Introduction

Over the last 50 years, globalisation and internationalisation in trade, communication, education and migration have highlighted the concept of population diversity. Peter Hershock (2012: 1) defines diversity as the ‘coexistence of many different kinds of things’ and the current emphasis values differences on the path to equity. While equality focuses on reducing discrimination, it implies the elimination of difference without necessarily recognising that individuals have different starting points and ongoing needs, which ignores the dynamic and distinctive nature of peoples and cultures (Hershock, 2012). By contrast, equity acknowledges and addresses cultural and demographic differences and enhances diversity. The shift in terminology from equality to equity underpins the current ethos of diversity in higher education: policies and frameworks, teaching and research, study options, workforces and student bodies. Proposing a framework for exploring interdisciplinary perspectives of diversity in science and technology policy, Andy Stirling notes that ‘diversity offers a means to promote innovation, hedge ignorance, mitigate lock-in and accommodate pluralism’ (2007: 715). This statement aligns with our project’s approach to diversity in open knowledge institutions: it aims to encourage openness in research through collaboration, diverse areas of knowledge, ideas, languages, cultures and workforces. Diversity is a group characteristic, but the multiple intersecting identities of individuals contribute to workforce diversity (Gibbs, 2014).

The Curtin Open Knowledge Initiative (n.d.) project produces a data warehouse of published research literature and its open access status. It includes analysis of institutional, national and international research output types, citations, social media and collaborations. Analysis is grouped by discipline, topic, publisher, funder, and research funding bodies. In addition, we collect and analyse workforce demographic data in order to explore correlations between institutional diversity and knowledge production. Understanding the transformation of universities into open knowledge institutions involves asking questions about who produces knowledge and how diverse is the knowledge disseminated through research and transmitted in teaching.

In this article, we explore origin (ethnicity, race, nationality, indigeneity), gender/sex and disability data in higher education institutional workforces across 24 countries. In discussing these diversity dimensions, we acknowledge the difficulty of using concepts that are culturally and biologically bound (Stolcke, 2000), and whose meanings differ globally: Western meanings of the dimensions discussed may overlay and differ from concepts held within other societies and epistemologies. Diversity and equality, in other words, are not universal concepts that can be ‘interpreted uniformly across cultures and countries’ (Klarsfeld et al., 2019). Medrano Valdez (2019) explains that in Andean societies, for example, ‘men and women are complementary opposites and the hierarchical status between them is functional’, not exclusive. In Yorubaland (Nigeria), however, the category of ‘woman’ did not exist before Western influences (Oyěwùmí, 1997).

This article opens a discussion on the variability of higher education workforce diversity data and the limitations of global comparisons. It begins by exploring the background of equality, equity and diversity in higher education. A discussion of our project follows, including the context of this study, the research methodology, results across nations, and finally its challenges and limitations. We then discuss the use of diversity data within our project.

Background

From the mid-20th century, challenges to equality in admissions and employment became part of the higher education landscape, with approximately 25 percent of countries globally introducing affirmative action programmes – many in the 5 years since 1990 (Jenkins and Moses, 2017). Affirmative action and equal opportunity policies in relation to workforce and student bodies focus on age, race, gender, socio-economic status, ethnicity, language, sexual orientation, disability, religion and caste. Some actions were in response to the legacies of imposed historical practices and assumptions of colonialism such as ‘patriarchy, racism, heterosexism, capitalism, and ableism’ (Senier and Miranda-Galarza, 2016: 393).

Diversity for Whom?

Many equity, diversity and inclusion (EDI) policies and legislative processes address diversifying student enrolment and student cohorts. This focus on student diversity, driven by international marketing and world university rankings, limits attention to the heterogeneity of faculty, academic and administrative staff (Buenestado-Fernández et al., 2019). Systemic cultural barriers and recruitment practices can limit student progression to academic workforces; the existence of international staff and students does not always indicate a diverse campus or educational outcome (Spencer-Oatey and Dauber, 2019).

Sara Ahmed (2012) reflects on the replacement of social justice-oriented equal opportunity and affirmative action policies in higher education, with policies of equity, diversity and inclusion seen as less threatening to reputations than earlier terminology that implied equality deficits. Diversity is ‘a conveniently empty word which can conjure up the image of a vibrant international, cultural and ethnic mix while concealing race- and ethnicity-based inequalities’ (Bowl, 2018: 685). University policies often act as marketing tools to promote diversity but without necessarily achieving diverse outcomes, and this economic focus can inhibit the resourcing and action of gender and racial equity policies and outcomes (Blackmore, Sánchez-Moreno and Sawers, 2015; Da Silva, 2016; David, 2015). In other words, the existence of an equality policy can become a proxy for achieving equality, while achieving diversity in institutions can remain elusive (Ahmed, 2012).

The powerful influence of world university rankings leads some universities to employ international academic staff without acknowledging different languages, cultural attitudes and practices, customs and academic traditions. For example, in South Korea, western academics hired to teach in English and build research activities were not aware of and did not fulfill the university’s expectations of collaborative research that highlights Korean contributions and learning Korean (Shin and Gress, 2018). Analysis by Buenestado-Fernández et al. (2019) of indicators in 127 higher education institutions in the Academic Ranking of World Universities (ARWU) found limited progress in reaching workforce diversity. Hundle (2019) notes that ‘the study of diversity requires serious, thoughtful investigations of global and transnational histories, theories, and practices of decolonization, as well as critical interrogations of key concepts of citizenship, community, and majoritarian and minority difference’ (316).

The Curtin Open Knowledge Initiative

Universities exist to support the creation and transfer of knowledge, and efforts by universities towards openness broaden the impact of higher education and research institutions. The Curtin Open Knowledge Initiative project grew from a critique of the limitations of the major commercial world university rankings. We identify and analyse institutional research performance and focus on bringing together open access publishing and open data in open knowledge institutions, building on a commitment to an open society and to open science. Achieving institutional openness involves the coordination and communication of policies and programmes and addressing challenges of diversity in research. This means engaging with diversity in knowledge and research production, and how knowledge is shared within disciplines, scholarly communities, across disciplinary boundaries and between universities and wider communities (Montgomery et al., 2020).

Diversity in Knowledge Production

A diversified workforce contributes to openness through the inclusion of diverse ideas, knowledges and languages in research and teaching. Underpinning this is diversity in leadership, administration and decision-making (Montgomery et al., 2021). Researchers representing diverse sexes/genders, ages, races and minority voices bring perspectives, knowledges and experiences that enrich and expand research. The New England Journal of Medicine recently acknowledged the need for more representative researchers and study populations, requiring the diversity details of participants to be provided for all articles (The Editors, 2021: 1431). The demographics of those producing research shapes knowledge, for example, through the gendered ways in which researchers are perceived in terms of their power and importance (King and Frederickson, 2021). We contend that core to achieving openness in knowledge creation, sharing, research and teaching is the diversity of institutional staff at all levels: academic, executive, professional, administrative and technical.

Ethnic and gender diversity in higher education research workforces produces a strong performance in research output and publication based on citations (Coalition of Urban Serving Universities et al., 2016); however, this varies by discipline. Analysis of research authorship diversity in higher education journals finds a predominance of English-speaking countries and affiliations with a limited representation of countries and institutions from the Global South (Williams et al., 2018; Fitzgerald and Jiang, 2019). Gender-based research output analysis shows global variation with overall higher productivity and quality from men, with only Brazil, Iceland and Kenya achieving gender balance (Opesade et al., 2017). Longitudinal productivity analysis of 1,523,002 Science, Technology, Engineering and Mathematics (STEM) scientists’ publication output and citations between 1955 and 2010 by Huang et al. (2020: 4615) identifies ‘gender invariants’: similar annual productivity and citation numbers for the same amount of work, but ‘gender-specific dropout rates’ affecting outcomes in the length of publishing careers and productivity.

To understand the progress of open knowledge institutions, the Curtin Open Knowledge Initiative project gathers and analyses publicly available workforce demographic data to build awareness of institutional diversity in research and knowledge production. We explore this data in parallel with analysis of research output, open and closed access publications, citations, institutional collaboration, publication sources, publishers, funders and social media events.

Data Characteristics

Discussion

The Open Knowledge Initiative project analyses demographic diversity data to highlight interactions with research performance such as publication, open access output and collaboration. For example, we include gender and origin data analysis where available within our interactive dashboards, providing an opportunity for managers and executives to review their institution’s diversity outcomes. To explore the intersections between diversity and research performance, we develop correlations between the percentages of women in academic and non-academic positions, open access publications, mentions of research in non-academic sources such as social media and institutional revenue or income data in universities. This raises questions about the reasons for different performances among national institutions. For example, wealthier, well-established and prestigious metropolitan institutions with high-ranking research profiles often present lower gender parity than less prestigious, regional universities with lower funding levels (Montgomery et al., 2021, Wilson et al., 2022 in publication).

The variations in workforce diversity data in this article reveal different geopolitical, economic, social and population movements, labour markets, traditions and priorities. Imperatives to collect data about workforces, indigenous populations, ethnic groups, gender and disability depend on the approaches of governments and educational authorities toward research, curriculum, funding and planning. These are counter-balanced by active institutional support for diversity and the individual’s decision to declare a disability and to self-identify a gender, ethnicity, race or indigeneity. Further investigation into the presence and absence of public demographic workforce data reveals the differing national priorities and outcomes underpinning the statistics: regional patterns where differences suggest underlying conflicts with systemic colonisation, patriarchy and marginalisation.

The absence of origin data for most African countries reflects complexities in the application and meaning of the concept of diversity. Ghana has over 80 ethnic groups that were forced together during colonisation, and acknowledging and prioritising such population diversity in higher education institutions is a challenge (Marcellus and Christian, 2012). Mamdani (2020) argues that ethnic differences in Rwanda are a construction of ‘colonial scholarship’ (42). At Makerere University, Uganda, Hundle (2019) notes that ‘problems of minority difference (whether racial or ethnic) are framed in both analytical and quotidian ways that emerge from the local and historical context: ethnic affiliation or ethnic politics; tribalism; or antiforeigner sentiment, anti-Asian sentiment, and even Aminism’ (311). Kenya, a ‘multi-ethnic society’ (Kisaka, Jansen and Hofman, 2019: 35), has more than 42 major and minor ethnic groups that are not represented equally in public university and college workforces, despite legal requirements for diversity. South Africa introduced the Employment Equity Act (EEA) in 1998 to redress inequities resulting from apartheid and to advance the employment of ‘designated groups, namely black people, women and people with disabilities’ (Kola and Pretorius, 2014: 1316). The Act requires higher educational institutions to develop employment equity plans. However, barriers to equitable access and participation in higher education persist due to poverty, social inequality and organisational inefficiencies (Pitsoe and Letseka, 2018). Black women, in two of the designated groups, continue to face ongoing gender and racial stereotyping, and encounter challenges with the publication of research and promotion in employment. Despite national mandates, not all individual institutions have equity policies with diversity strategies targeting black women (Ramohai, 2019). The categories ‘African’, ‘Coloured’, ‘Indian’ and ‘White’ (Table 1) reflect the separate universities for these groups established during apartheid (Mampane, 2019). The large body of migrant workers from across the continent is merged into the undifferentiated category: ‘African’. The struggle between ‘neoliberalism and managerialism [and] transformation’ after 1996 demonstrates the dangers of ‘decolonial discourse’ becoming rhetoric or metaphor because of systemic and structural blockages in the academy (Keet, Sattarzadeh and Munene, 2017: 4).

Asian countries in our selection provide gender data but none for race, ethnicity or nationality. India was an early adopter, introducing reservation policies, or ‘caste quotas’, aimed to address inequalities in education and employment following independence from the United Kingdom in 1947 (Jenkins, 2003; Robles and Krishna, 2012; Subramanian, 2019: 275). However, data collection has become complex, and colonial legacies such as the use of colonial caste labels persist (Jenkins, 2003). Diversity in Chinese higher education is multifactorial. Some suggest that the expansion and internationalisation of universities along Western and American lines since the 1990s, and the pursuit of higher rankings and educational status, has led to a lack of value placed on diversity and multiculturalism, as well as the displacement of indigenous Chinese knowledge and language diversity (Liu, 2016; Marginson, 2018; Wang and Leong, 2019).

Diversity, equity and inclusion are strategically important to European higher education institutions where cultural, legislative and institutional contexts strategies vary (Nason and Sangiuliano, 2020). Gender equality and gender mainstreaming in academic research are priorities for European research (European Union, 2019). In 2018, the government of the Netherlands sponsored the Westerdijk Talent scheme to appoint 100 women professors in universities. French universities, required to remain neutral because of the country’s secular status, cannot enact inclusive policies or gather racial and religious statistics (Claeys-Kulik, Jørgensen and Stöber, 2019). However, France contributes national gender parity data to the European Union, with gender equality programmes in higher education and research institutions (Ministry of Higher Education, Research and Innovation, 2022; Nason and Sangiuliano, 2020). A large gender imbalance in academic workforces persists (Claeys-Kulik, Jørgensen and Stöber, 2019). In the UK, the Athena Scientific Women’s Academic Network (SWAN) Charter was established in 2005 to increase women’s participation in STEMM fields. In 2015, the Charter expanded to include all disciplines, and professional, support and transgender staff (Advance HE, n.d.).

Most countries in the European Union collect nationality data, tracking international academic workforce mobility patterns, changing levels of global migration and movements of displaced peoples into Europe post World War II (Supik and Spielhaus, 2019). The UK collects racial data institutionally, with specific nationality data only at the national level. The Race Relations Amendment Act (2000) and Equality Act (2010) led to the Race Equality Charter 2014, which aimed to advance racial equality for Black, Asian and Minority Ethnic students and staff in higher education institutions. Similar to the Athena SWAN Charter for gender, the two charter schemes compete with one another for resourcing and positioning—an example of the challenges for intersectionality when race and gender are regarded as mutually exclusive (Cho, Crenshaw and McCall, 2013). Ireland adopted Athena SWAN in 2015. It has a global education strategy that encourages inward and outward staff mobility, and the Irish Higher Education Authority has recently announced that it will collect staff race and ethnicity data for the first time (Aodha, 2020). However, the comments about ‘toxic identity politics’, quotas and merit appointments that followed Aodha’s article indicate resistance to such moves.

Many Latin American and Caribbean countries aim to achieve diversity in education but with varied implementation by country depending on historical, political and social factors (Buenestado-Fernández et al., 2019). Chile and Mexico provide gender data, but no public ethnicity data.

In the United States, the ground-breaking 1964 Civil Rights Act prohibits discrimination in education by race, colour, sex, religion or origin, voter registration and employment. It aimed to redress the impact of slavery but has since broadened from its original intentions. However, financial factors (‘university capitalism’) and the ‘return of epistemic and political conservatism’ threaten affirmative action processes and studies (Santos, 2019: 232). Following opposition and legal challenges to racial affirmative action programmes and legislation, many have ceased to exist (DiAngelo, 2018). For example, the US Supreme Court upheld a claim of discrimination through the ‘preferential treatment’ of students from minority groups inherent in higher education affirmative action programmes as a violation of the Fourteenth Amendment (Da Silva, 2016: 193). A focus on gender equality programmes includes the National Science Foundation’s (NSF) ADVANCE funding of programmes to increase the numbers of women in STEMM faculties (Yen et al., 2019).

The United States and Canadian population race/ethnicity classifications, corresponding to their statistical agencies, provide detail on colonised peoples within higher education. However, Gasman, Abiola and Travers (2015) note that there are small percentages of diverse population groups employed in senior positions, and a lack of diverse leadership in eight elite United States Ivy League institutions, despite better student diversity figures. American Indians have ‘little to no pipeline into graduate education and thus academic leadership’ (8). The Universities Canada (2015) Principles of Indigenous Education meet federal and provincial government legislation requirements but these strategies are ineffective and ‘non-performative’ in addressing racism (Dua and Bhanji, 2017: 238). Indigenous scholars identify ongoing colonialism and the need for systemic change in Canadian institutions (Cote-Meek and Moeke-Pickering, 2020).

Australian data reflect recent developments. Government guidelines require universities to implement indigenous workforce strategies to increase the number of Aboriginal and Torres Strait Islander staff to ‘3 percent of the total workforce’ (Universities Australia n.d.: 32)—a reflection of parity with the Australian population—with at least one Aboriginal or Torres Strait Islander person employed in a senior executive role. However, strategies amended and adapted by individual institutions meant that baseline targets were not being met. The second report highlights some growth in indigenous staff numbers (Universities Australia, 2020) but Australian universities need to acknowledge systemic inequities to bring about cultural change (O’Sullivan, 2020). Australia adopted the Athena SWAN programme (Science in Australia Gender Equity, 2018), yet disparities persist: gender equality is downplayed within EDI programmes and world university rankings are prioritised both in Australia and in Aotearoa New Zealand (Bönisch-Brednich and White, 2021).

Aotearoa New Zealand tertiary institutions provide ethnicity data for Māori and Pasifika populations. However, despite equity obligations specified in Te Tiriti o Waitangi and policy aspirations to achieve diversity among researchers and across disciplines, the number of Māori and Pasifika scientists in Aotearoa public institutions varies (McAllister et al., 2020). Pihama et al. (2019) identify the importance of non-mainstream epistemological and cultural knowledges in curricula, along with pedagogies, research and methodologies. Public workforce gender statistics, ongoing gender pay gaps and slower progression through academic employment grades for women than men indicate persistent gender disparities (Brower and James, 2020).

Concluding Remarks

This article opens a conversation on the global variability of higher education demographic data. Our exploration of workforce data characteristics from a selection of countries and regions reveals gaps, variations and limitations of national and institutional data collection practices and priorities. These findings support others regarding the inconsistency of global diversity data (Alesina et al., 2003; Klarsfeld et al., 2019, Patsiurko et al., 2012). In short, when data gathering is influenced by economic and political decisions made by governments, education authorities and institutions it affects the recognition and acknowledgement of educational and employment rights of minority and emerging groups. Decisions taken on whether to collect or to not collect data in certain demographic categories can exclude, ignore and silence those who do not fit or do not wish to be categorised, ultimately compounding ‘inequality and discrimination’ (Desivilya et al., 2017: 92).

We have found, in relation to gender data, an indication that there is global progress towards gender equity in many higher education workforces – but gender cannot be fully accounted for within binaries. Data collected in a few countries recognise diverse gender identities but these small numbers are not reflected in published data, limiting awareness of gender diversity. Analysis of origin diversity highlights varying attitudes and practices across the world. We find inadequate recognition of different mental and physical abilities within workforces in most countries. Overall, diversity data fails to acknowledge intersectionality within workforces: the ‘multiple “axes” of power and difference that shape individuals’ positionalities’ (Rice, Harrison and Friedman, 2019: 409). Our analysis highlights the need to recognise how power within knowledge production is differentiated by country, and to see the importance of considering context, binaries and hierarchies in the use of data (D’Ignazio and Klein, 2020).

Our interest is to understand workforce diversity as a dimension of research production and knowledge openness within countries, regions and institutions. Together with longitudinal analysis of research output, this contributes to our project’s aims: to build comprehensive data analysis and tools for understanding the role of universities in knowledge-making and sharing; to encourage diversity and openness in research and teaching; and, to move beyond a reliance on incomplete and inadequate metrics employed by the major world university rankings. Through analysis of publicly available demographic data, we bring greater clarity to diversity in higher education workforces and knowledge production as pillars of open knowledge institutions. We encourage universities to examine diversity data in order to understand their own workforces, to promote multiple perspectives in research, teaching and management, as well as opportunities for knowledge sharing and critical thinking within and among institutions and communities. Evaluating diversity data can lead to a critical understanding of the impacts of policy achievements and outcomes on teaching and research practices, and can, therefore, challenge inequalities.

Early research into the COVID-19 pandemic lockdowns and restrictions on academic work practices shows a negative impact on research productivity outcomes for women and early career researchers (Australian Academy of Science, 2020; Esser et al., 2020; King and Frederickson, 2021; Woolston, 2020). Further analysis is needed to highlight global disparities and to reinforce the need for awareness to address higher education workforce diversity and equity. We hope that, by drawing attention to workforce demographics alongside research output analysis, we can encourage institutions to examine and understand the impacts of their practices and outcomes, to extend and maintain diversity in research and teaching.

Competing Interests

The authors have no competing interests to declare.

Data Availability

All data are publicly available. Sources are shown in Figure 1.