1 Introduction

Founded in 1999, Information Systems Frontiers (ISF) is a respected and highly cited journal. With a 2017 impact factor of 3.323, the journal has emerged as a quality outlet for publishing IS research by scholars from all over the world. The journal is highly ranked in various ranking systems. For example, Scimago Journal & Country recently upgraded ISF rank from Quarter 2 (Q2) to Quarter 1 (Q1) in Information Systmes, and the Australian Deans of Business Council (ABDC) ranks ISF in the ‘A’ category. ISF has enjoyed a consistent increase in the number of published articles from 1999 until today.

Motivated by the 20th anniversary of ISF, this paper reviews ISF papers between 1999 and 2018. This uncover the trends that have had more influence. The analysis considers a wide range of issues including the publication and citation structure of the journal, most-cited papers, the most influential authors, institutions and countries, and citing articles. Importantly, the paper presents a thematic analysis of the publications that appeared in ISF in the past 20 years. The thematic analysis is evidenced by two sources of data: First, a bibliometric analysis highlighting core topics within the past 20 years is presented. Second, a semantic analysis of keywords introduced by the authors themselves is applied. This will show that ISF is steadily concerned with core IS topics such as Enterprise Architectures and IS use and at the same time it is maintaining currency with emerging topics of increasing importance such as IS for social good, analytics and Internet of Things (IoT).

The aim of this paper is two folds: First is to provide an overview of the bulk of ISF papers. Second is to uncover trends in an evidence-based fashion. We use a bibliometric analysis to fulfil the first aim and at the same time provide early evidence towards uncovering key trends. For the IS community, uncovering trends in the emphasis across IS would be most instructive. Towards this, the paper will first provide a bibliometric analysis that is essentially a data extraction process. This will be followed by a correlation analysis supported by a visual analysis. Finally, an IS-based thematic analysis using keywords of each publication will be used to achieve the overall aim of the paper.

Hence, this paper is organised as follows: Section 2 presents the bibliometric methods and descriptive results. Section 3 presents the correlation analysis and a graphical mapping of the bibliographic material of ISF by using the visualization of similarities (VOS) viewer software. Section 5 analyzes the thematic structure of the journal, and Section 6 ends the paper summarizing the main conclusions and findings.

2 Bibliometric Analysis

2.1 Bibliometric Methods

In order to develop the bibliometric analysis, we first define the bibliometric indicators that we used in the analysis. Various indicators have been used in the literature (Alonso et al. 2009; Ding et al. 2014). This study aims to provide a simple and complete overview of the publications of the journal. Various interests and potential perspectives of readers are accomodated. There is no single indicator that can totally represent the results of a set of documents. Each expert may evaluate the results in different ways according to his or her interpretation. Some experts may give more importance to the number of publications as a measure of productivity while some other ones may give more relevance to the number of citations as a measure of popularity and influence. Many other indicators could also be considered. This work aims to provide different perspectives by presenting each analysis considering several representative indicators. It focuses on the number of publications and citations, the number of cites per paper, citation thresholds and the h-index. In some specific cases, the work also considers some other indicators including the world university ranking in the Shanghai and QS rankings and the results per capita at the country level.

The search process uses the Web of Science Core Collection database and analyses all ISF publications between 1999 and 2018. ISF has been available in Web of Science since 2001. Hence, the search identifies all results between 2001 and 2018. Nevertheless, for documents published between 1999 and 2000, the work develops a manual search by using the Cited Reference Search tool of Web of Science Core Collection. Here, the search identifies all the documents that have received at least one citation in Web of Science Core Collection. In addition, for those documents that have zero citations in Web of Science, the study identifies them through the webpage of the journal (https://springerlink.bibliotecabuap.elogim.com/journal/10796).

2.2 Bibliometric Results

ISF has published 855 documents between 1999 and 2018 if only considering articles, reviews and letters. Note that ISF has also published approximately eighty documents classified as editorials and corrections. Since these documents are not strictly a scientific contribution, they are not included in the bibliometric analysis. The 855 documents considered in the analysis have received 9888 citations up to august 2018. Thus, the cites per paper ratio of ISF is 11.56. The h-index is 43. That is, of the 855 documents, 43 have received 43 citations or more and the same time there are not 44 documents that have received 44 citation or more. Figure 1 presents the annual number of papers published each year in ISF. This increased from around twenty articles per year during the first years to eighty documents in the last years.

Fig. 1
figure 1

Annual number of papers published in ISF

Full size image

Table 1 presents several annual citation thresholds that the papers published in ISF has reached over time. Currently, 1% of the articles published in ISF have obtained more than one hundred citations and 4% more than fifty citations. Around 60% of the papers have received at least five citations and more than 95% of the documents at least receive one citation. ISF publications are clearly having a strong academic impact. The most cited papers published in ISF according to Web of Science Core Collection are shown in Table 2.

Table 1 Annual citation structure of ISF
Full size table
Table 2 The 20 most cited documents in ISF
Full size table

As Table 2 shows (full list of references is shown in Appendix), the paper by Klaus et al. (2000) on enterprise resource planning (ERP) has attracted the highest citation. This paper was one of the earliest studies in the IS field that tried to synthesize the literature to explore what and their benefits ERPs are about. Ross and Vitale (2000) paper on ERP ranks forth in the list and highlights that information systems integration is still a relevant topic to IS scholars in 2019.

Both Klaus et al. (2000) and Ross and Vitale (2000) papers made a number of foundamental suggestions for guiding research as well as teaching, which led to a lot of attention within IS and other diciplines. Furthemore, while just recently published in 2015, the papers by Whitmore and colleagues and Li, Xu, and Zhao on the Internet of Things rank second and third in the above table, which highlights the overwhelming interest in this topic.

Table 3 presents the 20 most cited documents in ISF. Fornell and Larcker (1981) paper on evaluating structural equation models with unobservable variables and measurement error is the most highly cited publication in this table. This paper is actually very highly cited in many other diciplines, thanks to its foundamental and practical guides on the use of structural equation models for business, technology, and social science research. Next in the list is Davis (1989)‘s popular Technology Acceptance Model (TAM), followed by Hevner et al. (2004) paper on design science in Information Systems research, which undoubtedly have inspired and impacted many studies in IS and other diciplines over the past decades.

Table 3 Top 20 most cited documents in ISF publications
Full size table

From the journal point of view (right side of Table 4), the self-citations of ISF are the most common ones which is a very common result for most of the journals. Other journals citing frequently ISF are Computers in Human Behavior, Decision Support Systems, Expert Systems with Applications and Enterprise Information Systems. Whilst most of the top citing journals are IS or IT journals, it is also worth noting that some other journals with a broader scope, specially in computer science, also appear among the Top 30.

Table 4 Citing articles of ISF: countries and journals
Full size table

3 Correlation Analysis and Graphical Mapping

VOS (Visualisation of Similarities) viewer (Van Eck and Waltman 2010) is a computer software that collects the bibliographic material and builds graphical maps by using bibliographic coupling, co-citation and co-occurrence of keywords (Blanco-Mesa et al. 2017; Cancino et al. 2017). We use VOS to produce a graphical mapping of bibliographic couplings in ISF. Recall that bibliographic coupling occurs when two documents cite the same third work (Kessler 1963). Co-citation appears when two studies receive a citation from the same third work (Small 1973). Co-occurrence of author keywords measures the most common keywords and those that appear more frequently in the same documents (Laengle et al. 2018).

First, let us consider co-citation of journals. That is, when two journals receive a citation from the same third journal (Merigó et al. 2018). The graph visualizes the most cited journals among all citations generated by ISF and the networks represent the co-citations between journals. Figure 2 presents the results for ISF considering a minimum citation threshold of 20 citations and visualizing the one hundred strongest co-citation links. Excluding the self-citations, MIS Quarterly continues to be the journal most cited in ISF. Some other highly cited journals in ISF are Communications of the ACM, Information Systems Research, Decision Support Systems and Management Science. A more interesting coupling analysis is on the most common keywords of ISF. This is very useful in order to see the leading topics that are published in the journal. For doing so, the work uses co-occurrence of author keywords (Gaviria-Marin et al. 2018) in order to identify the most frequent keywords in ISF and connect through a network visualization those keywords that appear frequently in the same articles. Figure 3 presents the results considering a minimum threshold of 3 occurrences and the one hundred strongest connections.

Fig. 2
figure 2

Co-citation of journals in ISF

Full size image
Fig. 3
figure 3

Co-occurrence of author keywords in ISF

Full size image

Knowledge management is the most common keyword used by the authors themselves to tag their articles. Other popular keywords are Security, Trust, Privacy, Semantic Web, and Webs Services. We used these keywords (shown in Figs. 3, 4 and 5) to identify higher level themes shown in Table 5. Before we examine these themes, it is instructive to further analyse the evolution of the keywords themselves. We examine them considering two different periods: 2003–2012 and 2013–2017 shown in Figs. 4 and 5 respectively. The figures show the results considering a minimum threshold of three occurrences and one hundred co-occurrence links.

Fig. 4
figure 4

Co-occurrence of author keywords in ISF: 2003–2012

Full size image
Fig. 5
figure 5

Co-occurrence of author keywords in ISF: 2013–2017

Full size image
Table 5 Common themes in ISF: General and temporal
Full size table

Figures 4 and 5 show the dynamic and changing state of the journal over their respective periods. During the last five years, many new emerging keywords, especially those related to modern technologies, have appeared in the journal. A comparison of the most frequent keywords reveal that the focus of IS research has shifted from Knowledge Management, Information Sharing, and Business Process Management research and in the 2003–2012 period to Open Data, Business Intelligence, Social Media, and Intenet of Things in 2013–2017 range. Interestingly, keywords related to Security and Privacy have almost equally received authors’ attention in both Figs. 4 and 5. This reflects the importance of research on IS security, and suggests that there are still various problems in this domain worth investigation.

4 Thematic Analysis Based on Keywords

We generalized the keyword analysis into a thematic analysis. A total of 3200 keywords as well as the number of their frequency of use were extracted and recorded in an Excel file. Keywords in our context are empirical attributes of the papers (as used by the authors in the ‘keywords’ list). In the earlier analysis paper (Bang 2015), keywords were synthesized using semantic analysis tools on the introductions of the papers. In our analysis, semantic analysis is applied to synthesize the themes, as described in what follows. Two IS experts, first and second author, undertook a semantic analysis on these keywords. This consisted of the following four steps:

  • Given that the majority of keywords only occurred once in our list, it was important to group them for extracting high level themes. Thus, each expert independently went through the keywords. Keywords with frequencies higher than 4 were selected for identifying high level themes. The keywords used were the ones identified by the authors by the paper (and used in Scopus for indexing).

  • The experts then grouped keywords into ‘themes’ based on their similarities. Each expert independently chose a name for each theme.

  • The experts then met and discussed disagreements. A reconciliation took place, which led to 10 themes with their underlying keywords.

  • Next, these themes were used to group additional keywords occurred from 1999 to 2017 to further increase the coverage beyond the cut-off of frequency of 4.

The list was turned into a 2 D structure (the Table 5 below). A detailed list of top keywords within each theme is shown in Appendix.

As the above Table shows, IS Use has ranked the highest in terms of frequency of the corresponding keywords during the past two decades. This reflects that despite recent changes in the technology space, IS scholars have been steadily concerned with core IS topics over the life of the journal. Other stable streams of research in ISF are Operations Management, Knowledge Management, and Enterprise Architectures; this represents a firm interest in researching these topics in the IS and ISF community over this period. This is also evident from Table 2, as the papers by Klaus et al. (2000) as well as Ross and Vitale (2000) on ERP and Carter and Weerakkody (2008) on IT Use are amongst top cited papers in ISF.

On the other hand, interest in IoT and Ubiquitous Computing, IS Risk Management, Analytics, and Web of Future have enjoyed the sharpest increase over the same period of time among contributors. This is evidenced by Whitmore et al. (2015) and Li et al. (2015) listed as top five cited papers in ISF in Table 2 on the subject of IoT as well as Ziegler and Lausen (2005)‘s paper on social media and trust as one of the top ten cited papers in this list. This shows that ISF is maintaining currency with emerging topics of increasing importance. The consistent increase of keywords within these themes suggest that interest in these topics will continue to the near future, and more studies can be expected within IS security, data analytics, IoT, and social media domains in the coming years.

Furthermore, the journal is increasingly becoming interested in papers that focus on IS societal good, and published a number of articles and special issues in this context. In the April special issue on societal impacts of big data and analytics, Gupta et al. (2018) proposed a simple framework that conceptualizes analytics for societal impact through (i) data and infrastructure, (ii) techniques for big data analysis and interpretation, and (iii) societal application domain, including areas such as environment, disaster response, life style, and resource management. In its following issue, the journal also examined the good, the bad, and the ugly aspects of social media platforms. It brought together a range of disciplines for the development of knowledge regarding the adoption, use, impact, and potential of these platforms (Dwivedi et al. 2018). This covered issues such as self-disclousure, sharing political content, and impact of social media on consumers’ acculturation and purchase intentions. The journal also hosted two special issues on the important topics of Distaster Management (DM) and Emergency Releief. The focus of the former was on how we use theories to provide assessment methods of DM readiness, and what are the theoretical challenges that are associated with integration of social media in DM information systems (Beydoun et al. 2018; Abedin and Babar 2018). The latter issue (Ghosh et al. 2018) examined the empirical results of various social media technologies such as Twitter, Facebook and WhatsApp for extracting vital and actionable information from social media content in disaster situations (Ghosh et al. 2018).

The journal maintained a rigourous IS focus. In fact, the number of studies concerned with IS research methods and theories have had a gradual increase in the past twenty years. This reflects ISF community’s increasing awareness and appreciation of theories and methodologies in IS research. A deeper analysis of keywords within this theme shows that case strudy, simulations, and design science are particularly popular methods of analysis amongst ISF authors.

5 Conclusion

This paper examined the focus of ISF over the past 20 years. The analysis revealed that ISF topics are largely into two groups: one group represents the anchor of ISF within the traditional IS themes. Another group reflects ISF research adapting and responding to key trends. Within the anchor group, are these topics: Operations Management, Knowledge Management, and Enterprise Architectures. This dichotomy was also reflected in an earlier study, Dwivedi et al. (2009), which examined a decade of Information Systems Frontiers research. The new themes we identified include IoT and Ubiquitous Computing, IS Risk Management, Analytics, and Web of Future. Interestingly, some of these were also subject of special issues flagged in that older survey Dwivedi et al. (2009) (e.g. Information Dynamics, Cyber Law, Knowledge Discovery in High Throughput Biological Domains, and Secure Knowledge Management).

Indeed, credit to the editorial team, this shows an underlying strategy to ensure that ISF is maintains currency and a firm IS identity. Indeed, the IS identity has also been reinforced lately in an increase in the number of studies concerned with IS research methods and theories have had a gradual increase in the past twenty years. Furthermore, the journal is increasingly becoming interested in papers that focus on IS societal good, and has published a number of special issues on topics such as how IS can contribute to emergency relief and disaster management, the bad, good, and ugly side of social media, and the societal impacts of big data analytics. We anticipate that these topics will feature more strongly within the next decade of ISF.