Tracing collaboration beyond the confines of academia through co-authorship of publications in the humanities

Eline Vandewalle^*, Raf Guns^** and Tim C.E. Engels^***

Centre for Research and Development Monitoring (ECOOM), University of Antwerp, Belgium

^*Eline.Vandewalle@UAntwerpen.be

0000-0002-9852-3373

^** Raf.Guns@UAntwerpen.be

0000-0003-3129-0330

***Tim.Engels@UAntwerpen.be

0000-0002-4869-7949

In this study, we use affiliation information for co-authored publications in the Flemish humanities to trace collaborations between people affiliated with universities and people affiliated with organizations outside of the higher education sector (non-academics). The data was collected from the Flemish regional research database VABB-SHW. We show how collaboration patterns with non-academic organizations differ between different humanities disciplines. We also consider the difference between people with dual appointments (both academic and non-academic) and “true non-academics” (non-academic affiliation only). Our preliminary results show that academics from the humanities do collaborate with non-academic organizations and that the collaboration patterns differ between the various disciplines. As the current paper presents a research-in-progress, we also discuss the directions the research will follow.

1. Introduction

Academic publications are most frequently (co-)authored by people affiliated with universities. However, there is also frequent collaboration between academics and organizations outside of academia. There is already some literature on collaborations between academia and industry (Abramo et al., 2011; Kotiranta et al., 2020; Meyer-Krahmer & Schmoch, 1998), mainly for STEM fields. In studies of university-industry relations, the focus tends to be on the commercialization of knowledge. This can be studied through patent data or spin-off contracts for example. Few studies have analysed collaboration between scientists and people working in the non-profit sector (Čada & Ptáčková, 2013).

Knowledge exchange in the humanities, however, is not primarily directed towards industry. The types of organizations collaborated with are varied, including companies, but also the government, NGOs, museums, archives and others. In this study, we focus on patterns of co-authorship between academics from the humanities and non-academics. Collaborations in the form of joint publication may indicate instances whereby research is put to practical effect, to be used by professionals. They may also indicate instances of the incorporation of knowledge gained through professional experience. The flow of knowledge between the academic and the professional world is not unidirectional. Furthermore, a study of co-authorship with non-academics may showcase the ways in which academics interact with companies, civil society and the government when conducting research. The various humanities disciplines may have a different profile of collaborations with non-academics, which can highlight the interactions between academia and society. Studies on publication patterns in the humanities have shown that the humanities are different in terms of publication culture to other sciences, but also that they are different from each other (Engels et al., 2012).

In this study, we are conducting an exploratory data analysis with the VABB-SHW database (henceforth VABB), a Flemish regional database for the social sciences and humanities. Included in the analysis is an overview of the types of organizations, the different disciplinary profiles in terms of collaboration patterns and the characteristics of publications written in collaboration with non-academic organizations in terms of language and publication type. We also discuss to what extent authors are affiliated with both an academic and a non-academic organization (dual appointments) or are uniquely affiliated with a non-academic organization. In the discussion section, we further reflect on the implications of this study.

2. Data and methods

We use affiliation data on the author-publication level to pinpoint instances of collaboration between academics and professionals in Humanities publications in Flanders. The data for this study was granted through the VABB, a Flemish regional database constructed in the context of a performance-based research funding system (PRFS) (Aspeslagh et al., 2022).

The VABB contains affiliation information for publications of more than one author from the Social Sciences and Humanities from 2011 onwards. Even though the VABB database contains both peer-reviewed and non-peer-reviewed publications, the collection of affiliation information was focused on peer-reviewed publications. The affiliation data was initially collected on the publication-level. The coding of organizations is based on the organizations of ROR (Research organization registry) where available¹. However, as many organizations, especially smaller local ones, are not found in ROR, additional organizations were added during the coding process. For an overview of the data collection process, see Aspeslagh, Engels and Guns (2021).

In order to create a useful subset of publications that could be considered instances of collaboration between academics and non-academics, we took all publications from the category “education” as academic organizations. We consider universities and schools of higher education to be “academic” organizations. These fall under the category “education” of the ROR organizational typology. In total, there are 8944 publications from a Humanities discipline that have been coded for affiliations. 2181 of these publications were co-authored with at least one author affiliated with a non-educational organization. For this study, 1146 of these publications were additionally coded on the author-publication level (these publications all have at least one author affiliated with a non-educational organization from Belgium).

The VABB includes metadata on the authors, the discipline, the publication type and the language of publishing. Publication types included are book chapter, edited volume, monograph, journal article, and conference proceeding. For the disciplines, we include the discipline Law as a humanities discipline, even though Law is also frequently categorized among social sciences or as a separate field, e.g. van Leeuwen (2013).

This study is based on an exploratory data analysis of the affiliation data of humanities publications. The main variables discussed are: organization type, discipline, language and publication type. The first part of the analysis was conducted on the publication level, with data on the types of organizations, disciplines and publication types. The second part of the analysis, on dual appointments, was conducted on the publication-author level.

3 Results

3.1. Types of organizations frequently collaborated with

6763 humanities publications are co-authored by people exclusively affiliated with educational organizations. Apart from educational organizations (other universities and schools of higher education), the publications from the humanities are most frequently co-authored by people affiliated with the non-profit organizations, followed by the government (Figure 1).

Figure 1: Overview of organization types most frequently collaborated with

A closer look at the data shows, however, that it is important to see what organizations are behind these broad types. Some of the non-educational organizations are strongly research-focused and therefore have strong connections with universities. The organization type ‘nonprofit’ includes not for profit organizations that are research institutes. The government category includes organisations from different levels of government, e.g. the European level, the federal level, the regional (Flemish) level, province and city levels. Facilities are mainly research facilities, which means that they also will have strong links with universities and schools of higher education. Meanwhile, for the category of healthcare institutions, university hospitals lead the charts. Universities and university hospitals are often very closely linked, with people listing their affiliations as both in the hospital and the associated university. The list of companies seems more easily identifiable as non-academic. Categories of companies include law firms, architectural firms and archaeological companies. The list of archives includes museums as well as archives (to avoid confusion, we will refer to these organizations as “archives and museums”). On the whole, it seems that research-focused organizations appear in almost every category of organization. A manual effort will be required to distinguish between the government organizations on different levels, to distinguish museums from archives, and research institutes listed among government and non-profit organizations. The results of this effort will be presented at the conference.

3.2 The location of organizations

More than half of the non-educational organizations collaborated with are Belgian organizations (55.9% of organizations). Other non-educational organizations listed on affiliations are based in France and Germany (incidentally Belgium’s neighbouring countries) followed by the US and the UK.

Table 1: Breakdown of non-educational organizations collaborated with by country

3.3 Disciplines collaborating with non-academic organizations

The disciplines most frequently collaborating with non-academic organizations are the generic categories of ‘Humanities (general)’ and ‘Other disciplines’, followed by ‘Linguistics’, ‘Art History’ and ‘History’ (Figure 2). The discipline classification scheme used allows for publications being associated with multiple disciplines. This means that it is possible for publications to be represented more than once in the bar chart below. It also means that non-humanities disciplines are represented when a publication was co-authored by someone affiliated with a non-humanities discipline (e.g. Social Health Sciences and Criminology).

Figure 2: Disciplines most frequently collaborating with non-academic organizations

An overview of the types of organizations collaborated with per discipline, shows that there are distinct collaboration patterns per discipline. An example is the discipline philosophy (see Figure 3), for which the largest proportion of the collaborations are collaborations with the government, followed by healthcare institutions. This can be due to publications focussing on ethics or bio-ethics co-authored with people affiliated with hospitals. The reverse can be seen in political Sciences, which includes very little collaboration with healthcare institutions, but many with the government and non-profit organizations. History and archaeology list comparatively many collaborations with archives (which, as mentioned before, include both museums and archives). For archaeology (see Figure 3), there are also many publications written in collaboration with companies and the government. These are often publications co-authored with people working for archaeological services of the government or people working for archaeological companies performing archaeological services required for developments. The authors from law who work for companies often include authors who work as professional lawyers beside their academic activities. The specifics of disciplines create different collaboration patterns with non-academic institutions.

Figure 3: Overview of organization types collaborated with for two disciplines

3.4 Characteristics of publications co-authored with non-academics

In terms of language, the top language in humanities publications overall is English. For publications written in collaboration with people affiliated with a non-educational organization, English is still the most commonly used language, although the percentage of Dutch-language publications is somewhat higher. The overall difference in terms of language distribution does not seem to be very big, with similar levels of English publications.

Table 2: Overview of languages of publications

A starker difference can be seen in the publication types. For humanities publications in general, 61,20 percent of the publications are articles, while 26,35 percent are book chapters. For publications co-authored with non-educational organizations, the numbers are 41,63 percent articles, 33,89 percent book chapters. It seems that in publications co-authored with people affiliated with non-educational organizations, a smaller proportion is journal articles compared with the total VABB dataset.

Table 3: Overview of publication types

3.5. Double affiliations vs. non-academics

One particular type of collaboration with non-academic organisations found in publications is the double affiliation. A double affiliation occurs when an author is affiliated with more than one organization at the same time. For all authors in all of the publications coded for affiliations, 29,29% are affiliated with more than one organization. Considering the publications 574 publications out of 1146 publications that were coded on the publication-author level include authors that are affiliated both with one of the five Flemish universities and a non-educational organization. This is 50,9% of the publications in the dataset.

The converse of the double affiliation is the non-academic. These are authors who are not affiliated with an academic organization. In this case, we look at organizations that are not of the type “Education”. For all authors in all of the publications, 26,99% are not affiliated with an educational organization and could thus be regarded as non-academics. Out of 1146 publications 619 are publications with at least one non-academic. This is 53,87% of publications. These are cases where there is a collaboration with people from outside the university and thus fits more neatly with the objective of this study, which is to look at the interaction between the university, the academic world, and the non-academic world. The highest proportion of non-academics are affiliated with companies, followed by healthcare organizations and the government.

4. Discussion

The data about collaborations with non-academic organisations show that scholars from the humanities collaborate with people from government institutions, non-profits, companies, hospitals, facilities, archives and museums. The collaborations discussed here are cases of co-authorship, which is a highly formalized form of collaboration. It can be assumed that co-authored publications are only the tip of the iceberg of all interactions with non-academic partners. Other types of interaction include informal contracts, educational activities (workshops and trainings) and contract research (more usual in applied technical fields). In a study into the role of informal collaborations between the SSH and non-academic partners, Olmos-Peñuela (2013, p. 4) points out that informal collaborations “tend to be highly interactive rather than the one-way flow of technology from academia to industry inherent in technology commercialization”. Co-authorship, even though it is also a formal way of collaborating, indicates a bidirectional flow of knowledge as well.

When considering collaborations with non-academics, it is important to distinguish between collaborations between people working for different organizations and collaborations where people are affiliated with more than one organization (double affiliations vs. non-academics). In the case of double affiliations, the knowledge flow happens almost automatically, whereas in the case of collaborations with non-academics an extra effort is required to establish a collaboration. A study on the effects of dual appointments on collaborations with private and public partners (Cattaneo et al., 2019) has shown that the effects of dual appointments on collaboration are not straightforward, whereby academics with dual appointments may face some barriers to collaboration, such as precarious contracts and competing demands from their dual employers. This is an important aspect to bear in mind when considering the potential of dual appointments for enhancing knowledge flows.

In some humanities disciplines, the connection between research, and the profession related to the discipline is still strong. This is true for law, architecture, archaeology and history for example. The professionals are in these cases not working in industry, but law practitioners working for a law firm or independently, architects with an architectural practice, archaeologists affiliated with an archaeology company or public service, historians working in archives or museums. It may be helpful in this context to consider the concept of “knowledge communities”, defined by David and Foray (2002) as a “community where a large proportion of members is involved in the production and reproduction of knowledge”, and whereby “most knowledge communities cut across the boundaries of conventional organizations” (David & Foray, 2002, pp. 14–15). Academics from different humanities disciplines are part of knowledge communities that do not necessarily end at the university gates. For example, lawyers share their law insights in dedicated publications, archaeologists frequently write professional reports (Börjesson, 2015) etc.

The interactions between the university and the non-academic world can also be seen in light of the triple (Etzkowitz & Leydesdorff, 1995) and quadruple helix models of innovation. In the triple helix model, the interaction between science, industry and the government is theorized, the quadruple helix models adds the public (civil society) to the mix. The collaborations between academics from the humanities and non-university organizations show co-production of knowledge between different players in the quadruple helix model, as both companies, the government, civil society organizations and academics are involved.

The collaboration between the university and non-university actors may show the ways in which the university is firmly rooted in society. As part of the university’s third mission, the university takes up the additional mission of engagement with society on top of its original missions of teaching and research. However, as Jongbloed et al. (2008) point out, there is considerable overlap between the university’s third mission and teaching and research. University-industry relationships may lead to joint research, and lifelong learning activities really are a part of ‘teaching’. In a similar vein, co-authored publications with non-academic organisations are a part of the university’s research output, but the collaboration with non-academics may also contribute to the university’s third mission, or societal engagement.

5. Conclusion

In this paper, we have shown the results of an exploratory data analysis of affiliation information from the Flemish humanities. As this is a research-in-progress, there are a few avenues open to explore and include in subsequent publications.

A first concern is to include a more suitable typology of organizations than the one currently used (which is based on ROR). This will allow for a more fine-grained distinction between organizations. Secondly, an expansion of the study to the social sciences will offer additional opportunities for comparison and show how collaboration patterns differ between the various disciplines of the social sciences and humanities. Based on the preliminary results presented in this paper, we expect disciplines to have unique patterns of collaborations with non-academic organizations, which can be linked to the way in which the disciplines and professions related to the disciplines are rooted in society. A third concern is author disambiguation. The VABB-SHW data is available on the author-publication level. Whereas scholars affiliated with Flemish universities have identifiers within the database, this is not the case for external authors. Author disambiguation is therefore required to trace collaboration patterns over the years. This would allow us to gauge to what extent collaborations are longer lasting, with joint publications written by the same people over a number of years. With information about career trajectories, we could get a better sense of people moving in and out of academia. The career paths of former PhD students is a major concern for universities as academic positions on the postdoctoral level are limited. People with a background in academia may end up working for non-academic organizations, but still engaged in scholarship and contributing to scholarly knowledge from their new positions. Another group that could benefit from further attention, is the independent scholar. These are researchers not affiliated with any organization, academic or otherwise. Orlans (2002) shows how independent scholars are understudied, and face various hurdles when conducting research (difficulty of attaining grants, access to library collections, databases and interlibrary loans…). The VABB-SHW does not hold career information, but CRIS systems in other countries may provide this information. Finally, extending the analysis to other countries might also be useful to see if the results are similar across countries.

Open science practices

The VABB-SHW database is available online: Aspeslagh, Peter, Guns, Raf, & Engels, Tim C. E. (2022). VABB-SHW: Dataset of Flemish Academic Bibliography for the Social Sciences and Humanities (edition 12) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.7573408. An anonymised version of the affiliation data can be found at https://github.com/Eline-Vandewalle/collaborations_non-academics and will be updated.

Acknowledgements

We would like to thank in particular Peter Aspeslagh, who has worked on the addition of affiliation data to the VABB-SHW database for the last two years.

Competing interests

We declare that we do not have any competing interests.

Funding information

ECOOM is funded by the Flemish Government.

References

Abramo, G., D’Angelo, C. A., & Di Costa, F. (2011). University-industry research collaboration: A model to assess university capability. Higher Education, 62(2), 163–181. https://doi.org/10.1007/s10734-010-9372-0

Aspeslagh, P., Engels, T., & Guns, R. (2021). The road towards structured affiliation information in a national bibliographic database. https://doi.org/10.21428/7a45813f.51be98be

Aspeslagh, P., Guns, R., & Engels, T. C. E. (2022). VABB-SHW: Dataset of Flemish Academic Bibliography for the Social Sciences and Humanities (edition 12) [Data set]. Zenodo. https://doi.org/10.5281/ZENODO.7573408

Börjesson, L. (2015). Grey literature – grey sources? Nuancing the view on professional documentation: The case of Swedish archaeology. Journal of Documentation, 71(6), 1158–1182. https://doi.org/10.1108/JD-09-2014-0137

Čada, K., & Ptáčková, K. (2013). Possibilities and limits of collaboration between science and NGOs in the Czech Republic. Journal of Cleaner Production, 49, 25–34. https://doi.org/10.1016/j.jclepro.2012.07.039

Cattaneo, M., Horta, H., & Meoli, M. (2019). Dual appointments and research collaborations outside academia: Evidence from the European academic population. Studies in Higher Education, 44(11), 2066–2080. Scopus. https://doi.org/10.1080/03075079.2018.1492534

David, P. A., & Foray, D. (2002). An introduction to the economy of the knowledge society. International Social Science Journal, 54(171), 9–23. https://doi.org/10.1111/1468-2451.00355

Engels, T. C. E., Ossenblok, T. L. B., & Spruyt, E. H. J. (2012). Changing publication patterns in the Social Sciences and Humanities, 2000–2009. Scientometrics, 93(2), Article 2. https://doi.org/10.1007/s11192-012-0680-2

Etzkowitz, H., & Leydesdorff, L. (1995). The Triple Helix -- University-Industry-Government Relations: A Laboratory for Knowledge Based Economic Development (SSRN Scholarly Paper No. 2480085). https://papers.ssrn.com/abstract=2480085

Jongbloed, B., Enders, J., & Salerno, C. (2008). Higher education and its communities: Interconnections, interdependencies and a research agenda. Higher Education, 56(3), 303–324. https://doi.org/10.1007/s10734-008-9128-2

Kotiranta, A., Tahvanainen, A., Kovalainen, A., & Poutanen, S. (2020). Forms and varieties of research and industry collaboration across disciplines. Heliyon, 6(3), e03404. https://doi.org/10.1016/j.heliyon.2020.e03404

Meyer-Krahmer, F., & Schmoch, U. (1998). Science-based technologies: University–industry interactions in four fields. Research Policy, 27(8), 835–851. https://doi.org/10.1016/S0048-7333(98)00094-8

Olmos Peñuela, J., Molas Gallart, J., & Castro Martínez, E. (2013). The role of informal collaborations in the social sciences and humanities. https://digital.csic.es/handle/10261/107943

Orlans, H. (2002). Independent scholars: A neglected breed. Society, 40(1), 12. https://doi.org/10.1007/BF02802965

van Leeuwen, T. N. (2013). Bibliometric research evaluations, Web of Science and the Social Sciences and Humanities: A problematic relationship? Bibliometrie - Praxis Und Forschung, 2, 1–18.