in print or digitally issued, and those that are both. The analysis of books holdings shows thus that the 82% result is similar or superior to Currie and Monroe-Gulick (2013), Gao (2015), Tavernaro and Salisbury (2015), Ke and Bronicki (2015) and inferior only to Vaaler (2018).

The results indicate that the library is effectively supporting the faculty’s needs for literature and sources. However, compared to other studies, the results for the percentage of journals to which the library is providing access is lower than Vaaler’s 96% (2018), as well as both Currie and Monroe-Gulick’s (2013) and Gao’s (2015) 92%. Therefore, it might be interesting for the library staff to analyze further the 10% of journals that are beyond the library’s reach. Other scholars, including Tavernaro and Salisbury (2015), and Ke and Bronicki (2015), limited their holdings assessment to a set list of the top most-cited journals, which resulted in a very positive outcome, with a high percentage of journal coverage. Similarly, the Gothenburg University Library shows a 100% incidence of access for the top twenty-five most-cited titles across all faculties.

Conclusion

The results of this study have contributed a great deal to expanding the library’s understanding of scholars’ use of resources and literature in the process of writing their publications. Expectations about what kind of documents are being used were both confirmed and subverted. Most notably, the Business Administration department stood out, with a high use of Books citations. Combined with the analysis of the most-cited journals, which revealed a pronounced focus on sociology, this establishes a similarity to the social sciences that distinguishes the Gothenburg Business Administration department from international counterparts analyzed in the study.

The Economics department proved to be more article-focused, as expected. At the same time, results also highlight the fact that Papers constitute an important part of citations, as the authors were aware, but it is nonetheless important to have this confirmed by rigorous analysis. Unsurprisingly, Economy & Society has a high incidence of Books citations, even though the three research sections displayed individual patterns. The Law department stood out the most, not only because Books were more cited than journal Articles, but also because of the wide variety of other document types represented, such as Gov. Docs, Reports and Online Resources. The understanding of the department’s citation culture would surely benefit from a more dedicated, future citation analysis, like some of the studies mentioned, focusing exclusively on Law.

In terms of collection management, one of the most immediate effects of the results is the opportunity to propose a potential reorganization of the budget in relation to the distribution of book expenditure. However, the results could also be used in other contexts, such as students’ teaching and coaching, where the relevance of gray literature like Dissertations, Papers, and Reports, among others, is sometimes underestimated. Furthermore, areas of study involving citations from dissertations or student theses could serve as complementary research to the evidence gathered in this study. For instance, Becker and Chiware (2015) examined both masters’ and doctoral theses across several departments, while Datig (2016) focused on undergraduate students. The Economics Library might benefit from following Datig’s example, since educational work is one of the library’s core activities. Datig’s method of using focus group interviews is another interesting starting-point for future research, that would expand upon the large amount of quantitative data sampled in this study.

Further, the analysis of the age of citations was enlightening, since it clearly shows that older sources are still relevant. This may change the common practice of the library to purchase literature a majority of which is between one and three years old, while resorting to interlibrary loans to acquire older titles. In addition, the purchase of journal archive collections derives more legitimacy from these results, relating to the fact that almost all faculty departments use large numbers of different journals. This is an interesting result of the 20/80 rule analysis, which, along with the examination of the most-cited journals lists contributes to an even deeper understanding of citation cultures. Overall, the Business Administration and Economics departments display a narrower use of journals, while Economy & Society, and Law, by falling outside of the 20/80 rule, depend on a broader range of titles. However, Business itself, despite sticking very closely to the Pareto Principle, has a very large number of journals among its 20% most cited titles. This makes it harder to identify core lists for three out of four departments, with Economics being the one with a somewhat discernible core. These institutional differences, however, are both noteworthy and useful, since they indicate how different approaches are necessary when library staff makes collection management decisions. The resulting data has proved a useful source of information, both in the case of lists of 20% most cited journals as well as shorter ones, including between ten and fifty titles, depending on what information the library staff might be interested in.

The analysis of most-cited journals for each department and for some of the research sections revealed interesting results, such as a high degree of affinities and differences across the faculty, as well as several instances of multidisciplinary interests. It also highlighted that, even though top-tier high-ranked publications are crucial and well-regarded across the faculty, many other journals are just as relevant and important to the scholars of the departments and research sections. Therefore, a contextualization of rankings is another necessary approach for librarians when evaluating which journals to purchase.

Finally, the assessment of library holdings and whether these are supporting scholars in their research was very informative. With a high percentage of coverage, the library is indeed providing access to the majority of literature cited. Further research into this area could provide a more in-depth understanding of citation culture. By looking at subjects, publishers, and other elements of cited books and journals—a common practice within citation analysis studies—the library could sample even more information to better organize its collections’ management. The in-depth study of sources that are not part of the local library catalogues is another fruitful area for further research. The Gothenburg University Library has a well-developed and well-used interlibrary loan system, which might already be helping scholars to access the 18% of books and 10% of journals used which are currently absent from the library holdings. Even from the point of view of holdings analysis, more dedicated studies involving a smaller number of publications and a wider range of variables could enrich the knowledge sampled in this paper. Further, it is important to consider what kind of relation can be found between the way scholars retrieve their sources, and what the library offers. To what extent is the library influencing citation patterns, and vice versa? Such questions could be an interesting starting point for qualitative studies that include interviews and focus groups as well as other methods, to seek answers directly from the scholars themselves.

The ambitious goal of conducting a simultaneous analysis on both a departmental and research section level across the whole faculty allowed for an approach where comparison produced new knowledge, and the value of a diverse citation culture proved to be crucial for the understanding of each citation culture. The Faculty of Business, Economics and Law is undoubtedly a dynamic environment where many different schools of research coexist. The research sections themselves are independent, with individual qualities and ways of using literature. This suggests that citation analysis conducted exclusively on the department or research section level could reveal even more specific and detailed data. The heightened awareness to which this study contributes has been the most valuable result, and it has shown how important it is for librarians to possess in-depth knowledge about scholars’ research practices.

Notes

1 The Economics library is one of the branches of Gothenburg University Library, with collections covering the subjects of Economics, Business, Law, Economic History, and Human Geography, among others. It is connected to the School of Business, Economic and Law, whose students and scholars make up the library’s main target groups.

2 This principle, when used within the field of citation analyses, describes a usage pattern where 20% of journals is expected to account for 80% of the articles, see Nisonger (2008).

3 The absence of certain titles from ranking databases or their low score when present should not necessarily be interpreted as a sign of low relevance or impact.

4 The author only uses Journals and non-Journals as document types.

5 Including Working Papers, Conference Papers and Discussion Papers, among others.

6 The fact that books constitute such a vital part of scholars’ citations also makes the definition of core journals more problematic.

7 As mentioned above, the topic of Ocean Governance is one of the research concerns of the section.