Exploring the status of diversity in policies and practices of Spanish universities. An asymmetric dual model.

Higher education across the world is currently in the throes of assuming a commitment to diversity. However, certain critical positions maintain that such evolution is still guided by market principles. Within such a context, this paper explores what attention is given to diversity in Spanish university policies and practices and how it relates to key productivity indicators. To do so, a study with a descriptive and correlational design was conducted, based on analysing institutional documents and surveying chief diversity officers, techniques which provided evidence about diversity philosophy and practices, respectively. The results revealed at least an average level of institutionalisation of diversity, although it did not demonstrate, in most of the areas, any association with indicators derived from a consolidated ranking by productivity in Spain. The conclusion is that Spanish universities have adopted an asymmetric dual model, in which neo-liberal ideas maintain their hegemony while, although subordinately, certain innovations have been consolidated in parallel in order to protect a number of vulnerable groups under the rhetoric of equity and social justice.

Introduction

In today's knowledge society, the role played by higher institutions in regard to the diversity of persons, groups and identities has been questioned and on occasions, greater emphasis is called for when attending to diversity at institutional level. In Spain, the status of diversity has been fostered by various nationwide regulations, but also by the academic community, where some firm steps have been taken in institutional policies by a number of universities (Márquez-Lepe et al., 2018). Nevertheless, a comprehensive and reliable diagnosis across the entire population of Spanish universities has still not been undertaken, nor have these developments been interpreted within the global paradigmatic framework in which such institutions operate. In this sense, the research reported on here seeks to identify the level of institutionalisation of diversity by Spanish universities (both public and private), taking into account a list of 24 indicators validated previously in an exploratory study of an international sample group (Buenestado-Fernández et al., 2019). These indicators were useful to design a checklist to obtain evidence about institutional philosophy and policy, as well as a questionnaire to collect data about actual practices in the field of diversity. Likewise, the level of institutionalisation indicated by data on diversity obtained with both instruments was associated with positions that institutions occupy in a national ranking based on meritocratic indicators of teaching, research, and innovation and technological development (Pérez-García et al., 2018), which enabled a certain level of interpretation on how diversity fits into universities' operating framework, dominated by market values.

Higher education, diversity and institutional response

The role of higher education in the knowledge society has evolved over recent decades. Both international organisations and academic players emphasise the status of the university as an agent of sustainability in a diverse society for which it assumes its responsibility (Santos-Rego, 2016; UNESCO, 2010). Accordingly, a growing interest is seen in this attribution in scientific literature that looks at the reforms undertaken by universities (Larrán-Jorge and Andrades-Peña, 2017). Within this context, an important task assumed by higher education as part of its social responsibility is training in global citizenship skills (Horey et al., 2018), but another very important subject is attending to individual and group diversity, particularly in reference to vulnerable groups, where universities design inclusive pedagogies (Stentiford & Koutsouris, 2020) and other strategies oriented towards the pursuit of inclusion (Baltaru, 2019). For example, they provide access to a greater number of citizens, but also qualify such access and promote personal, professional and social development among all groups of citizens (Smith, 2020). Research on this question has set its priority on sectors of students (Goastellec and Välimaa, 2019) but attention has also been paid to the rest of the university community, by assessing, for example, how teaching staff are selected, retained and promoted (Breetzke and Hedding, 2018; Gumpertz et al., 2017; Kaplan et al., 2018; Kelly et al., 2017; Moody 2012), or even how demographic composition impacts on certain positions of administrative responsibility (Bowman and Bastedo, 2018).

Furthermore, the dimensions, areas or institutional indicators that would be affected by inclusive attention to diversity in the university community as a whole vary greatly. The literature on this issue reveals that a number of elements need to be considered in a process of institutionalisation (Ahmed, 2012; Association of American Colleges and Universities, 2015; Buenestado-Fernández et al., 2019; Department of Education, 2016; Ferreira et al., 2014; Gause et al., 2010; May and Bridger, 2010; New England Resource Center for Higher Education, 2017): 1) institutional philosophy and policy towards diversity; 2) opportunity of entry for students from all protected groups, as well as the strategies for their participation and progress; 3) clear leadership in favour of inclusion and equity; 4) how support services operate for students from protected groups; 5) processes to foster evaluation, research and innovation in inclusion and equity; 6) an inclusive academic curriculum; 7) a climate and culture of inclusion that permeates the interaction and social and academic participation of all students and staff at the institution; and 8) projection of this inclusive culture within the community. Additions to such wide-ranging areas would include more specific ones such as staff—especially teaching staff—training, both in regard to the concept of diversity and, above all, in terms of teaching methodologies and techniques (Hitch et al., 2015; Llorent et al., 2020; Moriña, 2017; Moriña et al., 2015, 2020). In short, the agenda on diversity is very far-reaching and cannot be restricted to a very limited set of areas and groups (Chang, 2002).

Within the institutional response the international higher education community gives to diversity, historical, political and cultural factors still set differences in terms of prioritising groups. Thus, for example, gender is a cross-cutting issue in the western world, where notable progress in the institutionalisation of diversity has been made in some areas but not without considerable controversy (Leathwood and Read, 2009). Furthermore, in the USA, race and ethnicity, as well as socio-economic status, are the criteria for diversity (or protection) that are the biggest concern in regard to equal opportunities of entry and learning results in higher education, not forgetting the attention given to functional diversity (Lombardi et al., 2018) and other characteristics (Mukherji et al., 2017), and promoting the concept of inclusive excellence in all cases (Association of American Colleges and Universities, 2015; Department of Education, 2016). This kind of discourse in favour of diversity is rooted in the Civil Rights Movement of the 1960s and its prolegomena (one could even refer to the 2nd Morril Act of 1890, which prompted the creation of black colleges), where the Civil Rights Act of 1964 is the clearest precursor of affirmative action in the United States (Bell, 2012). The results of this movement have been positive and have resulted in indicators of various types (Okahana et al., 2018), but their pragmatic and market-based approach has also been criticised (Hode and Meisenbach, 2017; Thomas, 2018, 2020) and thus they have turned out to be somewhat limited compared to the scope of inclusive efforts (Foster and Fowles, 2018; Furquim and Glasener, 2017).

Affirmative action measures were also approved later on in other regions and countries, such as in South Africa with its transition to democracy in 1994 and the post-Apartheid era. These measures have had and are having positive effects on the demographic composition of higher education, as Kerr et al. (2017) analysed in the case of the University of Cape Town. However, graduation rates among beneficiaries of such affirmative actions are still not at the desired level and, although progress has been made, deep inequalities based on race, socio-economic status or gender currently still persist in the country (Pitsoe and Letseka, 2018) and extend to its universities in the form of epistemic violence or imposition of a Eurocentric vision (Heleta, 2016; Ndofirepi and Gwaravanda, 2019).

Across Latin America, ideological divisions have become more acute, with currents of thought in higher education that do not recognise the plurality of groups and individuals. By not doing so, they effectively exclude any minority group that does not adapt to the homogenizing cultural norms in power. The movement for indigenous native rights strives to denounce this position and has had certain influence in the realm of higher education (Lehmann, 2013), similar to what has happened in Canada since the 1970s (Pidgeon, 2016), where they have attempted to incorporate indigenous culture and knowledge into institutions or create structures parallel to the urban-centred Western model of higher education, like, for example, the intercultural universities in Mexico (Lehmann, 2013; Mateos-Cortés and Dietz, 2015). The application of this type of inclusive measures in the American continent entails certain obstacles that indigenous students themselves have reported (Hanne, 2017), and evaluations of their implementation and impact are still awaited in order to measure the extent of progress in regulations (Fajardo-Becerra, 2017). Rules in favour of indigenous, but also black students and candidates from poor backgrounds have also been applied in countries like Brazil (Balbachevsky et al., 2019), although results have become more visible in quantitative rather than qualitative terms—minorities participate in programs with lower social status—.

In Europe, along with gender (Klein, 2016; Rosa et al., 2021) and income—this latter criterion served through traditional financial support policies that include scholarships, loans or other benefits (European Commission/EACEA/Eurydice, 2018)—disability or functional diversity is the criterion of plurality that currently receives the most attention in the organisation of diversity services and programs in higher education (Biewer et al., 2015; Paz-Maldonado, 2020). The effectiveness of inclusive efforts, however, might be quite limited in some countries (Aust, 2018; Gibson, 2015) and, in any case, there is still a broad scope of action to be taken (Barkas et al., 2020).

In general, our overview of progress in different geographical areas reveals greater attention being paid in higher education to disadvantaged social groups, although practice—particularly, effective action—might be lagging behind rules and organizational measures. Even so, cataloguing vulnerable groups has served to define diversity and channel special attention. In some countries, the concept of protected features or groups has been introduced, for whom certain adjustments are made to facilitate their members' full participation in the learning processes and prevent any type of discrimination or exclusion (Equality and Human Rights Commission, 2014; O'Donnell 2016).

Context of the research and theoretical framework

In the case of Spanish universities, current state regulations establish the principle of non-discrimination and equal opportunities (Royal Legislative Decree 1/2013 of 29 November 2013) and urge institutions to establish positive actions and services. Looking specifically at physical disability, which is the priority criterion for protection along with gender, such actions are mainly aimed at gaining accessibility to buildings and public areas, adapting curricula, building awareness of disability, and procedures for welcoming and guiding new students.

As far as areas of institutionalisation of diversity in Spanish higher education are concerned, encouragement has been given to creating central support services. However, this measure may be serving to provide protection to gender and functional diversity but not to other vulnerable groups, as Márquez-Lepe et al. (2018) pointed out in a study on diversity policies at 8 different universities. When it comes to gender, these authors found that universities have developed comprehensive plans that include actions in labour, research and teaching fields. In these areas, they have addressed problems related to conciliation and co-responsibility, vertical segregation in access to academic positions, prevention of sexual harassment, promotion of non-sexist use of language, and research with a gender perspective or led by women. For their part, disability action plans are not comprehensive, but consist of specific protocols aimed at supporting and guiding students with disabilities, fostering collaboration with companies and other socio-labour actors, and developing tailored teaching resources and protocols. The variability in the production of these policies in the eight universities studied is, however, considerable. Despite this preliminary evidence, no overall diagnosis exists of policies and practices to address diversity in Spanish higher education, nor is it known how they are articulated within the neoliberal-tainted hegemonic operating framework. Regarding these deficiencies, two objectives are intended to be achieved in the current study: 1) to diagnose the level of institutionalization and coordination of diversity policies and practices in Spanish universities; and 2) to establish the relationship between diversity and productivity indicators in order to determine the fit of diversity policies and practices within the market framework.

The theoretical grounding would be in the proximity of Paulo Freire's Critical Pedagogy school (Freire, 2000), extended to analyse the impact of banking education on social groups, not only on people. Within this framework, if the dominant educational policies are generated and evolve according to market logic and if there is little room for moral considerations, it is difficult to foresee significant developments of an inclusive approach in the education system, particularly at levels such as vocational training or higher education, in which teaching is aimed at providing employability skills so as to compete on the market. In opposition to this dominant trend, the Critical line defends critical consciousness as a learning process that seeks to empower people to overcome global and local alienation and transform the world (McLaren et al., 2007). In this way, through consciousness raising, people are progressively freer (Freire, 1974), such that education in itself can be seen as a liberating practice of a political nature (Freire, 1983; Giroux, 2011). Education ceases to be a means of liberation when it imposes a certain worldview—that of the dominant group—which renders that of minority groups invisible. In this hierarchical scenario, the guidance of a critical pedagogy that adapts adequately to new educational environments becomes a necessity, as do critically participatory academics/activists to be involved in anti-hegemony transformation processes (Apple, 2011, 2016).

From this critical approach, universities are conceived as transformative agents who have a responsibility to contribute to the social mobility, inclusion and social development (Paz-Sánchez et al., 2018). However, this normative position does not match neoliberal productivity goals and competition mechanisms that have been strengthened in higher education in the world during the last decades (Cerro-Santamaría, 2020). Although importance given by students and employers to global rankings depend on several factors (Souto-Otero and Enders, 2017), classification of universities could be a good example of the market trend. Indicators that are valued in rankings are far from inclusion or social development. They are based on factors such as institutional reputation, citation impact of published research results, or even faculty Nobel Prize and Fields Medal winners (Hou and Jacob, 2017). Universities that rank at the top of global classifications belong to a small number of white-majority countries, with the United States and the United Kingdom at the forefront, and with a loop being observed between the economic context and the results: the more favourable the economy of the environment, the higher the scientific output which, in turn, enhances a better position of universities in the rankings (Luque-Martínez, 2015; Luque-Martínez and Faraoni, 2020). Therefore, it could be contended that the market has a relevant influence on university policies and practices and, if certain innovations are not acknowledged in rankings—e.g., those about diversity, equity and social justice, institutions are less likely to commit to them. But if universities still want to adhere to a more inclusive profile, then they must urgently address the cognitive dissonance of promoting productivity rankings and, at the same time, committing to diversity and equity (Stack, 2020).

In short, this paper assumes that today's university has incorporated the concept of quality in terms of efficient productivity in a market-based context, thus rendering the value of individual and group diversity invisible. However, critical pressure from numerous social groups has led to an implementation of processes to institutionalise diversity but which are probably subsumed by the more general neoliberal current without adequately connecting the processes and achievements in diversity with others linked to productivity. These assumptions allow us to anticipate three hypotheses. Two of them are proposed under the first objective of the study:

The standard of institutionalisation of diversity in Spanish universities is not yet at an advanced stage in most of the areas; and, as it is a recent trend, limited progress would affect both public and private universities (i.e., empirical averages of institutionalisation of diversity policies and practices will remain below the midpoint of the response scales used, without significant differences by type of institution).

The coordination between policies and practices is still scarce (i.e., no significant associations will be found between diversity policies and most of the practice areas).

As for the second objective of the study, a third hypothesis is tested:

Diversity policies and practices are progressing independently of productivity goals (i.e., no significant associations will be identified between diversity and productivity indicators).

Method

Design

A descriptive and correlational design was used to evaluate the presence and characteristics of policies and practices to institutionalise diversity in Spanish universities (both public and private). A two-phase study was conducted. In the first stage, as a means of diagnosing institutional philosophy and policy, an ex post facto study was performed through documentary analysis and review of information accessible via university websites. Website content analysis has already been used in other research as a source of information about the commitment of higher education institutions to diversity, inclusion, equity and social justice (e.g., Ford et al., 2019; Ford and Patterson, 2019; Holland and Ford, 2020; LePeau et al., 2018; Wilson et al., 2012; Wilson and Meyer, 2009), so that this tradition was continued in our first stage. This data collection strategy is more objective than survey procedures when researchers focus on normative information, since limited cognitive ability of respondents to retrieve stored data is avoided, even if, on the other hand, content analysis includes an interpretative component when linking text units to codes (Krippendorff, 2013). However, a deductive content analysis approach was used in the study, reducing subjective influences on coding. Secondly, in order to explore institutional practices, data were collected via a questionnaire sent to chief diversity and equality officers at Spanish universities. Unlike philosophy and policy, most websites do not include diversity practices, except, in some of them, for the information distributed through annual diversity reports. However, since the study was aimed at carrying out a diagnosis of current diversity actions, the administrative managers who were in charge of translating institutional policies into practice could provide the most complete and recent information. In fact, chief diversity officers have a leadership role in the process of institutionalisation of diversity (Stanley et al., 2019) and, as such, they are key figures in the implementation of diversity practices in higher education (Leon, 2014; Worthington et al., 2020), and have already participated in other samples when studying diversity actions (e.g., Buenestado-Fernández et al., 2019; Griffin et al., 2012, 2019).

Sample groups

Initially, all the population units in both studies were included, which accounted for 84 Spanish universities (50 public and 34 private). It was only necessary to discard three private institutions due to the lack of information provided on their websites. Therefore, the final number of sample units in the first part of the research consisted of 81 universities (96.4% of the population).

In the second stage, we were able to contact via email the heads of diversity and equality services at 79 universities and invite them to take part in an online survey. 42 chief officers from 35 universities (44.3% of those invited, representing 41.7% of the population of Spanish universities)—28 public and 7 private—responded to the survey (at 7 institutions, replies were received from the heads of both services—diversity and equality). Two cases were ruled out following identification of systematic and extreme bias in their responses to all questions, thus reducing the number of universities considered in our critical analyses to 33 (27 public and 6 private).

Instruments

Two instruments were used—one non-interactive for the first stage (a check-list), in order to record the presence of diversity indicators in institutional regulations, and another interactive one in the second stage (a questionnaire), aimed at obtaining data on diversity actions.

Check-list for identifying diversity policies

For the purpose of the first stage of the research, five indicators corresponding to area 1 of institutionalisation—philosophy and policy—were included on the check-list (Annex 1) of 24 on the original list validated in a previous study by Buenestado-Fernández et al. (2019). These authors listed 24 institutionalisation criteria that they identified in the literature and grouped into four dimensions: “2 general (philosophy and institutional policy, and institutionalisation strategies aimed at the university community) and 2 specific (institutionalisation strategies aimed at teaching and research staff, and institutionalisation strategies aimed at administrative managers within the institution)” (Buenestado-Fernández et al., 2019, Tools and Procedure section, para. 3; see the complete list of indicators in https://doi.org/10.1371/journal.pone.0219525.s002). This was the original material of an e-Rubric subjected by these authors to a content validation procedure, after which the wording of several items was altered, although the number and essential meaning thereof was maintained. What was now done to build the check-list was to take five of the seven indicators from the philosophy and policy section of the original list (i.e., items on alignment with institutional statement, strategic planning, definition of diversity, institutional culture, and accreditation). The two remaining indicators of this section (i.e., items on the connection of actions to the community context, and collaboration with external entities) were included in the second instrument of the study—a questionnaire, along with the indicators in the remaining three sections, since it was judged that a more precise information could be collected about them from key informants. In order to detect the presence of the five philosophy and policy indicators in institutional provisions through content analysis (Flick et al., 2004), a total of 167 official documents (strategic plans or, failing that, bye-laws; diversity plans; and equality plans) were extracted from institutional websites. Using the qualitative analysis software Atlas.ti (v.7), we proceeded, after a preliminary scanning of documents, to the systematic reading of each of them, as well as to deductive coding or theoretical-guided coding consisting of assigning categories of phenomena (i.e., the five descriptive codes/indicators of philosophy and institutional policy on diversity) to segments of text (Miles et al., 2014; Miles and Huberman, 1994). In the check-list shown in Annex 1, each code label (in italics) is followed by its description, crucial to identify its related text units. In order to obtain an index of institutionalisation of diversity, the number of indicators identified in the regulatory documentation for each university were added up. The result, that ranged from 0 to 5, was considered as the score in the area of diversity philosophy and policy.

Questionnaire on diversity practices

This survey was designed by adapting the remaining 19 indicators from the original list cited above (Buenestado-Fernández et al., 2019), which were estimated to cover eight areas of institutionalisation of diversity practices encountered in the theoretical review (access, participation and achievement; leadership; support services; evaluation, research and innovation; curriculum; climate and culture; community projection of inclusive culture; and staff training. See numbering of areas in Table 1). Item 1 of the questionnaire corresponded to indicator 5 of the original list, while the other items (2–19) followed a continuous and correlative numbering to that of indicators 7–24 in the original list. The correspondence of items to practice areas of institutionalisation is as follows: items 6, 7, 8 (area 2); item 4, 13 (area 3); items 14, 15, 17 (area 4); items 11, 12, 18, 19 (area 5); item 9 (area 6); items 3, 5, 16 (area 7); items 1, 2 (area 8); item 10 (area 9). The questionnaire is shown in Annex 2. Google Forms was used to administer it online to chief diversity and equality officers at Spanish universities, who were asked, with prior informed consent, to express to what extent they agreed with the assertion included in each item, using a 5-point Likert response scale (1 = "Strongly disagree"; 5 = "Strongly agree"), except for two elements (13 and 15), where the response was dichotomous (Yes/No). Before the survey was sent out, it was approved by the University of Cordoba Bioethics and Biosafety Committee.

Table 1

Descriptive statistics and correlations between scores in the nine areas of institutionalisation of diversity.

Institutionalisation area	M	SD	2	3	4	5	6	7	8	9
1. Philosophy and policy	2.67	1.41	.130	.093	.203	.067	.563∗∗	.317	.285	.082
2. Access, participation and achievement	3.70	0.86	--	.770∗∗∗	.810∗∗∗	.708∗∗∗	.513∗∗	.774∗∗∗	.725∗∗∗	.446∗∗
3. Leadership	3.74	1.05		--	.700∗∗∗	.613∗∗∗	.392∗	.798∗∗∗	.616∗∗∗	.495∗∗
4. Support services	3.62	0.86			--	.661∗∗∗	.465∗∗	.750∗∗∗	.596∗∗∗	.528∗∗
5. Evaluation, research and innovation	3.04	0.80				--	.623∗∗∗	.549∗∗	.672∗∗∗	.548∗∗
6. Curriculum	3.59	1.04					--	.580∗∗∗	.698∗∗∗	.394∗
7. Climate and culture	3.97	0.70						--	.758∗∗∗	.475∗∗
8. Community projection	4.05	0.81							--	.447∗∗
9. Training	3.55	0.92								--

Note. The mean for the area of philosophy and policy is interpreted on a scale of 0–5, while the questionnaire reply scale used to measure institutionalisation in eight practice areas ranged from 1 to 5.

∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001.

As far as the seven universities where two central service managers replied, first we took the average of both replies for each item (their answers never differed by more than two points on the scale in any item). Then, in the 33 institutions with valid responses, the score indicating the level of institutionalisation in each area was determined from the average obtained in the relevant rating scale items. For its part, the score in regard to the level of institutionalisation in the set of practice areas was determined by the mean of the 17 replies in scale format. As indicated by the response scale, averages ranged from 1 to 5.

The reliability of the questionnaire containing 19 items (including two Yes/No options) was tested using Cronbach's alpha and revealed high internal consistency (α = .939).

Data analysis

As a previous step to hypothesis testing, relative frequencies for the five indicators in the area of philosophy and policy were computed, and basic descriptive statistics (means and standard deviations) were calculated for the general distribution of scores assigned to universities according to the number of indicators identified in their regulatory documentation, as well as for the same general distribution by type of university (public or private). On the basis of the data received from the chief diversity and equality officers, descriptive statistics for the 17 questionnaire items with a scaled reply were analysed, as were those that refer to each of the eight areas in which these items were theoretically grouped. In addition, frequencies were obtained for dichotomous items 13 and 15. Finally, descriptive statistics were computed for the generic variable (level of institutionalisation of diversity practices).

The first hypothesis was tested by using one-sample t-tests to confirm the limited level of institutionalization of diversity, and t-tests for independent samples to check for differences between public and private institutions. In the case of diversity practices, a complementary non-parametric test (Mann-Whitney U-statistic) was used because of the small size of the groups under comparison. The midpoint of the diversity policy and practice scales was taken as the theoretical reference (if below it, the level of institutionalization would be considered limited, while if above it would be judged as advanced).

As for the second hypothesis, bivariate correlations were conducted between the nine areas of institutionalisation of diversity, as well as between the five philosophy and policy indicators and the eight practice areas, and between the general index of institutionalisation of diversity philosophy and policy and that of diversity practices. Apart from relating general indexes of institutionalisation, both distributions were z-transformed (this was done for the 33 universities for which both types of data were available—the scale was different in the two sets). Differential standardised scores were computed and plotted in a bar chart to help explain correlational results.

Lastly, the third hypothesis of the study predicted that diversity policies and practices are progressing independently of productivity indicators. As a measure of productivity, a solid ranking of universities in Spain was selected—U-ranking (http://www.u-ranking.es/en/nosotros.php), designed by the BBVA Foundation and the Valencia Institute of Economic Research. It is constructed using indicators referring to three university missions (teaching, research, and technological innovation and development). Institutions are classified both overall and on each of these three large areas, with performance corrected to take into account the institutional size. The sample group in the 2018 edition of the ranking comprised 61 universities. The association of this ranking (overall index and partial indexes of teaching, research, and innovation) with the two general distributions of institutionalisation of diversity (philosophy and policy, and practices), and also with the different areas of institutionalisation of practices, was tested by using bivariate correlations.

All the analyses were performed using the IBM SPSS software package (version 25.0).

Results

Level of institutionalisation of diversity

Institutional policies

Our review of the presence of the five indicators of diversity philosophy and policy in institutional documents revealed that four of them are contemplated in a majority of universities' basic regulations, while one is only present in about a quarter of them. Specifically, while approximately two-thirds of institutions include diversity in their stated mission (66.7%) and have an active plan for diversity (69.1%), less than a quarter of them professionally accredit the activities they perform in this respect (24.7%). On the other hand, although the definition of diversity and the incorporation of diversity into the institution's organisation and operating mode are most often contemplated in the regulations (54.3% and 51.8%, respectively), that does not occur in almost half of the universities.

An average of 2.67 was obtained for institutionalisation of diversity policies on a scale of 0–5 (SD = 1.41), i.e. approximately half of the indicators are contemplated in mainstream universities, although the variability between them is broad.

One-sample t-test result showed that it was not possible to reject the null hypothesis, t (80) = 1.06, p = .292, hence the evidence pointed towards an average visibility of indicators in regulation documents (M = 2.67), that was not different from the theoretical average of the scale (M = 2.5). What it was possible to reveal was the scant level of institutionalisation in private universities (M = 1.81, SD = 1.38) compared to public settings (M = 3.20, SD = 1.16), t (79) = 4.89, p < .001, d = 1.10.

Institutional practices

In terms of the 17 indicators with a scaled reply, chief diversity and equality officers highlighted the collaboration with external entities (M = 4.30), the existence of sufficient support mechanisms that facilitate participation by members of the university community (M = 4.23), the operation of support and advisory mechanisms in diversity matters (M = 4.21), the extension of activities to the entire university community (M = 4.18), and activities to raise awareness within the university community regarding diversity (M = 4.12). On the other hand, limitations are mainly found in the indicators of evaluation, research, innovation, and resource endowment, where the average is around the midpoint of the scale (between 2.83 and 3.18).

In regard to scores in the areas of institutionalisation in which the 17 indicators were theoretically grouped (Table 1, areas 2–9), these range from medium to medium-high, with the greatest scores obtained in the area of inclusive climate and culture (M = 3.97) and the projection of an inclusive culture in the community (M = 4.05). For its part, the most moderate score corresponds to the area of evaluation, research and innovation in inclusion and equity (M = 3.04).

The evidence associated with the dichotomous items reinforces the scores in two of the areas, since 78.8% of the universities have a high-level formal manager who exercises leadership in the promotion of diversity (area 3), while in 93.9% of the institutions, there exists a formal entity that coordinates diversity programs (area 4).

Finally, the general index about institutionalisation of diversity practices reached a medium-to-high level (M = 3.61, SD = 0.70), differing quite significantly from the midpoint of the reply scale used, t (32) = 5.02, p < .001, d = 1.77. Therefore, unlike the level of institutionalisation of diversity policies, the result for practices stood significantly above the theoretical average of the scale. However, it should be considered that the first sample is larger than the second, since in the case of practices, the only source is the opinions of the chief diversity and equality officers who agreed to take part in the survey (33 institutions), while the institutionalisation of policies was based on almost the entire population (81 website analysis).

In addition, as had previously been the case in regard to the institutionalisation of policies index, significant differences were tested between the degree of institutionalisation of practices at public universities (M = 3.66, SD = 0.68) and private universities (M = 3.37, SD = 0.78), although this time the value of t was not associated with an error lower than .05, t (31) = 0.92, p = .36. The Mann-Whitney U test confirmed the result, U = 60.00, z = -.982, p = .35.

To sum up, the first hypothesis of the study is mainly rejected: none of the two diversity general indexes remained below the midpoint of the scales used. What is more, the empirical mean of diversity practices reached a medium-to-high level of institutionalisation. Apparently, a moderate effort in philosophy and policy to promote diversity might have been efficient in terms of actions. One exception to the acceptance of the hypothesis was identified: public universities have made more progress than private universities as far as the institutionalisation of diversity philosophy and policies is concerned.

Relationship between institutional policies and practices

From the correlation analysis of all the areas of institutionalisation (Table 1), two findings can be highlighted. Firstly, most of the correlations between diversity practices are high or moderate-to-high. Secondly, the area of philosophy and policy fails to achieve a statistically significant correlation with seven of the practice areas and only does so with the area of curriculum.

In view of such a weak association between regulations and practices, it was decided to delve further into the relationship between the area of philosophy and policy and the rest by reviewing to what extent each item in area 1 covaries with the practice areas. As can be seen in Table 2, indicator 1 (alignment with institutional statement) correlates significantly with areas 5, 6 and 9, while indicator 5 (staff accreditation) does so with area 8. In other words, while diversity is integrated into institutional statements of mission, to a greater extent, diversity practices are also institutionalised in evaluation, research and innovation processes, as well as in curriculum and staff training; and the greater the consideration given to diversity activities in accreditation and professional development processes, the greater the projection of inclusive culture in the community. The three remaining regulatory indicators do not achieve statistically significant correlations with any of the areas of institutionalisation of practices. Thus, the most generic and abstract indicator apparently predicts some institutional practices, while the more operational indicators do not, except that of accreditation very timidly.

Table 2

Correlations between the five indicators in the area of institutional philosophy and policy, and the eight areas of institutionalisation of diversity practices.

Institutionalisation of practice areas	Area 1 indicators (philosophy and policy)
	1	2	3	4	5
2. Access, participation and achievement	.279	.049	-.202	-.030	.169
3. Leadership	.103	-.023	-.081	-.023	.162
4. Support services	.290	.209	-.156	-.058	.239
5. Evaluation, research and innovation	.466∗∗	-.192	-.003	-.208	.057
6. Curriculum	.439∗	.079	.328	.106	.301
7. Climate and culture	.187	.186	.131	-.112	.326
8. Community projection	.208	.134	.069	-.154	.413∗
9. Training	.425∗	.068	.018	-.227	-.092

∗p < 0.05; ∗∗p < 0.01.

As can be deduced from the absence of a correlation between the institutionalisation index of philosophy and policy and seven of the eight areas of institutionalisation of practices, the general index of institutionalisation of these practices did not reach a statistically significant correlation with the general index of philosophy and policy regarding diversity, r = .23, p = .189.

Differential standardised scores showed that institutionalisation of diversity was found to be higher in the policies index than in the corresponding practices index at 19 institutions (57.58%), while in 14 cases (42.42%), this ratio was reversed (Figure 1). The average of the absolute differential standardised scores was slightly higher in the latter difference (M|zpol = 0.83, M|zprac = 1.13). This descriptive result is applied to almost 40% of Spanish universities, where policies are seen to precede practices in a higher number of institutions than the reverse case. In any case, the variability across institutions is high (in Figure 1, the length of the bars is inversely related to internal consistency: the longer the bars, the lower the consistency between diversity policies and practices). The absence of a generalised pattern explains why indexes do not correlate.

Figure 1

Differential standardised z-scores (zpolicies – zpractices) for 33 Spanish universities. Note. Negative differential z-scores are for universities with a higher index of institutionalisation of diversity practices than policies, while positive differential z-scores correspond to the reverse case.

Taking into account the results presented, particularly the small number of significant correlations between diversity policies and practices, the second hypothesis has to be accepted. Spanish chief diversity and equality officers paint an optimistic picture about diversity practices, although this is scarcely related to institutional philosophy and policy. This latter area is only associated with the curriculum, with the first policy indicator having the highest weight in the association.

Relationship between institutionalisation of diversity and productivity indicators (U-Ranking)

Once the bivariate correlations had been tested between U-ranking and diversity indexes (Table 3), it was concluded that none of the coefficients that associate the general areas of institutionalisation with the ranking indexes attain statistical significance. However, when the indexes of institutionalisation of each practice area were correlated with the U-ranking indexes, one area was identified, specifically climate and culture, which covaried negatively with the of U-ranking's overall classification and with the teaching dimension. Therefore, an inclusive climate and culture that permeates social interaction and participation is inversely associated with the institution's meritocratic performance, although such an effect is only slight in size.

Table 3

Correlations between institutionalisation of diversity indicators and indicators for the U-ranking.

Institutionalisation of diversity indicators	U-ranking indexes
	Overall	Teaching	Research	Innovation
Philosophy and policy area (N = 61)	.140	-.134	.230	.171
Practices areas (N = 30)	-.264	-.313	-.070	-.257
2. Access, participation and achievement	-.178	-.354	.037	-.181
3. Leadership	-.328	-.265	-.172	-.311
4. Support services	-.056	-.186	.096	-.113
5. Evaluation, research and innovation	-.177	-.217	-.024	-.179
6. Curriculum	-.083	-.036	-.048	-.175
7. Climate and culture	-.381∗	-.386∗	-.196	-.330
8. Community projection	-.333	-.307	-.163	-.259
9. Training	-.295	-.244	-.114	-.250

∗p < 0.05.

To sum up, results mainly confirm hypothesis 3, since productivity and diversity indexes do not correlate, except for climate and culture.

Discussion

This study has tested three hypotheses grouped into two objectives (diagnosing institutionalization and coordination of diversity policies and practice, and establishing the relationship between diversity and productivity). Firstly, results have been useful to identify a level of institutionalisation of policies and practices on diversity in Spanish universities above the initial expectation (hypothesis 1). It would correspond to a medium level of visibility of philosophy and policy, as identified from regulatory documentation, while the perception reported by chief diversity and equality officers points to a medium-to-high level in ongoing actions in different areas of institutional practice, although this assertion is discussed in greater detail below.

It is probably macropolitical and legislative initiative at state level that is driving progress in institutional policy at Spanish universities in regard to diversity and also in their practices, beyond micro-institutional documentation. Prominent among such legislative instruments is the consolidated text of the General Law on the Rights of Persons with Disabilities and their Social Inclusion (Royal Legislative Decree 1/2013 of 29 November 2013), which calls for inclusive education to be guaranteed at all levels of the education system for people with disabilities. It explicitly mentions the establishment of positive actions and services by universities. Likewise, Law 3/2007 of 22 March 2007 on the Effective Equality between Men and Women is a highly relevant precursor of the campaign to create equality services or units in different types of institutions (including higher education).

Together with this national trend, the difference found in the institutionalisation of diversity policies between Spanish public and private universities might represent a result that diverges from other areas of the world (data about diversity practices were similar for both types of institutions, but the small number of private universities in the second sample [n = 6; 17.6% of Spanish private institutions in the data collection year] does not allow to consider the comparative result as consistent). In the United States, for example, as is seen in the information obtained from the university websites analysed by LePeau et al. (2018), there are no differences in the commitment to diversity between public and private institutions, but it is also true that the history and evolution of universities in both countries is very different. In Spain, most private institutions came to life during the current democracy (i.e. after 1978) and, just as they score lower on the institutional diversity philosophy and policy index, they also fare poorly in terms of national productivity rankings (a very different situation, for example, from the Ivy League in the USA). Thus, in the U-Ranking, average productivity for the 48 Spanish public universities in the sample (almost the entire public sector) is significantly higher on average than at the 13 private institutions under assessment, t (59) = 2.98, p = .004, d = 0.77 (M = 1.05, M = 0.85), and no private university is found in the top third of the overall ranking. In other words, average academic development in private universities in Spain is lower than at public universities, and this is true in both dimensions of the dual model (neo-liberal and the one associated with diversity policies).

Perhaps the most surprising result of the study and probably its most significant limitation refers to how those responsible for diversity and equality services perceive the actions taken at their universities in terms of diversity, equity, and inclusion. It is possible that the medium-to-high level of diversity practices is associated with some self-protective bias (Schaller et al., 2017). In social psychology, it is sufficiently well-known that individual self-esteem interacts with group identity in such a way that giving a more positive image of the group—in this case, of member institutions—would reinforce individual self-esteem. This is a fundamental prediction of Tajfel's Theory of Social Identity (Tajfel and Turner 1979; see review in Scheepers and Ellemers, 2019). This effect would be further enhanced by the fact that professional self-esteem plays a part in overall self-esteem. In other words, chief diversity and equality officers may have up-valued their daily work attending to diversity and equality, which would also have contributed to strengthening their overall self-esteem in interaction with their social identity. This phenomenon might have resulted in the inflation of scores—a phenomenon that is not coherent with limitations in diversity practices that have been found in Spain and some Latin American countries (Paz-Maldonado, 2020)—and also in a certain halo effect occurring, which could have raised the internal consistency of the questionnaire.

Secondly, bearing in mind the limitation on questionnaire data quality and the consequent caution, philosophy and policy area was associated with actions in the curriculum area but not with other diversity practices. The weakness of coherence between diversity statements and facts was expected (hypothesis 2), as has already been evidenced by other researchers in different national contexts (e.g., Balbachevsky et al., 2019; Elwick, 2020; Thomas, 2018). Even so, innovation in the curriculum must be highlighted because it can be very effective in the medium term to encourage development in other areas, such as entry access for ethnic-cultural minorities to universities (Garibay and Vincent, 2018). In fact, curriculum has been the subject of prolific research given its potential to bring about transformation. Critical pedagogues have been writing on this subject for decades (Apple, 1979; Carr and Kemmis, 1986; Kemmis and Fitzclarence, 1986) and have framed it within a concept in which education is a social activity of a political nature—a concept broadcast by Freire (1974), 1983—with its development regulated according to democratic principles. From the critical paradigm, new curricular concretions of a theoretical and practical nature adapted to current contexts continue to be sought. For example, in reference to higher education, Killick (2018) has proposed Critical Intercultural Practice, based on Critical Pedagogy, as an adequate approach for combating the neo-liberal agenda and for promoting equity in students' experiences and results, diversity as a valuable learning resource, and the empowerment of students as defenders of social justice. This type of curricular practice would be the most appropriate way of achieving the double challenge of responding to student and contextual diversity and at the same time training citizens to live together in a globalised world. The mission facing academics from this critical paradigm would be enormously demanding (Apple, 2016), but it could be effective in transforming what James Thomas (2018, 2020) has called “diversity regimes”—a formal institutionalisation of diversity, but without structural changes.

Thirdly, confirmation of hypothesis 3, related to the second objective of the study, resulted in a very relevant outcome: university policies and practices that address diversity with an inclusive approach bear little relation to the indicators of efficiency and productivity that are used in classifications such as U-Ranking. Only inclusive climate and culture were associated slightly in our data to the overall classification of U-ranking and to teaching, but in a negative sense, which would indicate that the extension of equity and social justice values and attitudes may progress in contrary to that of productivity indicators. In fact, in U-ranking, consideration is not given to criteria associated with inclusive practices but rather to access to funding, productivity, quality and internationalisation (Pérez-García et al., 2018). Therefore, our prediction that a response to social demands for diversity policies and practices does not find its fit within universities’ classic efficiency-seeking mission would be confirmed. This fact could be understood within the context of what can be called asymmetric dual operation, in which the neo-liberal approach would continue to be notably predominant while, in parallel and in response to social demands and legislation of a higher order, innovations are gradually introduced that aim to protect certain vulnerable groups, which is still seen as a subaltern trend.

The asymmetric dual model could also explain the lack of articulation between diversity and the market in higher education in other countries. For example, in the United States, where the neo-liberal model has a strong presence in universities, institutional discourse has nonetheless been shown to favour diversity and inclusion. Indeed, in a study based on an analysis of website contents, Wilson et al. (2012) confirmed 75% of the 80 USA higher education institutions in their sample alluded to diversity in their mission statements (in our study, the percentage approached these authors' figure in a sample group of a similar size: 66.67%). However, heterogeneity in regard to the commitment assumed by universities is remarkable (LePeau et al., 2018) and, in any case, the mere concept of diversity as the plural and representative presence of minorities is not incompatible with an economist-based approach (Ahmed, 2012; Klein, 2016), although it could be perceived as inconsistent by university officers (O'Sullivan et al., 2019). In fact, this kind of focus represents a very limited approach that lacks transforming potential (Chang, 2002) and is not disjointed from each institution's priorities in respect of its position within the hierarchical system of higher education that is governed by mercantile laws (Evans et al., 2017). Certain justification of the measures of affirmative action can even be identified in individualism, market and group hierarchy (Goldstein-Hode and Meisenbach, 2017), perpetuating White privilege (Bhopal, 2019). Within this framework, higher education cannot be understood as a liberating action, but as a banking practice (Freire, 2000) that maintains a climate of oppression and exploitation (Rosa and Clavero, 2021).

Conclusion

The first conclusion of the study is that institutionalisation of diversity is underway in Spanish universities. Based on the most consistent data—those related to philosophy and policies, the level of institutionalisation of diversity would be medium. Secondly, coherence of diversity policies and practices is low, except for the curriculum, with the variability across universities being high. Thirdly and lastly, classic higher education productivity indicators are hardly related to indexes of diversity policies and practices. Therefore, even with its inherent limitations, the study has been useful for several purposes: 1) to establish an early diagnosis of diversity policies and, to a lesser extent, diversity practices at Spanish universities, 2) to detect certain imbalances between policies and practices in the progress made to date, and 3) to propose pedagogical-critical interpretation of the evidence based on the identification of an asymmetric dual model that would explain the parallelism with which two types of unequally weighted goals are pursued—those of the market, of a hegemonic nature, and those of diversity, of a subaltern type.

As a suggestion for university policy leaders and for organizations that conduct global or national university rankings, it would be valuable to include diversity indicators into the operationalisation of higher education quality when institutions are ranked, at least if inclusion remains a goal. Some authors have already begun to propose them (Khan et al., 2019), and this may be an appropriate path to convergence of the current duality of goals. However, “cosmetic diversity” changes (Ford and Patterson, 2019; Hoffman and Mitchell, 2016) or, as is the case of some global rankings, the representation of human beings as institutional goods (e.g., the inclusion of the number of international students as a quality indicator) (Ford and Cate, 2020) might not be the way to overcome the current asymmetric pattern. The cognitive dissonance of promoting rankings and, at the same time, diversity and equity has to be addressed (Stack, 2020). To solve the disharmony, decisions on diversity indicators as part of institutional quality could be more adequately taken within a large partnership between political and economic actors, higher education institutions and the civil society. At the same time, an array of diversity areas and factors must be considered, and not just those that are easily quantified, such as demographic composition. Finally, diversity indicators should consider as many diversity criteria as possible, in such a way that all the minorities that are disadvantaged in power relations could benefit from policies and actions.

In the research domain, alongside the perception expressed by those responsible for central services, one way of establishing better quality indicators in terms of institutional practices for diversity in the future would be to ascertain the opinion of other members of the university community (students, teachers, admin staff, community agents). Furthermore, in upcoming studies, it would be interesting to delve deeper into the interpretation participants give to the questions in the survey stage, to learn, for example, about institutional action models in matters of diversity, protected groups that are associated with the actions they are being asked about, or how respondents’ concepts of equity and social justice bear upon their replies. This could be done via qualitative studies through interviews, as Park et al. (2017) did in a sample of university students in order to clarify the meaning they gave to the items in a survey on diversity. In addition, case studies would provide a more comprehensive view of the articulation of diversity-related curricular and organizational practices in the day-to-day functioning of specific higher education institutions that pursue inclusive excellence.

In summary, the current study has provided an initial overview on the institutionalization of diversity in Spanish universities and its relationship to the market hegemonic framework, but now research on inclusive universities, particularly that conducted from critical perspectives, should hear the voice of all actors, understand their interpretations on issues under investigation, and diversify methodologies and techniques. In doing so, there will be more opportunities to clarify the conditions under which progress on the convergence of diversity and productivity can be made.

Declarations

Author contribution statement

José-Luis Álvarez-Castillo: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Wrote the paper.

Carmen-María Hernández-Lloret, Hugo González-González, Luis Espino-Díaz: Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data.

Gemma Fernández-Caminero: Performed the experiments; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Funding statement

This work was supported by (project number EDU2017-82862-R) and PhD Scholarship (grant number FPU2017/02463).

Data availability statement

Data associated with this study has been deposited at Mendeley Data under the accession number https://doi.org/10.17632/wck98bvrwz.1 and https://doi.org/10.17632/bg8dfsx8pn.1.

Declaration of interests statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.