Abstract
While considerable research has examined white privilege in educational settings, particularly in Great Britain and the United States, less attention has been paid to in-service teachers’ white privilege attitudes, especially in European contexts. This study addresses this gap by analyzing cross-sectional data from 536 teachers across 44 Belgian secondary schools, using multilevel modeling. First, it explores the correlation between teachers’ white privilege awareness (cognitive) and remorse (affective) with the schools’ ethnic composition and teachers’ ethnic prejudice, socio-economic background, level of education and gender. Second, it investigates the relationship between these characteristics and teachers’ involvement with multicultural teaching. The findings highlight the importance of teachers’ individual characteristics in explaining variability in teachers’ white privilege attitudes and demonstrate the importance of these attitudes, particularly the affective component, in explaining variability in multicultural teaching. This study discusses the implications of these findings for future research and social policy.
Ethnoracial 1 educational inequality remains a key issue in the field of education with far-reaching consequences for ethnoracialized students, and society at large (Stevens and Dworkin, 2019). Ethnoracialized adolescents systematically experience lower educational attainment and academic well-being. Numerous underlying mechanisms produce and reproduce this inequality. However, in both theory and practice, it is mostly viewed from a disadvantage-perspective (Powell et al., 2005; Solomon et al., 2005), which focuses solely on the factors that contribute to the disadvantage faced by ethnoracialized students.
One such overlooked dimension underpinning educational inequality is whiteness. Whiteness is the overt and subliminal socialization processes and practices, power structures, laws, privileges, and life experiences that favor the white 2 racial group over all others (Helms, 2017: 718). Like all racial groups, the white racial group is not defined by biological traits, but is socially and historically constructed. Whiteness permeates all aspects of society, including education. A disadvantage-focused approach only focuses on the struggles of ethnoracialized students without interrogating how whiteness influences the academic chances and well-being of white students (Garner, 2007; Solomon et al., 2005).
Bringing whiteness into the discussion of educational inequality is the primary objective of Critical Whiteness Studies (CWS). Building on the groundwork of Critical Race Theory, Critical Whiteness Studies examine how whiteness operates as a dominant yet invisible norm in society and education (Bennett et al., 2019; Jupp and Badenhorst, 2021; Ladson-Billings, 1998; Matias and Boucher, 2023). It challenges dominant narratives such as equal opportunity and neutrality by highlighting the role of whiteness and its most frequently discussed expression: white privilege. CWS argues that educational inequality cannot be meaningfully deconstructed without acknowledging the dominant presence of whiteness within education. Empirical work within this framework typically consists of small-scale studies employing a range of qualitative methods, predominantly conducted in the United States or United Kingdom (Bennett et al., 2019; Jupp and Badenhorst, 2021).
This contrasts with the field of (counseling) psychology, where various quantitative instruments have been developed to measure whiteness as a multidimensional concept (Hays et al., 2023; Schooley et al., 2019). However, these studies tend to adopt a strong individual focus and aim to support white counselors in understanding their white identity and how whiteness plays a role in social inequality, so they can support clients in an optimal way (Hays et al., 2023). Consequently, the empirical evidence is mostly based on (under)graduate students and/or directed toward counseling psychologists (Hays et al., 2023; Pinterits et al., 2009; Schooley et al., 2019). These studies are also predominantly conducted in the United States or United Kingdom.
This study advances the field by integrating the theoretical-conceptual lens of Critical Whiteness Studies with quantitative tools from psychology, applying this interdisciplinary approach to the under-researched context of in-service teachers in Flanders. Teachers play a crucial role in either reproducing or challenging educational inequities, making their perspectives on whiteness particularly important.
Furthermore, we focus on white privilege attitudes. Since McIntosh’s (1989/2005) influential metaphor of the “invisible knapsack,” white privilege is often used as a compelling entry point to help people understand how the system of whiteness operates, and as such, it has received significant attention in both academic scholarship and the field of (teacher) education (Bennett et al., 2019; Jupp and Badenhorst, 2021; Lensmire et al., 2013). However, second-wave CWS scholars have raised important critiques of how this concept has been taken up. While white privilege was initially framed as a consequence of whiteness as a system, its application in teacher education and educational studies has often narrowed the focus on privilege toward individualized self-reflection and moral accountability. Second-wave CWS calls for a more nuanced and complex understanding of whiteness and white privilege; one that considers its intersectionality with other identity dimensions, accounts for the influence of context, and acknowledges the limitations of raising individual awareness as a sole solution.
While taking this critique into account, we deliberately chose to focus on white privilege as a conceptual entry point, given the study’s context in Flanders (the northern, Dutch-speaking region of Belgium). In Flanders, ethnoracial educational inequality is pronounced (De Meyer et al., 2023; Schleicher, 2019), but how whiteness and white privilege play a role in this inequality has received very limited scholarly attention; mirroring trends across much of Northwestern Europe (Stam, 2020; Thomas et al., 2023; Wekker, 2016). This study draws on data collected in 2014–2015, capturing how white educators in Flanders understand white privilege prior to key societal developments such as the refugee influx, the rise of (extreme) right-wing politics in Europe, and the global resurgence of antiracist movements following the murder of George Floyd in 2020 (Hasmath and Kay-Reid, 2021). While the data do not reflect these more recent developments, they nonetheless provide a valuable baseline and shed light on the perspectives of white in-service teachers in a context that remains significantly under-examined in educational research.
Taking all of the above into account, we pursue two aims:
To explore white privilege attitudes (affective and cognitive dimension) among in-service teachers, and how these relate to ethnic prejudice, socio-demographics, and the ethnic composition of schools.
To explore how white privilege attitudes connect to multicultural teaching practices. One important assumption in the multicultural framework is that teachers must be able to step out of their own frame of reference and adapt their teaching practices to the diverse reality of their students (Banks and Banks, 2019). However, this theoretical assumption has little empirical basis and the relationship between white privilege attitudes and multiculturalism is under-researched.
Understanding ethnoracial educational inequality from a whiteness perspective
Race and ethnicity, alongside gender and class, are socially constructed categories through which education is structured and obtains meaning. Education is built on the foundation of white supremacy: not extremist hate groups, but the power dynamic that positions white people as superior and central, systematically privileging white individuals over all others (Gillborn, 2005). For example, curricula in Europe are often Eurocentric and exclude contributions from around the globe (Bhopal, 2018). This exclusion deprives racialized and ethnically minoritized students of the opportunity to learn about their cultural heritage and reaffirms white culture as the norm. Hence, this leads to a dual process in education.
First, if you fit the white norm, this creates “an invisible package of unearned assets” (McIntosh, 1989/2005: 30). For example, white teachers and students are rarely required to reflect on or denounce their race within the classroom (Bennett et al., 2019; McIntosh, 1989/2005). They navigate the system with unexamined ease, as their racial identity is framed as neutral or normative (Garner, 2007; Nayak, 2007). Whiteness is the frame of reference, dictating norms and beliefs that are deeply embedded in the structure and everyday practices of education, yet it is rarely named or scrutinized (Bhopal, 2018). While whiteness is the dominant system, its effects at the individual level are not uniform (Jupp et al., 2019; Jupp and Badenhorst, 2021). All individuals perceived as white benefit from the system, though not to the same extent, as these advantages intersect with other forms of structural inequality, such as class and gender. Importantly, while white individuals benefit from systemic privilege, they also carry the burden of complicity in sustaining a racially stratified society.
Second, whiteness marginalizes those with non-white ethnoracial identities through discrimination, stereotypes, and systemic exclusion, operating as a normative position from which other ethnoracial identities are constructed as deviant or lacking (Garner, 2007; Matias and Boucher, 2023). This leads to a deficit framing of ethnoracialized students, portraying them as unwilling to adapt to “the” norms. It fosters a discourse that positions white students as “worthy,” while their ethnoracialized peers are seen as disadvantaged and “not worthy” (Bhopal, 2018). An analysis of Norwegian teacher education policy and curriculum documents empirically supports this argument by showing that the figure of the “Norwegian pupil” is represented as cognitively superior to the “multicultural pupil” (Fylkesnes, 2019).
Therefore, understanding ethnoracial inequality in education requires including how whiteness and white privilege function (Garner, 2007). Only then we understand that educational inequality is not simply a matter of disadvantage, but the outcome of a system that normalizes whiteness while marginalizing those who do not conform to it.
Variability in white privilege attitudes
The teaching profession operates within the broader system of (invisible) whiteness, yet it is inhabited by individual teachers who can engage in reflections on whiteness, white identity, and privilege. Over the past decades, qualitative studies have provided valuable insights into how white teachers come to recognize—or avoid recognizing—their white racial identity and associated privileges. A central framework for understanding this process is Helms (2017) White Racial Identity Development model, complemented by Tatum’s (1994, 2017) work in the educational context. Both scholars emphasize that developing a critical white identity is neither automatic nor linear. Rather, it is a dynamic, iterative process marked by moments of insight and regression.
A central reason for this complexity is the emotional weight of confronting whiteness. Facing the realities of systemic racial inequality and one’s own role within it, can elicit a range of affective responses, including fear, defensiveness, guilt, and even anger (Amos, 2010; Jupp et al., 2019; Matias et al., 2017; Swim and Miller, 1999; Tatum, 1994). These emotional reactions are an integral part of how individuals process this new knowledge. Emotional discomfort, therefore, often becomes a site of either resistance or transformation.
This wide spectrum of responses is reflected in the systematic review by Jupp et al. (2019), spanning 25 years of race-evasive white teacher identity studies. Silence or disengagement is an important theme, relying on the normalized invisibility of whiteness to avoid racial dialog altogether. Yet, those who are (made) aware react in divergent ways: resistance and trying to reconstruct familiar worldviews is a well-described strategy. Others may become partially aware and reflective, yet their actions remain inconsistent or contradictory—caught between new insights and old habits. Finally, the review shows that becoming truly anti-racist requires specific reflexive whiteness pedagogies; while anti-racism is both possible and necessary, it poses a serious challenge.
While qualitative studies in particular have helped to illustrate and develop a deeper understanding of this complexity (Jupp et al., 2019; Jupp and Badenhorst, 2021) quantitative studies have the potential to chart broader patterns in larger populations and test specific hypotheses regarding this variability. For quantitative research to add to theory development in this field, it needs to consider the complexity of whiteness, for instance by conceptualizing white privilege attitudes as multidimensional—encompassing cognitive, affective, and behavioral components (Pinterits et al., 2009). Selecting instruments that reflect this complexity allows for a more accurate and meaningful assessment of teachers’ engagement with white privilege. This study adopts such an approach by focusing specifically on the cognitive and affective dimensions as measured by the White Privilege Attitudes Scale (WPAS).
White privilege awareness is the cognitive sub-dimension and involves the conscious understanding of white privilege. It is seen as a continuum from no awareness to critical consciousness (Pinterits et al., 2009). However, as previously discussed, white privilege attitudes are not purely about recognition. The emotional responses (i.e. guilt, anger, fear) are assessed in the second, affective subdimension of the White Privilege Affect Scale, namely white privilege remorse. Including this affective dimension is essential, as it is an integral part of the process of becoming aware (Helms, 2017; Matias et al., 2017; Tatum, 1994).
The behavioral component, such as greater willingness to engage in collective action (Ulug and Tropp, 2021), is a crucial outcome of cognitive and affective engagement. In this study, this is addressed indirectly through self-reported multicultural teaching practices (see below), which are theorized as downstream effects of the cognitive and affective dimensions (Swim and Miller, 1999).
Current knowledge about how people score on white privilege awareness and white privilege remorse attitudes and its correlates, both in the general population and among teachers specifically, is limited. However, based on existing empirical and theoretical evidence, respondents may vary in terms of how they score on these two characteristics, which in turn can be explained by a complex interplay of socio-demographic (e.g. gender), contextual (e.g. ethnic composition) and situational (e.g. being confronted with) characteristics, experiences and other beliefs (e.g. ethnic prejudice; Blevins and Todd, 2022; Coleman et al., 2021; Garner, 2007; Pinterits et al., 2009).
Ethnic prejudice
White supremacy operates through a dual process: it centers whiteness as the norm while simultaneously marginalizing non-white groups (Matias and Boucher, 2023). To deepen our understanding of how teachers engage with white privilege, we include a closely related attitude: ethnic prejudice. While distinct, both are grounded in the same underlying ethnoracial structure. White privilege attitudes reflect the degree to which individuals are cognitively and emotionally aware of the unearned advantages conferred by whiteness. In contrast, ethnic prejudice captures another affective dimension: the internalization and reproduction of beliefs that position non-white groups as inferior. Allport (1954) defines ethnic prejudice as “an antipathy based upon a faulty and inflexible generalization,” whether directed toward a group collectively or individuals due to their group affiliation. Whereas white privilege attitudes focus on the ingroup, ethnic prejudice entails biased attitudes directed toward the outgroup. Empirical research among university students demonstrates that heightened ethnic prejudice correlates with diminished levels of awareness and remorse (Blevins and Todd, 2022; Pinterits et al., 2009).
Socio-demographic characteristics
Since not all white individuals benefit equally from whiteness, as class and gender form intersecting systems of superiority and oppression, it is plausible that these intersections shape one’s broader awareness and/or understanding of structural power and inequality dynamics in society more generally. In this study, we focus on gender, educational attainment and social-economic background (parental SES, given that all teachers occupy the same professional role). Some studies among university students in the United States have shown that men report lower white privilege awareness compared to women (Blevins and Todd, 2022; Pinterits et al., 2009). The same is true for white privilege remorse (Pinterits et al., 2009). A parallel explanation has been given for the association between lower parental SES and higher white privilege awareness among adolescents (Grossman and Charmaraman, 2009), but results from another study among university students point in the opposite direction (Blevins and Todd, 2022). As for educational attainment, a study among university students shows large differences between undergraduate and graduate students in scores on white privilege awareness and remorse scales, with graduate students having higher scores (Pinterits et al., 2009).
Ethnic student composition
There is a need for a better understanding of how contextual factors relate with white privilege attitudes (Blevins and Todd, 2022; Coleman et al., 2021). The school’s ethnic student composition determines teachers’ opportunities to interact with ethnoracialized students and have intercultural experiences (Blau, 1968; Moody, 2001). According to a meta-analytic test of intergroup contact theory, intergroup contact in itself is sufficient to reduce ethnic prejudice (Pettigrew and Tropp, 2006). When certain conditions are met, such as repeated, high-quality contact, the effect becomes even stronger. The main drivers behind this effect are reduced anxiety about intergroup contact and increased empathy and perspective taking (Pettigrew and Tropp, 2008). However, the school’s ethnic composition gives an indication of the opportunity to interact and familiarize with ethnoracialized students but does not directly reflect how often teachers interact with these students, as within school segregation is possible (e.g. according to tracks; Van Houtte and Stevens, 2015). Furthermore, the driving mechanisms are focused on the outgroup, while white privilege awareness and remorse require an inward-reflection about taken-for-granted processes related to being white. Hence, the question remains if intergroup contact will also foster white privilege awareness and/or remorse.
A recent U.S. study among 1285 white college students shows a marginally significant relationship between growing up in communities with a higher percentage of white residents and showing less white privilege awareness (Blevins and Todd, 2022). A year-long qualitative study conducted in a school in the U.S. with a student population of predominantly Latina/o students and all white teachers showed that teachers still seemed little aware of their own white privilege (Marx, 2008). A recent study by Rand (2021) shows that almost all white teachers who teach primarily Black, Latinx or Asian students see it as their responsibility to help their students to analyze and respond to social injustices, while 23% of the white teachers with primarily white students do not see this as their responsibility. Hence, the question remains whether the ethnoracial student composition influences the reflections of white teachers about their white privilege?
White privilege awareness and remorse and multicultural teaching
White individuals can differ in their cognitive awareness and affective reactions to this awareness, and both elements can drive behavioral attitudes, like multicultural teaching. Multicultural education aims to “reform schools and other educational institutions so that students from diverse racial, ethnic, and social-class groups will experience educational equality” (Banks, 2004: 3). It constitutes a framework that captures school practices addressing cultural diversity, emphasizing cultural pluralism in curricula, social justice, and equity in both academic outcomes and socio-emotional development (Bennett, 2001). When teachers adopt a multicultural approach, it can reduce ethnic prejudice among their students and promote inclusive teaching practices, intergroup perspective-taking, and a sense of belonging for ethnoracialized students (Aragón et al., 2017; Meeussen et al., 2014; Verkuyten and Thijs, 2013; Vervaet, 2018).
The concept of multicultural education, especially earlier conceptualizations, has been critiqued for resting on the idealistic assumption that recognizing and affirming cultural differences alone can foster equality, while overlooking the structural inequities and power dynamics that underpin inequality (May, 2009). This has led to new or extended frameworks like culturally responsive teaching, anti-racist education, or critical multicultural education (Gay, 2018; May, 2009; McDonough et al., 2023; Nieto, 2017). These approaches recognize the influence of power structures, move beyond superficial acknowledgments of culture, call for critical reflexivity, and center lived experiences of ethnoracialized students—elements often overlooked in earlier models. Moreover, research on multicultural education has the tendency to focus on those students who are disadvantaged without discussing whiteness or white privilege. As Helms (2017) observes, the dominance of whiteness allows white individuals and institutions to determine when racism is acknowledged, and when it is replaced by more comfortable narratives of “diversity” or “multiculturalism.” Thus an explicit link between whiteness and multicultural education is necessary, otherwise it risks being a distraction, rather than a framework of educational equity.
However, the theoretical progress made over time has not always translated into widespread change in practice (McDonough et al., 2023). In Flanders and the Netherlands, teachers tend to place a strong emphasis on the implementation of various cultural viewpoints and diverse examples into the traditional school curriculum (Agirdag et al., 2016; Geerlings et al., 2019). Theoretically, this corresponds with Banks (1993) “content integration” dimension and could be seen as a first step to take. Other domains, such as multilingualism and religious practices, are still frequently approached through an assimilationist lens (Hagenaars et al., 2023). Thus, while the critique needs to be taken into account, grounding educational practices in the multicultural framework remains relevant—particularly when such basic dimensions have yet to be fully realized in everyday practice.
There is limited information about the underlying motives that stimulate teachers to adopt more multicultural content in their lessons (Agirdag et al., 2016; Vervaet et al., 2018). Attitudes play an important role (Gay, 2010), for example, less ethnically prejudiced teachers are more involved in multicultural teaching (Vervaet et al., 2018), but we are not aware of studies that focus on the role of white privilege awareness or white privilege remorse among in-service teachers in Europe.
Understanding whiteness in a European context
There is no fixed or universal meaning of whiteness; as a social construct, it gains context-specific significance depending on geographical and historical conditions (Frankenberg, 2004). The level of attention paid to race and more specific whiteness in the U.S. remains largely unparalleled in Northwestern-Europe (Bhopal, 2018; Dixon and Telles, 2017). In many Nortwestern-European settings, the public and academic discourse focus mainly on categories such as ethnicity, nationality, culture, or migration status, as the discussion of race remains a taboo (Dixon and Telles, 2017; Osanami Törngren and Suyemoto, 2022).
A noteworthy manifestation of this is evident in the Dutch language (Essed and Trienekens, 2008), which does not use “White versus Black,” but “autochthonous” (us) versus “allochthonous” (the other). Though not framed in phenotypical terms, the logic parallels racial categorization: autochthonous individuals—those with ancestry in Northwestern Europe—are seen as civilized, speak the “right” language, and embody “proper” European values (Alba and Duyvendak, 2019; Essed and Trienekens, 2008). They are assumed to belong without even needing to acknowledge their ethnoracial identity. They are the invisible norm, the mainstream. In contrast, allochthonous individuals—typically those with roots in non-European societies—are viewed as incompatible with national norms and remain symbolically excluded, regardless of legal citizenship. Additionally, as described by Goldberg (2006) and recently reaffirmed in the work of Osanami Törngren and Nyström (2024) on Sweden, this “us” versus “them” dichotomy is not only anchored in ancestry or cultural alignment, but also fundamentally in phenotypical whiteness. A “real” European is white, while racialized others are grouped into categories such as “allochthonous” or “immigrant,” implying that they both lack the appropriate ancestry and phenotypical characteristics.
Hence, although discussing race is a taboo in Northwestern-Europe and skin color is not seen as the primary marker of difference, the combination of racial, ethnic and cultural markers is important, and it is undeniable that whiteness is embedded in cultural and institutional norms. A clear dynamic of privilege exists for the “autochthonous” population, while those labeled as “allochthonous” face systemic disadvantages. Given this persistent imbalance, the concept of whiteness and the privilege associated with it remain essential for understanding and addressing ethnoracial inequalities in Northwestern-European education.
Methods
Data
The data used are part of the Racism and Discrimination in Secondary Schools (RaDiSS II) project. To ensure adequate variability and a sufficient number of cases across levels of urbanization and student ethnicities, a multistage sampling approach was employed (for further details: Vervaet, 2018). The data are collected in 45 secondary schools, with student populations ranging from zero to all ethnic Belgian students. Within these schools, we surveyed students and teachers in Grade 6 (comparable with Grade 12 in the American educational framework). Data collection took place during 2014–2015. Following the distribution of an information letter, teachers were invited to fill out an anonymous paper-and-pencil questionnaire voluntary and to return the questionnaire to the researchers via post free of charge. A total of 669 out of 1584 teachers completed the questionnaire, resulting in a response rate of 42%. Given that we are interested in the concept of white privilege, we restricted the analyses to ethnic Belgian teachers. Only 4.5% of the teachers had another than Belgian descent. Missing data were handled by excluding cases listwise. The number of missings for each variable never exceeded 5%. This resulted in a sample size of 536 teachers in 44 schools. In these schools, 3371 sixth grade students also filled out a paper-and-pencil questionnaire (for more information: Vervaet, 2018). The questions related to ethnicity were used to create the actual ethnic student composition variable (see below).
Participants
The study sample consisted of 536 secondary school teachers in Flanders. Of these, 64.0% identified as female, which reflects the current gender distribution among Flemish secondary school teachers. The participants had an average age of 42 years (range: 23–65) and an average of 15 years of teaching experience (range: 1–41 years). All participating teachers identified as ethnic Belgian. In terms of school context, 20% of the teachers worked in schools where 86% or more of the student population was ethnic Belgian, 46% taught in schools with an ethnic Belgian student population between 50% and 85%, and 34% in schools where less than 50% of the students were ethnic Belgian. The student sample consisted of 3371 sixth-grade students, of whom 50.1% identified as female. The average age was 19 years. Regarding ethnic background, 70.8% were ethnic Belgian, 5.9% had roots in Turkey, 8.2% in Morocco, and 4.8% in Eastern Europe (for more details: see below).
Instruments
White privilege attitudes: Awareness and remorse
These attitudes were measured each by a 5-point Likert scale ranging from absolutely disagree (=1) to completely agree (=5) with four items, closely inspired by the work of Pinterits et al. (2009). An example item of awareness is “Autochthonous people have certain advantages in society that allochtonous do not have.” An example item of remorse “I am angry because I know that as an autochthon, I have certain advantages.” The exploratory factor analysis showed the two underlying dimensions as intended with factor loadings on White Privilege Awareness items (WPA) ranging from 0.528 to 0.849 and on White Privilege Remorse items (WPR) ranging from 0.730 to 0.805. WPA had a mean of 3.37 (SD = 0.86) and a Cronbach’s Alpha of 0.828. WPR had a mean of 2.23 (SD = 0.86) and a Cronbach’s Alpha of 0.870.
Multicultural teaching includes 10 items measured on a six point Likert-scale (adapted from: Vervaet, 2018). Teachers indicated the extent to which they integrate examples, data, and information from a variety of cultures and groups in their subject content (Banks, 1993). Teachers could answer each item from absolutely disagree (=1) to completely agree (=5) and not applicable (=6). The last category was recoded to 1, because when teachers answered “not applicable,” it is fair to assume that they did not apply this in their subject. An example item is: “During my lessons at school, I reflect with my students on different cultures.” The scale was created by the mean scores on the 10 items, resulting in possible scores from 1 to 5, with a higher score indicating that teachers practice more multicultural teaching. Exploratory factor analysis reveals that there is one underlying dimension. The item loadings range between 0.404 and 0.804, and Cronbach’s Alpha for the multicultural teaching scale is 0.866. The mean is 3.05 (SD = 0.83).
Gender is operationalized as a dichotomous variable (male = 0, female = 1).
The highest educational attainment of the teachers is a categorical variable with three categories: secondary education degree (=0), teachers who have completed a 3 year bachelor trajectory in higher education (=1) and teachers who have completed a 4 year or 5 year bachelor and master trajectory in higher education (=2). The first category consists of 8.2%, the second of 23.5% and the last category of 68.3%.
Parental socio-economic status was measured using the International Socio-Economic Index of Occupational Status (ISEI; Ganzeboom et al., 1992), derived from the International Standard Classification of Occupations (ISCO-88). This is a metric variable that ranges from 16 to 90. We used the highest score out of the two parents to determine teachers’ socio-economic background. The mean score was 53.01 (SD = 15.25).
Teachers’ ethnic prejudice was evaluated using 18 items primarily focusing on ethnic minorities perceived as economic and cultural threats to Belgian society and adapted from a scale developed by Billiet et al. (1990). The three prominent and visible groups of ethnoracialized people living in Flanders include the Turkish, Moroccan, and Eastern European communities. Consequently, we repeated the six items that assessed respondents’ biases toward these three groups and then computed the average over all items to assess the ethnic prejudice. Sample items include, Turkish/Moroccan/Eastern European people are a threat to our culture and habits and Turkish/Moroccan/Eastern European people take advantage of the support from welfare. Responses were recorded on a 5-point Likert scale, ranging from absolutely disagree (=1) to completely agree (=5). This metric variable has a mean score of 2.49 (SD = 0.61) and a Cronbach’s Alpha of 0.946.
Ethnic composition is measured in two different ways. One individual-level variable probing the teachers’ perception of their own students, and one school-level variable that captures the actual ethnic composition of the whole student population in Grade 6.
Perception of ethnic composition own students is measured with a single question: “In your opinion, how many of your students (%) at this school are allochthonous?.” The average is 42.89 with a large standard deviation of 33.31, which is in line with the broad range of school ethnic composition from 0% to 100%.
Actual ethnic student composition is a variable derived from the ethnic background of all students participating in the study. Ethnicity was primarily determined based on the birthplace of the student’s maternal grandmother, a common method in Belgium given that many students of immigrant heritage are second or third-generation Belgian nationals (OECD, 2008). If these data were not available, we relied on the birth country of the student’s mother. If both types of data were missing, we used the student’s birth country. Subsequently, the proportion of students of other-than-Belgian descent in Grade 6 was calculated for each school. This proportion ranges from 0 to 1, with an average of 0.4 (SD = 0.28).
Data analysis
Given that the data consist of a clustered sample of teachers nested within schools and the dependent variables are metric, the most appropriate method for all analyses was multilevel regression analysis (MLwiN 3.04 software). We specified two levels: level 1 refers to the individual teachers (N = 536) and level 2 refers to the schools (N = 44). All metric variables were grand mean centered. We used random intercept models. In the tables, the unstandardized coefficients are presented, however in the text where pertinent, we discuss the standardized results. The analysis was conducted in two parts. In the first part, we focused on getting a better understanding of characteristics related with white privilege awareness and white privilege remorse. Therefore, we performed two parallel step-wise analyses. Each time, we estimated an unconditional model, to check if the intercept variance at the second level was significant. The unconditional model of both white privilege awareness and white privilege remorse showed that an insignificant part of the total variance is situated at the school-level. Due to the insignificance, we applied multilevel modeling to account for the clustered data, but we did not add the actual ethnic student composition variable.
In the next steps, we started to build the model: first, the socio-demographic variables (gender, educational attainment, parental SES); second, perception of ethnic composition own students; third, ethnic prejudice. In the second part, multicultural education became the dependent variable, as we aimed to examine the relationship between white privilege awareness, white privilege remorse and multicultural education. Once again, the model was built step-wise. After a check of the unconditional model, we started by adding the same socio-demographic variables and the two measures of ethnic student composition; second, white privilege awareness; third, white privilege remorse; fourth, both white privilege awareness and white privilege remorse and finally, ethnic prejudice was added. This step-wise model allowed to gain a thorough insight into how different beliefs relate with multicultural teaching practices, while considering the socio-demographic variables and the ethnic student composition. Multicollinearity between ethnic prejudice, white privilege awareness and remorse was assessed using the correlation matrix and variance inflation factors (VIF), with all values falling within acceptable ranges, indicating no concerns.
Results
Correlates of white privilege awareness and remorse
The unconditional model (Model 0, Table 1) showed that 4.5% of the variance of white privilege awareness is situated at the school level and this is not significant. We can deduct from the fact that there is no significant variance at the school-level that variance in teachers’ awareness is not related with the actual ethnic student composition. Teachers’ perception of the ethnic composition of their own students was not related with white privilege awareness either. Regarding the socio-demographic variables, higher parental SES is related with more awareness. Teachers who studied in higher education for 4 years or more have more awareness compared to those teachers who finished only secondary education. Finally, ethnic prejudice is inversely related to white privilege awareness.
Correlates of white privilege awareness. Results of step-wise multilevel regression analysis (Teachers N = 536, Schools N = 44).
Unstandardized B coefficients are presented with standard errors in parentheses; −2 Log Likelihood values are reported for model fit.
p < 0.05. ***p < 0.001.
The results for white privilege remorse are very similar (see Table 2). First, the unconditional model showed that only 3.1% is situated at the school-level and this is not significant. The absence of significant variance at the school level suggests that differences in teachers’ remorse are not correlated with the actual ethnic student composition. Furthermore, teachers’ perception of their students’ ethnic composition did not correlate with white privilege remorse. Regarding the socio-demographic variables, higher parental SES is related with more white privilege remorse. Teachers who studied in higher education (short-term or long-term) have more remorse compared to those teachers who finished only secondary education. Teachers who hold more prejudiced attitudes show less white privilege remorse.
Correlates of white privilege remorse. Results of step-wise multilevel regression analysis (Teachers N = 536, Schools N = 44).
Unstandardized B coefficients are presented with standard errors in parentheses; –2 Log Likelihood values are reported for model fit.
p < 0.05. **p < 0.01. ***p < 0.001.
White privilege awareness, white privilege remorse and multicultural teaching
The unconditional model showed that 14.3% of the variance in multicultural teaching is situated at the school level. Actual ethnic student composition is significantly and positively related to multicultural teaching: More students with migration roots in the school is related with more multicultural teaching. The perceived percentage of teachers’ own students has no explanatory value once the school-level variable is added (not shown in table). Regarding socio-demographic characteristics, teachers who studied in higher education for four or more years apply more multicultural teaching than teachers who only obtained a secondary education degree (see Table 3). When added uniquely, both white privilege awareness (β = 0.126) and white privilege remorse (β = 0.165) are related with more multicultural teaching. However, when added together to the model (Model 4, Table 3) the association with awareness is no longer significant and only white privilege remorse is significantly related with multicultural teaching (β = 0.136). In the last model with white privilege awareness, white privilege remorse and ethnic prejudice, the prejudiced attitudes of teachers (β = −0.190) are a stronger predictor of multicultural teaching than white privilege remorse (β = 0.078; Model 5, Table 3).
White privilege awareness, remorse and multicultural teaching. Results of step-wise multilevel regression analysis (Teachers N = 536, School N = 44).
Unstandardized B coefficients are presented with standard errors in parentheses; –2 Log Likelihood values are reported for model fit.
p < 0.05. ***p < 0.001.
Discussion
Ethnoracial disparities are a major challenge in the field of education. This study does not focus on the disadvantages faced by ethnoracialized students, but instead focuses on the role that whiteness plays in shaping these disparities. Although whiteness is an important element in understanding inequality, it is frequently overlooked in research, particularly in Northwestern Europe and within the context of quantitative research. Therefore, this study aims to gain a better understanding of white privilege attitudes among in-service teachers in Flanders, Belgium. First, we focus on the attitudes themselves and explore how ethnic prejudice, socio-demographic characteristics and the ethnic composition of the student body correlate with white privilege awareness and remorse. Second, we focus on the relationship between white privilege attitudes and multicultural education. Is more white privilege awareness and remorse associated with more multicultural teaching practices?
With the first research question, we aim at gaining a nuanced view on how teachers perceive white privilege, as very little is known about white privilege attitudes and its correlates among teachers. While there are a few studies on ethnic prejudice (Vervaet, 2018), these typically focus on the white majority’s attitudes toward ethnoracialized groups—essentially, the white group “looking out.” In contrast, this study centers on white privilege awareness and feelings of remorse, which involve the white group “looking in”—a more introspective and self-critical perspective (Garner, 2007). Contrary to what we might expect, the (objective or subjective) ethnic student composition does not relate to white privilege awareness and remorse. Hence, although we know that intergroup contact is a powerful mechanism to lower prejudice (Pettigrew and Tropp, 2006), our findings suggest that mere exposure to diversity is likely to be insufficient to foster the kind of introspection central to white privilege attitudes. Another recent study in the U.S. among white college students shows a marginally significant relationship between the ethnic composition of the community and white privilege awareness, suggesting that long-term, passive exposure also has limited impact (Blevins and Todd, 2022). One possible interpretation is that the optimal conditions under which contact occurs (e.g. high quality, equality in status) may be insufficiently present in school settings. Future research should focus on how and under what conditions contact becomes meaningful. Another possibility is that mechanisms beyond intergroup contact are more relevant in developing white privilege awareness and remorse in educational contexts.
Furthermore, the finding that almost all variance in both white privilege awareness and remorse is situated at the individual level and not at the school-level, suggests the importance of the former over the latter and the need for approaching white privilege awareness and remorse as individual-oriented attitudes. Put differently, as whiteness shapes every aspect of society, this dominant, omnipresence can explain why no differences can be found between schools. Building on white teacher identity studies (Jupp et al., 2019), and taking these findings into account, it will be essential to institutionalize reflexive pedagogies that engage large cohorts of pre-service teachers over extended periods. Such pedagogies should focus on how whiteness—and its consequences, such as white privilege—operate within society in order to effect meaningful and lasting change.
In line with studies among (under)graduate students in the U.S., more ethnic prejudice correlates with less awareness and remorse (Blevins and Todd, 2022; Pinterits et al., 2009). Hence, this relationship is now reaffirmed in a completely different geographic context and among a completely different population. This underscores that there is a connection between the “look out” perspective (prejudiced attitudes toward ethnoracialized others) and the “look in” perspective (self-reflection on white privilege).
Male and female white teachers do not differ in their white privilege attitudes. Higher educational attainment and parental socio-economic status do show positive associations with awareness and remorse (Pinterits et al., 2009). Research displays contradictory results for the association between parental socio-economic status and white privilege attitudes (Blevins and Todd, 2022; Grossman and Charmaraman, 2009), but this study points in the direction of higher parental socio-economic status relating with more awareness and remorse. This finding is especially interesting since this is an adult population with the same profession. It also suggests that occupying a more intersectionally disadvantaged position (e.g. being female or from a lower socio-economic background) does not automatically lead to greater awareness of white privilege. Rather, these results point to the importance of socialization processes—such as family norms or educational exposure—in shaping attitudes toward privilege. Future research could look into the possibility that parental socialization mediates the relationship between parental socio-economic status and white privilege attitudes.
The next research focus, namely the relationship between white privilege attitudes and multicultural teaching, shows two results that warrant further discussion. First, a considerable proportion of the variance in multicultural teaching is situated at the school-level, explained largely by the actual ethnoracial composition of the student body. In schools with more students with migration roots, teachers will also apply more multicultural teaching. The perception of individual teachers about the percentage of their own students has no additional predictive value. All is explained by the school-level composition variable, underscoring the influence of the structural school environment in shaping teaching practices.
Second, the results of this study show that both more awareness and more remorse are linked with more multicultural teaching. While this cross-sectional study cannot answer if white privilege awareness/remorse is a prerequisite to multicultural teaching, it does suggest that teachers who are aware, will apply it more. Future research is necessary to refine this relationship. Another notable finding is that teachers’ application of multicultural teaching is related more with their feelings of remorse (affective) than their awareness (cognitive). This suggests that emotional engagement with privilege is more impactful than cognitive understanding alone. In line with the literature, emotional responses are recognized as a crucial part of becoming truly aware and engaged in processes of white racial identity development and anti-racist practice (Helms, 2017; Matias et al., 2017; Tatum, 1994). A limitation of this study is that we did not assess where teachers are situated within white identity development stages, which could influence how they process and respond to privilege. Nevertheless, the overall pattern indicates that—although emotional reactions such as resistance are possible—remorse appears, in this case to motivate rather than inhibit multicultural teaching. This highlights the importance of addressing the affective dimensions of whiteness in teacher education.
Next to the role of white privilege awareness and remorse, we explored how these attitudes behave when ethnic prejudice is taken into account. The results show that both white privilege remorse and ethnic prejudice are related with multicultural teaching. Hence, a teacher who is less prejudiced and has more white privilege remorse will apply more multicultural teaching than a teacher who is equally low in prejudice but lacks remorse about their white privilege. This suggests that both the reduction of negative attitudes toward the outgroup and affective engagement with one’s own privileged position contribute to multicultural teaching practices. When comparing the two, however, the relationship between ethnic prejudice and multicultural teaching is the strongest. This may indicate that white privilege awareness and remorse exert their influence more indirectly, possibly by shaping ethnic prejudice. Future research could explore potential mediation paths—for example, whether greater awareness leads to more remorse, which then reduces prejudice, or whether lower prejudice allows for greater engagement with privilege. Examining these pathways would enrich our understanding of how teachers move from attitude to action.
This finding also raises broader questions about the dominant framing of multicultural teaching—both in how it is conceptualized by teachers and in how we measure it. It may be that multicultural teaching, as currently understood, aligns more closely with a disposition of openness toward ethnoracialized students (“welcoming the other”) than with a critical reflection on one’s own racial position (“interrogating the self”). This could explain why a measure targeting the outgroup (i.e. ethnic prejudice) is more strongly related to multicultural teaching than a measure directed toward ingroup introspection (i.e. white privilege attitudes). This concern has already been addressed in frameworks such as “critical multicultural education” (May, 2009). Therefore, it will be essential to incorporate these adaptations to the multicultural framework into both survey instruments and teacher education practice.
It is also crucial to consider the limitations of this study. First, the data are collected in the school year 2014–2015, hence, social, political and educational developments may have influenced teachers’ perspectives and practices in ways not captured by this dataset. For example, public discourse surrounding whiteness and white privilege has evolved considerably—shaped by events such as the global protests following the murder of George Floyd—which have contributed to greater awareness and dialog (Hasmath and Kay-Reid, 2021). At the same time, the rise of (extreme) right-wing parties and increasing polarization around issues of race and identity may have counteracted these efforts. Given these contrasting trends, it is difficult to predict how teachers’ privilege-related attitudes and commitments to multicultural teaching may have shifted. This highlights the importance of replicating this work with more recent data and comparing findings in light of the current socio-political climate.
Second, as this study is cross-sectional, we assume the directions between the variables based on theoretical arguments, but only longitudinal data could empirically test the directions between these variables. Third, a mixed-method study with a sequential exploratory design would be ideal to investigate both the determinants of white privilege and how it relates to various educational and broader outcomes. Such a design could, for instance, start with a more inductive, explorative research phase to identify key determinants and outcomes related to white privilege within specific, theoretically important educational contexts. This in turn could feed into the development of a subsequent quantitative research phase that tests these more inductively developed insights on larger, more representative populations. Fourth, given the normative and sensitive nature of the study’s subject matter, responses may be influenced by social desirability bias. All measures are self-reported. While self-report measures are widely used and offer numerous advantages, they are also prone to social desirability and can be limited in capturing less conscious biases. This should be considered when interpreting the results. Future research could benefit from incorporating implicit measures to complement self-reports.
In conclusion, theoretical-conceptual work on white privilege clearly demonstrates the importance of whiteness and white privilege in understanding the workings of the education system and ethnoracial inequalities (Bhopal, 2018; Garner, 2007; McIntosh, 1989/2005). However, it often remains undiscussed in academia, policy and education in Northwestern-Europe, while deconstructing the monocultural reality is necessary to achieve real change in terms of ethnoracial inequality. With this study, we have contributed to this deconstruction, not only by providing insight into the variability of in-service teachers’ attitudes toward white privilege and its relationship to multicultural teaching, but also by adding to the empirical work of studies that have approached white privilege with quantitative research methods, and this in a European context. By broadening both the geographical contexts and research methods, we can begin to address the perspective gap of white privilege in the field of education.
Footnotes
Authors’ contributions
Fanny D’hondt: Conceptualization, Methodology, Formal analysis, Writing-Original draft, Writing- Review & Editing; Jolien Inghels: Writing- Original draft, Writing – Review & Editing; Mieke Van Houtte: Supervision, Writing - Review & Editing; Peter A.J. Stevens: Conceptualization, Funding acquisition, Supervision, Writing- Review & Editing.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funded by Research Foundation-Flanders as part of the research projects: Strategic Basic Research – S004119N and G062114N.
