Inequality of opportunity in selection procedures limits diversity in higher education: An intersectional study of Dutch selective higher education programs

the pre-university population. Only one cluster (Artificial Intelligence and Industrial Design in 2020) did not show significantly different odds of admission based on any of the ten back- ground characteristics of applicants. The intersectional analyses show that there are vast differences in acceptance rates between group ALL and group NONE (Table 3). Additionally, Fig 2 and the online figures in S1 File show the acceptance rates for groups which possessed some, but not all, of the background characteristics positively associated with odds of admission. Table 1 and the figures prove that the discussion on inequality in selective admissions cannot be based on single background characteristics only: we need an intersectional understanding of inequality. However, the char- acteristics which resulted in higher odds were not always what one would expect, based on intersectionality theory. For example, men did not always have higher odds of admission, nor did applicants from the highest SES categories or without a migration background. Addition- ally, whereas international evidence suggests predominantly lower odds for rural applicants [32, 33], this was not always the case in our study: in several study clusters, applicants from weakly to non-urban postal codes had higher odds of admission, whereas in others, their odds were significantly lower. This shows that (dis)advantage is context-dependent. In this section, we discuss the significance and implications of our findings. The variable which most often related to inequality of opportunity, was migration back-

ground. In almost all these cases, applicants with a migration background had significantly lower odds of admission (both at HBO and university programs) than applicants without a migration background. This is problematic, as Dutch and international research shows that children with certain migration backgrounds face numerous obstacles in the educational sphere that limit their potential, starting from a very young age [3, 25, 26, 28]. The inequality based on migration background in admission to HE could not be explained by socio-economic characteristics, as there are several SES indicators included in the multivariable model, which already results in adjusted ORs for migration background. This finding is in line with interna- tional research on inequalities based on ethnic background in admission to HE [19, 30, 31]. In clusters where socio-economic indicators influenced the odds of acceptance, applicants from lower and average SES backgrounds were usually (but not always) at a disadvantage. This is disconcerting as well, since these applicants have previously faced several obstacles in and beyond the educational system [55]. As discussed in detail in the results, applicants who do not belong to the highest SES groups are already (strongly) underrepresented in the applicant pools of the majority of clusters. For the period of 2016–2018, Mulder et al. (2022) found that having healthcare professional parents resulted in significantly higher odds of admission in Medicine, Clinical Technology,

Dentistry and Pharmacy. The data presented in this study show that this was no longer the case in 2020 for these programs. However, in other healthcare related study programs, appli- cants with healthcare professional parents did have significantly higher odds: For HBO pro- grams, these are Allied Medical Care, Nursing, Midwifery, Dental Hygiene, Denturism, Optometry, Physiotherapy, Psychomotoric Therapy/Psychomotricity, and Sport Studies. For university programs, these are Psychobiology and Psychology. The previously detected inequality based on migration background and sex (Mulder et al., 2022) are no longer found in the 2020 cluster of Medicine (SU1_2020) and Dentistry and Pharmacy (SU2). We speculate that the broad attention towards potential inequality of oppor- tunity in their selection procedures in recent years has led to effective changes to selection pro- cedures. However, it remains important to note that our analyses concern the nation-wide applicant pools for these programs. A study by Fikrat-Wevers et al. [56] showed that within individual Medicine, Pharmacy and Clinical Technology faculties, inequality of opportunity in selection still existed. Their multi-site study indicated that men performed significantly poorer PLOS ONE Inequality of opportunity in Dutch selection procedures limits diversity in higher education PLOS ONE | https://doi.org/10.1371/journal.pone.0292805 October 13, 2023 17 / 24