The persistence of bias in education: A call for research to move policy and practice from aspiration to results

Abstract

While it seems the world has made significant social and political progress in the last 50 years by, for example, in many countries decriminalising homosexuality, ensuring greater freedom regarding gender identity, acknowledging women’s rights, or removing racist policies, in the last few years, the world seemed to have turned back into a direction that jeopardises the progress made. Educational institutions are often thought of as progressive spaces; however, the institutions themselves operate in a societal and political context that reflect beliefs – wrong or right – that societies hold. Education researchers and policymakers have attempted to conceptualise a response to address bias, but challenges remain. How policies that highlight the importance of inclusive learning environments, and a recognition and celebration of diversity translate into practice and whether the kind of change envisioned is actually taking place in classrooms and lecture theatres is not well documented. While studies have focussed on establishing the links between bias, teacher behaviour and student learning, few studies have documented effective and sustained efforts to dismantle bias in education. We take the stance in this paper that policies aimed at reducing bias can set parameters aimed at shifting practices, however, what is also needed are interventions at every level of the system to change biased beliefs of those implementing policies, leading in schools and teaching students in classrooms. However, typical professional development only highlights bias theoretically and is inadequate to change teachers’ beliefs. Without disrupting beliefs that perpetuate bias, policies remain aspirational, and intervention outcomes remain elusive. We know from decades of research that bias persists in educational systems, institutions and practices; what we need is research situated at the intersection of policy and practice so that we move from identifying inequities to documenting ways to address them.

Prejudice, bias and stereotypes continue to exist in today’s societies. While it seems the world has made significant social and political progress in the last 50 years by, for example, in many countries decriminalising homosexuality, ensuring greater freedom regarding gender identity, acknowledging women’s rights, or removing racist policies, there remain issues of diversity, equity and social inclusion for many groups including the ones named in these examples. Indeed, in the last few years, the world seemed to have turned back into a direction that jeopardises the progress made.

As prejudice, bias and stereotypes exist in society, they exist in educational institutions (e.g. Blank et al., 2016; Yang et al., 2022). Educational institutions are often thought of as the haven of virtue, educating the new generation; however, the institutions themselves, their teachers, students and families operate in a societal and political context and thus bring and are influenced by beliefs – wrong or right – that societies hold. While we seem to grudgingly accept that there is bias in society, we seem surprised when reports from education contexts – from early childhood to tertiary or out-of-school contexts – note young people’s experiences of bias and document the related inequitable outcomes (e.g. DeCuir-Gunby & Bindra, 2022; De Kraker-Pauw et al., 2016; Lautenbach and Antoniewicz, 2018; Lynagh et al., 2015; Meissel et al., 2017).

Education researchers and policymakers have attempted to conceptualise a response to address bias, but challenges remain. Policies in western societies, and increasingly globally, have highlighted the importance of inclusive learning environments, and a recognition and celebration of diversity. However, how these policies translate into practice and whether the kind of change envisioned is actually taking place in classrooms and lecture theatres is not well documented. Inroads have also been made in the study of explicit bias and teacher expectations that influence teacher behaviour and, as a result, impact student learning and achievement (Denessen et al., 2022). However, more deep-seated implicit bias and its influence on teaching and learning have had less attention from researchers (DeCuir-Gunby & Bindra, 2022). Furthermore, while studies have focussed on establishing the links between bias, teacher behaviour and student learning, few studies have documented effective and sustained efforts to dismantle bias in education (for an exemption, see Rubie-Davies et al., 2015).

We take the stance in this paper that policies aimed at reducing bias can set parameters aimed at shifting practices, however, what is also needed are interventions at every level of the system to change biased beliefs of those implementing policies, leading in schools and teaching students in classrooms. However, typical professional development only highlights bias theoretically and is inadequate to change teachers’ beliefs. Without disrupting beliefs that perpetuate bias, policies remain aspirational, and intervention outcomes remain elusive.

Historical context

We start with the premise that ‘there is a need to analyse structural constraints upon political action in historical terms, examining both the establishment of institutions which privileged some and disadvantaged others, and the development of ideologies which set out to justify inequalities’ (Hill and Varone, 2021: p. 48). As Vandenbroeck (2007) details in the context of European early childhood education, nations first grappled with the transitions from their perception of the homogeneous society to recognise that systems served some groups more than others, for example, the 19^th century saw significant cultural and economic gaps between the upper and the lower working class. Colonisation and later labour import and migration in the 1950s and 1960s have further chipped away at the mirage of homogenous societies. While multicultural societies or states have existed throughout history, modern multiculturalism emerged in western societies after World War II in response to the historical wrongs of institutionalised racism and the Holocaust that the war had laid bare.

Multiculturalism not only included dismantling racism but aimed at protecting minority communities of all types and ensuring access to opportunities. The educational response to multiculturalism, however, often saw the goal of education in assimilation – enabling young people from a minority (mainly ethnic minorities) to achieve or behave the same as those from the majority group – rather than celebrating and learning from diversity (see, for example, Hathaway and Fletcher, 2018; Vedder et al., 1996). Early multicultural education thus often failed to challenge the norm in which the minority is the exception. Furthermore, it missed acknowledging the influence of power. These critiques gave rise to an anti-bias approach that emerged from the US context and recognised power relationships, but still used a dualistic model of the majority versus the minority group. A further development in education was the rise of culturally responsive teaching (Irvine and Armento, 2001; Ladson-Billing, 1994), which focused on teacher practices and integrating student culture into teaching (initially the focus was on ethnic diversity). However, the issue remains that for culturally responsive teaching to be impactful, teachers had to acknowledge and reflect on their inherent biases.

Over the past 20 years, there has been a greater realisation of the complexity inherent in diversity. Not only have we come to recognise greater diversity in how people identify (for example, the recognition of diversity within the LGBTQ+ community), there is now greater recognition that all people have multiple identities and that this intersectionality can bring about additional biases. Intersectionality can include multiple group affiliations on which basis people are marginalised (e.g. race, class, gender) as well as personal attributes that engender additional biases (such as physical appearance, age). In education settings, this intersectionality is such that different forms of bias intersect to perpetuate societal bias while also perpetuating experiences of discrimination and marginalisation unique to the school context. The norms of acceptable behaviour in schools, for example, I result in higher incidence of subjective behavioural infractions for non-white male students, where race and gender intersect such that behaviours are deemed ‘disruptive’ that might be viewed as ‘engaged’ if exhibited by a white female student.

In recent years, scholars and practitioners alike have identified and challenged social organisational processes in efforts to not just celebrate diversity but to push back against historical inequality and exclusion. As Myers (2013) writes, ‘Diversity is being invited to the party. Inclusion is being asked to dance’. An equity lens asserts that giving everyone the same thing (i.e. ‘equality’) is not enough to ensure everyone succeeds, but it is instead necessary to provide people with the specific tools they need to succeed. Most recently, the addition of ‘justice’ to diversity, equity and inclusion (DEI) approaches (sometimes referred to as ‘JEDI’; Hammond et al., 2021; Martinez and Truong, 2021) adds an important structural critique – that of examining the broader systems and institutions that perpetuate these biases, and to the work of dismantling barriers to resources and opportunities so all individuals and communities can participate fully and thrive.

Finally, there seems a recognition that ‘there are important differences in how diversity is perceived, [is] embedded in different cultural and historical contexts’ (Vandenbroeck, 2007: 23). However, as argued above, as much as we seem to have made progress in our understandings of the need to uphold principles of diversity, equity, inclusion and justice in educational institutions, individualisation, de-traditionalisation, neo-liberalism and the rise of populism present new challenges to delivering on these principles. Examples of debates, protests and rejections of notions of diversity and the power of privilege can be seen on a daily basis in different contexts around the world.

Policy efforts to reduce bias in education

In recent years, there have been policy efforts to create more socially just, equity-minded schools to combat bias. In the United States, policy efforts at the state and school district level have focused on recruitment and retention of teachers and school leaders whose ethnic/racial background mirror their students (Smith et al., 2022). For example, in Oregon, state policy since 1991 has striven to increase teacher diversity in an indirect effort to reduce bias in education. Despite multiple updates to these policy efforts aimed at greater accountability to meet these diversity goals, the teacher workforce remains predominantly white: while 40% of the state’s students identified as racially or ethnically diverse in the 2021–2022 school year, only 15% of teachers identified as such (Educator Advancement Council, 2022). This gap has remained unchanged since 2010 and clearly requires coordinated statewide efforts at the stages of training, recruitment and retention.

In New Zealand, The Education and Training Act 2020 legislates that schools ‘reflect and integrate’ te reo Māori (the indigenous Māori language), tikanga Māori (‘cultural values and practises’), mātauranga Māori (a concept that refers to Māori knowledge) and te ao Māori (Māori worldview) in the schooling system (Smith, 2023). The implementation efforts of this national policy include a programme called Te Ahu o te Reo Māori which aims to grow and strengthen schools’ ability to integrate te reo Māori into classrooms. This programme was piloted in four regions in 2019. An evaluation of the pilot programme reported overall promising results, but identified several challenges including relevance across school levels, the need for coaching, mentoring and support, and the limited availability of relief teachers to cover classes of teachers participating in the pilots (Smith et al., 2020).

The policy approaches in these two countries share the goal of fostering inclusivity in schools, and both are high on aspiration but low on implementation details. Policy implementation ultimately determines outcomes: ‘the consequences of even the best planned, best supported, and most promising policy initiatives depend finally on what happens as individuals throughout the policy system interpret and act on them’ (McLaughlin, 1987: 172). Well-crafted policies can serve as instruments that set parameters around behaviours to help shift the practices – and ultimately the beliefs – of teachers to interrupt the impacts of implicit bias on students. Current policy efforts in the United States and New Zealand fall short of this by not specifying how their aspirations are intended to be implemented; without this, the likelihood of real change is limited. As McLaughlin (1987) notes, ‘Policy makers can’t mandate what matter’ (p. 172). In order for schools to become more just, inclusive spaces, those at the front line (i.e., teachers) need to be gripped with an urgency to act. As Carr (2021) argues, ‘Good policy requires an understanding and acceptance of the need for change, as well as a convincing vision of how the proposed change will make more people better off’ (p. 72). Teachers need to believe that changes to their practices will directly affect not just students’ social-emotional learning, but their academic outcomes and future success.

Policy implementation efforts are shaped by our history, experiences and communities that present core persistent challenges (Le Fevre, 2020). Aspirational policy without clear implementation guidelines runs the risk of becoming symbolic policy (Hill and Varone, 2021), in which the ‘mobilisation of bias’ confines decision-making to ‘safe issues’ that ‘suppress conflicts and prevent them from entering the political process’ (Hill and Varone, 2021: 28). Schools have taken strides to become culturally responsive, but critics point out that this can present a deficit perspective of ‘othering’ when a nod to non-dominant cultures means the occasional novel by a minority author or a once-a-year celebration of ‘world cultures’. Rather than nudges around the periphery that result in toeknism but little change, ‘decision-makers need courage – courage to analyse the problem in order to be able to make decisions based on evidence rather than just to be seen to be doing “something”’ (Dalziel, 2021: 106). Repeated efforts to ‘do something’ to address bias without changing teachers’ beliefs are inadequate and ultimately breed pessimism that systemic change is possible.

The task of changing beliefs, especially of those who historically hold power, can feel insurmountable, but Hill and Varone (2021) remind us that ‘power is not fixed and immutable…. It is possible to seize opportunities to redefine assumptions and divert the flow of power into new directions’ (p. 31). Historical injustices and bias can be dismantled when policies are coupled with organisational structures that reduce barriers to change. Combatting bias through policy needs to move beyond aspiration with the knowledge that ‘not all ways of tackling a problem will be palatable or tolerated’ (Parkin, 2021: 193). We need ‘to ensure we understand which lever to pull in order to deliver a specific policy outcome for our communities – and to focus our energies on that particular strategy’ (March, 2021: 257).

Resurgence of bias

Policies aimed at combat bias face growing social movements away from such efforts. In the United States, what began as an effort to highlight racial injustice, White Privilege cards have instead heightened tensions. The farcical cards originally created by Joel Patrick, who is Black (see OfficalJoelPatrick.com) have the inscription: ‘White Privilege Trumps Everything. Member since: birth. Good thru: death. Card holder: Scott Free’ (Womack, 2022: 2). As one Oklahoma paper (Moss, 2022) described:

With its glossy white background and black writing, the card could easily be mistaken for a credit or debit card; however, a closer look indicates something much different, with the following language printed on the back: ‘This card grants its bearers happiness because it’s the color of your skin and not the choices that you make that determines your ability to be successful’.

The cards have been maliciously misappropriated by white students bearing the cards, and quickly spread to commodification, with the cards for sale on Amazon (Womack, 2022). The sale of White privilege cards could have been used the basis for a unit on social unrest, but legislative bills in place in many US states ban lessons that could contribute to students feeling discomfort or guilt because of their race or gender (Fortin and Heyward, 2022). In the face of these policy restrictions, educators face potential censure if they try to directly confront bias.

In the United Kingdom, new policies have been enacted aimed at curtailing educator bias against students with natural Afros. In one case reported on by The Guardian, a student was told her hair was ‘getting too big’ and was suspended (several times) for not conforming to the school’s policy that required ‘afro style hair [to]… be of reasonable size and length’ (Williams, 2022: 2). After a court battle, the UK’s Equality and Human Rights Commission issued new guidance to schools curtailing racially biased appearance policies: natural Afro styles are allowed as a protected characteristic, but as argued above, policy enactment does not translate into fidelity of policy implementation. Indeed, the 2022 ruling is implementation guidance based on the UK’s 2010 Equality Act banning discrimination against hairstyles ‘associated with race or ethnicity’ (Equality and Human Rights Commission, 2022: 3), which clearly did not result in the end of racial bias in UK schools. Clearly, institutionalised, systemic bias needs confronting at the individual as well as the legislative level, as it is ultimately educators’ beliefs and behaviours that perpetuate or curtail bias in school contexts.

Interventions aimed at dismantling bias in education

Our scientific understanding of how and when stereotypes and prejudiced attitudes can shape judgement and behaviour has advanced; however, a large gap remains between this research and its practical application. Evidence is growing that certain frameworks and organisational and interpersonal strategies can be more successful than others in promoting equity and inclusion, and thereby reducing incidents of bias (Schmader et al., 2022). A rich social psychology literature indicates there are numerous pathways through which bias can be disrupted, including through changing social norms, fostering inclusive interactions, increasing personal motivation for anti-bias efforts, increasing awareness of when biases are expressed, and/or training people to counteract their own or others’ biases (Schmader et al., 2022). Increasingly, studies are revealing that popular approaches may not significantly alter biased behaviour. For example, though anti-bias trainings are ubiquitous, unless a training cultivates genuine motivation to foster inclusion by either management or employees, anti-bias trainings can spark backlash and contribute to hostile workplaces and greater inequities (Emerson, 2017). Though many anti-bias programs focus on increasing awareness and understanding of implicit bias, increases in awareness has not been shown to lead to behaviour change that reduces bias.

Similarly, techniques that involve engaging with another’s perspective appear unfruitful, at least in short term implicit bias reduction, while other techniques, such as exposure to counter-stereotypical exemplars, may be more effective (FitzGerald et al., 2019). Broader cultural norms for inclusion are likely to inform targeted group members’ attributions for – and resilience to – encounters with biased behaviour, which has implications for their health and well-being (Berkowitz et al., 2017). The most effective interventions create partnerships across identity lines, when a critical mass of people within an organisation or community work together toward the shared goal of creating an inclusive culture that reduces the harm that bias causes and fosters well-being for all (Schmadeer et al., 2022).

In education, we propose that teacher beliefs and behaviour are central to the success of anti-bias interventions, though the evidence for this assertion is to be found not in anti-bias research, of which there is little, but instead in research focused on social-emotional learning, neuroscience and behaviour, and culturally responsive teaching. Key to all of these diverse domains are a teacher’s belief in a growth mindset perspective – that bias is not set in stone but, instead, that acting more equitably, in a way that undercuts bias, can be learned and made more effective through trial and error (i.e. practice). Teachers’ growth mindsets are critical to the work of reinforcing classroom and school norms that value equity and inclusion, to fostering inclusive interactions, and to creating classroom and school cultures in which students (and adults in schools) commit to trying to counteract their own and others’ biases. For teachers’ growth mindsets to have the most positive impact on students, teachers need to model and reinforce hard work (their own and others’), practice and hone effective strategies, and elicit help from other people. Teacher behaviours rooted in growth mindset are powerful because they are social environmental cues that can draw out a student’s nascent growth mindset (Walton and Yeager, 2020), and potentially affirm that anti-bias beliefs and behaviours are beneficial and legitimate in the school context.

Common elements run through successful initiatives that can promote educational equity and inclusion in schools that are not referred to as anti-bias interventions per se but, as mentioned, are drawn from social and emotional learning, neuroscience and education, and culturally responsive teaching. At the interpersonal level, a student’s feelings of belonging in school – inclusive school cultures – may be founded on the dedication of teachers and other school personnel to building positive relationships with students that are marked by rapport, effective communication, high expectations and high supports, and rooted in the increasing knowledge of a student’s unique characteristics – personal, cultural and familial – that are framed as assets and catalysts for a student’s school success (Patrick et al., 2021). Simple strategies may have high impact, such as when teachers take care to pronounce their students’ names correctly (Moore et al., 2020). Teachers’ consistent use of microaffirmations with their students can create classroom cultures that foster inclusion, which is to say, attending to and commenting when they notice students acting in ways that are prosocial or ‘catching kids being good’, and framing praise in relation to specific, observable behaviours that reflect knowledge or skill acquisition rather than immutable characteristics the student has no control over (Johansson et al., 2020).

Restorative practices seek to minimise the negative impacts on students of teacher subjective discipline actions, such as being sent out of the classroom to be ‘disciplined’. To provide one example of a restorative anti-bias intervention, in 2013–2014 a secondary school in Oregon, USA trialled a five-module professional development curriculum for teachers and administrators. Pre–post results indicated gains in staff knowledge of restorative practices (p < 0.0005), staff use of restorative practices (p < 0.0005) and staff use of proactive circles (p = 0.003). Staff also improved ratings of the discipline system as conducive to building relationships (p = 0.006). Student perceptions also improved of disciplinary fairness (questions include ‘students are treated the same in the discipline process at my school irrespective of their race/ethnicity’; and ‘the disciplinary rules and related processes, such as office referrals, used in my school… are fair; make my school a better place to learn’). Student race predicted perceptions of disciplinary fairness at pre-implementation, but not at post-implementation. Office discipline referrals decreased from 16 to 4%. White–Latino disparity in discipline decreased from 2.68 to 0.47% (Vincent et al., 2016). The anti-bias effects of restorative practices on school discipline have been documented elsewhere (Weber and Vereenooghe, 2020; Gregory et al., 2016; Kline, 2016), including in a randomised controlled trial (Augustine et al., 2018), and yet bias persists in educational practices and outcomes.

One of the barriers to dismantling bias in education is the tendency for school systems to jump on ‘quick fixes’ of a programme that has had success elsewhere, like the intervention trialled in Oregon a decade ago. Inherent in restorative approaches in schools is the recognition that positive reinforcement is more likely to lead to desired student behaviours than punitive discipline. This recognition has led to the endorsement of a ‘magic’ formula of five positive interactions for every one negative interaction; researchers point out that the endorsement of any specific ratio may have strong face validity but little empirical support, and they call for greater nuance (Sabey et al., 2019).

A further barrier to disrupting biases in education is relying on individual equity-minded teachers to attend to and actively counter microaggressions that are rooted in and can perpetuate negative stereotypes (Hammond, 2014; Sue and Spanierman, 2020). The systemic nature of biases requires addressing the broader social, economic, political and environmental determinants of students’ learning. Two examples are the provision of delicious and nutritious food in schools to all students, not just to those who qualify based on family income (Martinelli et al. 2022), or the creation of built environments whose broad doorways, hallways, elevators and easily accessible sidewalks are of benefit to all students, including those with disabilities (Titchkosky, 2011). Similarly, schools’ recognition during COVID that a students’ ability to attend and take part in distance learning might be due to their family’s socioeconomic status or rurality rather than a student’s desire to attend and learn has led to a decoupling of attendance from grades, and to a widespread consideration of grading as an equity issue in K-12 schools (Feldman, 2018). Through efforts such as those above that address social determinants of learning and school engagement, school policies, practices and personnel seek to provide students with equitable access to resources while undercutting potential stigma that might be associated with accessing those resources. The burgeoning field of anti-bias educational research would do well to not only investigate the efficacy of interventions at dismantling bias, but at related processes of stigmatization/destigmatization of those who might benefit most from anti-bias initiatives.

Call for research to inform policy and practice

In this paper, we have argued that policy alone is insufficient to dismantle bias in education. Neither are typical approaches to professional development or programs that rely on individual teacher action. Rather, policies that move beyond aspiration by including context-specific levers for implementation (see Smith and Thier, 2017) coupled with system-wide interventions that interrupt the historical, systemic and pervasive biases in education to increase equity of opportunities and outcomes. We offer examples of well-intentioned policies and programmes above, but we don’t yet feel there is sufficient empirical evidence spanning a wide array of contexts and cases to provide concrete guidance to policymakers and educators.

More empirical research is needed to better inform policy and practice agendas. Studies are needed that examine how educators are responding to (or obstructing) policies aimed at combating systemic bias. More is needed to understand how organisations operationalise policies aimed at decreasing bias into expected behaviours. Further, we know that teacher defensiveness and/or dismissiveness can be a barrier to embracing changed behaviours, opening a research agenda to document how these barriers are overcome. Finally, research is needed that examines ways to increase implementation fidelity, after policy enactment, after professional development and after an intervention ends. We know from decades of research that bias persists in educational systems, institutions and practices; what we need is research situated at the intersection of policy and practice so that we move from identifying inequities to documenting ways to address them.