1 Introduction

Knowledge is widely acknowledged to be one of the main engines for economic and social development of a country [³³, ³⁵]. Universities and research centres, either public or private, play a key role in both generating and disseminating this knowledge [²⁸, ⁵⁴]. Through the research mission, they generate cutting edge discoveries, expanding the boundaries of science, while the “third mission” implies the dissemination and exploitation of this knowledge, contributing to social growth and economic development [³, ²², ⁵⁸]. Moreover, knowledge spillovers stimulate other research institutions to commercialize their research findings resulting in acceleration of economic growth. The establishment of university-business collaborations (UBC) is therefore central to this process in order to facilitate this knowledge flow from academia to industry [²¹].

Reasons for engaging in UBC have been widely documented in the literature. From the standpoint of universities, an important body of the literature has examined the function of the university and their departmental characteristics [⁵²], as well as the technology transfer infrastructures that boost the commercialization of research results [⁴⁴, ⁶¹]. Other studies have shown that scientific productivity of a researcher is positively related to academic engagement in the industrial domain [⁸, ³¹, ³²]. Going a step further, several works also found a positive connection between the quality of the research conducted and the probability of researchers participating in commercialisation activities [²⁵, ⁵¹, ⁶²].

From the industrial side, knowledge sharing between science and industry is a preliminary step for innovation [³⁷]. In order to leverage markets and remain competitive, companies need to be continually alert of new developments. In this respect, universities offer firms a wide access to a variety of experts in various disciplines as well as appropriate infrastructure and state-of-the-art technologies, which can shorten the life cycle for industrial products [⁵⁷, ⁶⁶, ⁶⁸]. Partnering with universities not only provides firms with an adequate cost-benefits extension of the R&D resources (cutting-edge knowledge, equipment, facilities), but also helps firms to take advantage of emerging potential business opportunities, and evaluate the quality of the research into the company and update internal capabilities and skills [³⁰]. Partnering with university scientists is also beneficial for companies, as this form of alliance gives legitimacy to research results [³⁶].

Furthermore, universities can also conduct new research in specific fields that are of interest for firms [¹⁵]. Aiming at narrowing the gap between science and industry, many universities have created specific units and designed explicit programs to assist in this endeavour. Technology Transfer Offices (TTOs) and science parks are two clear examples. Acting as knowledge brokers, they are a bridge between academics, companies and venture capitalists, and their main objective is to facilitate the process of knowledge and technology transfer between the university and the company while infusing an entrepreneurial culture of research [¹⁷].

Nevertheless, despite the great support given to UBC, there are still significant impediments limiting their potential. Barriers mainly relate to the motivation incentives that drive academics and firms to engage in such a partnership. Science and industry operate differently. Their daily activities are highly tied to a specific organizational culture, mission and organizational practices [⁵⁹]. Accordingly, goals might signal opposite directions. First, companies cannot evaluate the quality of the invention a priori, and researchers may have difficulties in evaluating the commercial profitability of their inventions [⁴⁵]. Second, poor communication channels and low interest in academic research are other reasons that prevent universities and business from cooperating [⁶]. Universities have strong incentives for conducting basic research that leasing to new knowledge. On the other hand, industries seek solutions that make their operations and processes more competitive, their products more attractive, and consequently enable them to become more profitable [¹, ⁵⁵]. Third, time-span is another critical factor. University research projects tend to require long periods of time, while industry demands short cycles in order to compete in the market and obtain competitive advantage [¹⁶, ²⁶, ¹²].

Literature on UBC is abundant. The different stakeholders involved in this process (i.e., the universities, firms, and individual researcher) might explain this large corpus of both theoretical and empirical studies, as different approaches, and different points of view have been explored [²²].

Another explanation for this diversity relies on the variety of forms in which UBC materialize, ranging from casual interactions in events such as meetings, conferences, recruitment of university graduates, or staff mobility, to more sophisticated agreements such as collaborative joint research, contract research, consulting, consortia, alliances, trade associations, interlocking directorates, equipment or other facilities [²², ⁷].

Although this list is quite exhaustive, scholars converge on the difficulties of categorizing all potential mechanisms for UBC to take place [¹⁰]. All these mechanisms had been classified into six categories: (1) personal casual relationships, (2) personal relationships, (3) third party, (4) formal targeted agreements, (5) formal non-targeted agreements and (6) creation of focused structures [¹³]. Similarly, had been distinguished seven tools through which to strengthen UBC: (1) joint curriculum design and delivery, (2) lifelong learning, (3) student mobility, (4) professional mobility, (5) joint R&D, (6) commercialization of joint R&D, and (7) entrepreneurship [²³]. Whatever the mechanism used, the formalization agreement is a very important step, as it monitors and regulates the relationship, avoiding conflict and mistrust between the parties [⁵⁶].

Today, research in UBC enjoys good health and can be considered a recognised scientific field, with a myriad of researchers studying from different theoretical and practical perspectives.

Many research institutions; associations and international networks have also been created based on UBC at their core.

Some examples include the Triple Helix Association (TH) and the University Industry Interaction Network (UIIN). These associations organize seminars and conferences that constitute unique vibrant forums where academics and practitioners discuss the newest advances in this field. Similarly, specific journals and conferences have emerged aiming at providing a forum for discussion. Some specific journals that clearly deal with this topic include: Research Policy, Technovation, Journal of Technology Transfer, Journal of Engineering and Technology Management, R&D Management and the International Journal of Technology Management.

Given the amount of research generated around the different mechanisms through which to articulate technology transfer processes between universities and business, there is an urgent need for reviewing the state of the art from its theoretical inception, in the early 1980s, to the present.

We had decided to start this analysis from 1980 due to the fact that in that year the Bayh-Dole Act was enacted in the USA, which allowed universities to appropriate patents derived from public funds. In the same line, researchers working on public funds were stipulated to disclose their inventions to the technology transfer office [⁹, ⁵⁰].

This Act fostered the commercialization of university research [³⁸, ⁴²]. It is important to note that since this relevant Act, other countries outside of the USA such as Denmark, Germany, Austria, and Norway have reformed their intellectual property laws by giving universities ownership of publicly funded inventions, similar to the Bayh-Dole Act. [⁶⁴].

By using an extensive range of bibliometric indicators, in this study, we identify the most influential journals, authors, and papers.

We also analyse which countries and research institutions are taking a leading role in this particular field. The information was obtained from the Web of Science (WoS) database, considered one of the most relevant databases in the academic domain. Some prior studies have adopted a similar approach, but concentrate on specific geographical areas.

Had been presented a study based on joint scientific publications between universities and industry in the UK, covering two decades (1980-2000) [¹⁸]. Similarly, there is an analysis that examines public and private research collaboration between universities and industry in Italy for the period 2001-2003 [²]. In the same line, but without using bibliometric indicators, had been analysed academic articles in the field of UBC, starting in 1990 and ending in 2014 [⁵]. Our study differs from previous ones for considering the most current tools to represent a research area with bibliometric indicators, and for adopting a global geographical perspective [³⁴, ⁵³].

The remainder of the paper is organized as follows. We first describe the methodology. The next section, presents the results, including the most influential journals, the most cited papers, the most productive and influential authors, the main institutions in this area of knowledge and the most relevant countries. Next, we discuss the main implications. The paper ends with some concluding remarks alongside indicators for future works.

2 Methodology

The search process takes as a basis the Web of Science (WoS) database. Despite the fact that other databases could have been used (e.g. Scopus, EconLit, Google Scholar) we selected this database as it has been acknowledged to beof high quality and one of the main sources of citation information in the world [⁶⁷, ⁵³]. WoS includes more than 15,000 journals and 50,000,000 articles that encompass all the known sciences [⁴⁸]. Information is classified into research categories, research areas, articles, authors, journals, institutions, and countries.

Today WoS distinguishes 250 categories that are grouped in 150 areas. For the purpose of this study, the focus is given to the WoS Core Collection, which covers 12,000 of the most recognised journals worldwide as well as Open Access journals in the sciences, social sciences, arts, and humanities, with coverage since 1900. Despite The Social Science Citation Index (SSCI) have received some criticism about ideological bias in journals' inclusion, it could be just a result of chance [⁴⁰].

The study was done via the web, and it was consulted less of a quarter of the total journals indexed in SSCI. The first step in the search process was the identification of meaningful keywords that unequivocally return papers that fall within the topic of our interest: technology transfer links between academia and industry. We were therefore interested in papers that contained at least two different ideas: a collaboration or partnership agreement (mechanism articulating the technology transfer process), and the actors involved (industry and university).

Because literature has referred to these concepts using a variety of terms, we first elaborated a list containing all potential synonyms (see Table 1).

This process was indispensable to determine the inclusion/exclusion criteria to apply in the bibliometric study.

The search was conducted during July and August 2016. The inclusion criteria for accepting papers were: a) document type: article or review, b) language: English, c) timespan: all years, d) indexes: SCI-EXPANDED, SSCI, A&HCI, CPCI-S, CPCI-SSH, and ESCI. This research strategy returned an initial set of 696 records, from which 673 were journal articles and 23 reviews.

It is worth noting that these publications refer to the period comprising 1980 to 2016 (see Figure 1), coinciding with the enactment of the Bayh-Dole Act in 1980.This reform introduced important modifications to how universities could exploit new technologies and research developments created in the university setting, having significant political and management implications for all those involved in the commercialisation of university research results [⁶⁰, ⁴³]. Aiming at identifying the most influential journals, the most relevant articles, the most productive authors, as well as the leading institutions and countries researching in this particular field, we used several key indicators for measuring the bibliographic material. The purpose is to give a general overview of the bibliographic material [¹⁴].

Fig. 1. Evolution of papers published by year in university business collaboration and technology transfer from 1980 to 2016 

First, we relied on the citations/paper ratio, which permits to identify the most influential articles [⁴⁸]. Second, we used the h-index a measure that integrates publications and citations in the same formulation, by connecting the number of papers n that has received n citations [³⁴]. This index measures the productivity of a researcher and total impact of the papers.

Thus, researchers with a similar h-index are comparable at the level of scientific impact, although their number of articles and citations differs. (e.g., if a researcher has an h-index of 50, it means that he has a set of 50 articles that have received at least 50 or more citations. The h-index can be utilised in authors, countries, journals, articles, and universities.

Thus, it allows making a holistic analysis of a certain field of research, taking into account several different items [¹¹]. The Impact Factor (IF) for 2015 has been included in Table 3 to present the quality of each journal.

Ranking is development according to percentage of University Business Collaboration (UBC) and Technology Transfer papers in the Journals published between 1980-2016 at Web of Science Core Collection. The total records found during this period were 696

The impact factor analyses the influence of a journal over a two-year period, dividing the total citations received in year n-1 and n-2 by the total number of articles published in year n-1 and n-2. The impact factor in recent years has received numerous criticisms because it has been argued the ease of manipulation of results through self-citations or related techniques [¹⁹]. Impact Factor considers the two previous years. However; longer time periods of citations and/or sources could be considered, but then the indicator would not be current enough [²⁹].

The percentage of papers in UBC of any given journal (TP-UBC/TP) in Table 3 is also included. To evaluate the citation rate of papers in UBC, in Table 7, The countries information is classified according to citation structure with thresholds of, >250, >100, >50 citations.

In order to map the bibliometrical material we employed Science mapping. This science is considered a specific science, in which the different scientific fields are structured in a conceptual, intellectual and social mode [²⁰]. Additionally, the study uses VOS viewer software that provides easy-to-interpret graphical representations of the bibliographic material and has the functionality to construct maps based on citation, co-citation co-authorship, bibliographic coupling and co-occurrence data [⁶⁵, ⁴⁷]. Bibliographic coupling takes place when two articles cite the same third article [³⁹].

Co-occurrence identifies the most common keywords used in the articles.

This list of keywords generally can be founded on the first page of the paper [⁴¹]. Figure 3 and Figure 4 present the mapping of the most cited authors between 1980-2016 and 1917-1918. The mapping is focused on authors so we look for citation, co-citation, and co-authorship with a threshold of fifty cites and one paper.

Fig. 2. Top-20 Research areas in university business collaboration and technology transfer from 1980 to 2016 

Fig. 3. The Most cited authors from 1980 to 2016. Colours indicate the citation impact of different authors, and the size of the node, the productivity of them. The link width shows the citations frequency between authors 

Fig. 4. The most cited authors from 2017-2018. Colours indicate the citation impact of different authors, and the size of the node, the productivity of them. The link width shows the citations frequency between authors 

Citation analysis counts the number of times that document A cites document B and vice versa [⁴⁹]. Co-citation results when the study A and study B receive a citation from the same C study [⁶³]. Co-authorship identify the scientific output of researchers as well as their joint publications [⁴⁶].

The graphical visualization is showed through network where the size of the node increases with the number of publications and the network connection the relationship between them. VOS is available free of charge, and more information about the program can be obtained at the following link: http://www.vosviewer.com.

Table 2 shows the research categories in which the 696 records fell. Only the top 25 research categories are displayed, however, they cover almost the entire sample (95.97%). The category with the highest number of articles is Management (334 articles), followed by Development Planning (137 articles), Industrial Engineering (105 articles) and Business (89 articles).

Concerning the research areas, Figure 2 reveals that Business and Economics accounts for 57.47% of the total volume, followed by Engineering (26.72%), Public administration (20.69%), and Education & Educational Research (11.92%). Overall, these results mirror the Triple Helix model of university-industry-government relationships [²⁷].

Indeed, the most recurrent categories are business (industry), education (university) and public administration (government).

3 Results

This Section summarises the main results of this paper. First, we provide a comprehensive analysis of the most influential journals in the domain of technology transfer processes aimed at fostering university-business collaborations. The analysis is limited to the top-20 journals. Second, following a similar strategy, we concentrate on the most cited articles. Next, we focus on the most prolific authors. Lastly, we study which institutions and countries act as drivers in this particular field.

4 Discussion and Conclusions

By adopting a bibliometric approach, this study contributes to the current literature by giving a global view of the academic research in technology transfer mechanisms through which UBC can be established. We argue that our study overcomes the limitations of previous studies that adopted a similar approach, by not only focusing on a specific country or territory, but also adopting an international perspective. The period of analysis considers publications from 1980 up to 2016 due the fact that in 1980 the Congress of USA enacted the Bayh-Dole Act, which eased the commercialization of university research and thus university-business collaboration as well [³⁸, ⁵⁰, ⁶⁴]. The results show that the USA, England, Italy Spain and the Netherlands are the leading countries in this area, and all of them have shown a significant increase in their production over the last 10 years. The analysis also reveals that the major categories in which these publications fall have a strong focus on business, public administration and education.

The logic behind this lies in the fact that these areas are the ones more closely related to growth and economic development.

Regarding the study of the leading institutions, the countries hosting them are Spain, the USA, Belgium, and the UK.

Top institutions include the Consejo Superior de Investigaciones Científicas (Spain), followed by The Imperial College London (UK) and KU Leuven (Belgium). All these institutions have a similar h-index, thus the productivity and impact factor are fairly comparable. Concerning the analysis of the most cited researchers, Perkmann, D’Este, Muscio, Salter and Leydesdorff (co-author of triple helix model) stand as important big names in this area, publishing their works in some of the highest standing journals in this field, such as Research Policy, Technovation and the Journal of Technology Transfer.

We believe this study can help not only academics but also practitioners to identify the leading authors, institutions, and most influential publications in this field. This research might be particularly useful for anyone interested in obtaining a global picture of the current literature on UBC, particularly PhD students or researchers that start investigating this topic. Although we have followed a rigorous methodology, the study is not free of limitations. The bibliometric approach of this study is only informative. The analysis is biased to the records displayed in only one database, the Web of Science.