Really 'doing better' on racism

The abhorrent killing of George Floyd in Minneapolis at the end of May 2020 shocked the world. After being stopped by police for allegedly using a counterfeit $20 note, his arrest culminated in police officer Derek Chauvin kneeling on Floyd's neck for almost nine minutes, resulting in his death. Chauvin has since been charged with second-degree murder following public outcry at the incident, which was captured on camera and posted online for worldwide viewing.

Our aim in this article is not to comment on the death of George Floyd, but to offer a commentary on the response to this incident by institutions across the western world. Specifically, we turn to the response of universities who, following the re-emergence of an active and vocal Black Lives Matter movement, have been quick to respond. These responses have often taken a standard form: 

1. Identify that there are attainment gaps between White and Black and Ethnic Minority (BAME) students, or under-representations of BAME academics in senior leadership positions, or a pay gap between White and BAME employees;
2. Attribute these gaps to systemic racism;
3. Assure members of the community that the institution will do better, usually in a Twitter thread with associated hashtags and links to institutional student support services;
4. Announce the extension of existing programmes (e.g., implicit or unconscious bias training) with a view to showing the institution's action on racial inequality. 

In making these announcements, universities have signalled their anti-racist credentials and committed to action. But will the suggestions that these institutions are making bring about positive change for their Black and Minority Ethnic (BAME) students and academic colleagues?

In this article we want to unpack the stages outlined above, paying attention to how our discipline of psychology can inform discussion and push social change in a more evidence-based way. In short, we argue that there is clear evidence of racial inequality in both student outcomes and staff career trajectories, but the standard response of universities is unlikely to improve matters. More positively, we suggest that there are evidence-based approaches available that appear effective in closing student attainment gaps.

Racial attainment and representation gaps

Almost every university announcement related to the Black Lives Matter movement has highlighted the relative achievements of White vs. BAME students in terms of their attained degree classifications, and of the proportion of university professors who come from BAME (usually Black) communities. For this reason, it is important to consider these so that we can establish the nature and extent of racial inequalities in academia.

In terms of student attainment, data from the Office for Students (2019) show that around 82% of White students achieved a first-class or upper-second-class degree classification (known as a 'good' degree). Among Black students, this figure is 60%, while among Asian students it is 72%. Importantly, this pattern is observable across all potential entry levels (i.e., these racial inequalities of outcome are consistent across A-Level or BTEC grade profiles), suggesting that some degree of inequality is introduced within university environments. 

The racial context of university professorships looks slightly different. According to the Higher Education Statistics Agency (HESA), 85% of professors self-identify as White. This is approximately equal to their representation among the UK population and among academic staff (both 87%). Black individuals make up less than 1% of the professoriate, which represents a substantial under-representation when compared to the general UK population (3%), UK-domiciled PhD students (3%), and academic staff (2%). With these statistics in mind, it is clear that there is under-representation of Black individuals in senior university leadership positions, although this might partly reflect both historically lower Black representation in the UK population (1.5% in the 1991 census) and presumed historically lower Black entry to PhD programs than today.  

Universities as sites of 'systemic racism'?

The policies mentioned at the beginning of this article stem from an admission among senior academic leaders that their institutions and departments are systemically racist. In practice, this means that the racial disparities in attainment and representation are explained by invoking the view that structures and policies placed as barriers to BAME individuals in higher education, with White individuals having an easier experience. 

While university management teams should clearly do whatever is necessary to remove any such barriers, the corresponding challenge is to identify the specific structures and policies at fault rather than continuing with 'business as usual'. Across the UK, universities have led the way in enacting progressive initiatives to 'decolonise the university' and the Race Equality Charter. These projects are specifically designed to address systemic issues of inequality in academia among historically disadvantaged groups. In particular, the Race Equality Charter is a relatively new initiative that has yet to formalise and specify its long-term objectives in characterising a more equal future. As with any initiatives that aim at top-to-bottom changes in institutional practices, these projects will certainly be slow to unfold and be acted on.

With this context in mind, it is understandable that universities wish to take action now to address systemic racism, with many implementing 'immediate' responses to concerns about racism within academia. But will these quick fixes lead to the kinds of gains that we all desire?

Keeping up appearances?

The most common policy announcement across universities in response to the issue of racial inequality is the expansion of existing implicit or unconscious bias training programmes for everyone on campus. The mandatory nature of these schemes has the potential to change the experiences of university students and staff in significant ways. Although these initiatives look good on the surface, the evidence underpinning them shows that these ideas rest on shaky foundations.

A key tool used in the fight against attainment gaps in higher education has been the introduction of implicit bias training packages. Many organisational websites discussing diversity initiatives invite visitors to take an Implicit Association Test (IAT), and current approaches to delivering implicit bias training often require attendees to do the same to assess their level of unconscious prejudice. The IAT is a brief reaction time measure developed at Harvard University where participants must quickly link images of people from different groups (e.g., Black faces and White faces) with positive and negative evaluation words. The assumption of the test is that if, for example, you are significantly quicker to associate Black faces with negative words (compared to the same negative words when they are paired with White faces), you have an implicit bias against Black people that may impact your interpersonal interactions. 

The IAT (and the notion of implicit bias more generally) is not without its critics. Aside from the obvious doubts about whether meaningful conclusions can be made about reaction time differences measured at the level of tens of milliseconds (or hundredths of a second), the evidence base related to the effects of implicit bias training programmes is questionable. In a recent analysis of the effects of these trainings, Forscher et al. (2019) found that they produce, at best, small changes in implicit biases, with baseline IAT scores potentially inflating the extent to which biases actually exist due to their level of measurement (reaction times in milliseconds, rather than observable behaviours). Lai et al.'s (2016) earlier review was slightly more charitable. They reviewed nine different interventions and reported how all programmes produced some reductions in implicit bias. However, these effects did not last for any meaningful amount of time after the training (typically vanishing within a few days). Any changes that are observed typically occur among minority group members, or those who held the least biased views at the beginning of the training (Chang et al., 2019).

Setting aside this important issue about the lack of long term effectiveness of implicit bias training, there is only minimal evidence that scores on the IAT predict any tangible behavioural outcome (i.e., racism out in the world rather than in the respondent's mind). Meta-analyses by Blanton et al. (2009) and  Oswald et al. (2013) found that the test was not able to explain any discernible behavioural outcomes (e.g., a willingness to actually discriminate against BAME individuals). 

Away from criticisms of the effectiveness of formal training programmes, the enactment of diversity-based initiatives have been argued to increase racial bias in other ways. In a review by Dover et al. (2020) it was suggested that fairness, inclusion, and competence signals sparked by initiatives such as implicit bias training risk alienating those from majority groups (risking poorer engagement among these individuals), and implicitly communicate the message that under-represented or minority groups need support in order to succeed. Howell et al. (2017) also found that receiving feedback about one's implicit bias (as is the standard practice when completing an IAT) often leads to defensive responding, and lower intentions to engage in egalitarian behaviour. This leads to serious ethical questions about whether institutions should be delivering training that is designed to reduce biases, but might actually have the opposite effect.

The disparity between the observed empirical evidence for implicit bias measurement and bias-reducing policies on the one hand, and the enactment of such policies in higher education institutions on the other, suggests a desire to look like universities are doing something about a perceived problem, even if that 'something' is ultimately ineffective, or actually harmful to efforts to reduce racial biases and inequalities. That is, these appear to be signals of institutional values rather than considered moves to use the social scientific evidence base to affect real change. In highlighting this issue of image promotion, we now need to turn to the issue of affecting real change. What can we do to actually make a difference to the attainment and representation gaps between White and BAME individuals in higher education?

SCALE-UP

When identifying racism as systemic it is necessary to look at rules, policies, and tangible barriers to attainment that specifically disadvantage some groups and not others. In the absence of specifically biased policies (as is the case in higher education, generally speaking), the curriculum and how it is taught become the target for reform. Recent discussions about decolonising the curriculum have prompted lively debate on social media. However, this approach could contribute towards a learning environment that alienates no student and enhances critical thinking – if done in a careful and non-tokenistic manner, without prioritising lived experience over the scientific method. An example of this might be considering the context within which scientific knowledge is acquired, and considering different cultural approaches to group formation, gender relations and mental health, for example.

However, even at universities where the core curriculum content is to remain the same, there is convincing evidence that changing our teaching practices could have the desired effect of closing attainment gaps between White and BAME students. One method that has been shown to do this is the SCALE-UP model. This is a flipped teaching approach, with an emphasis on student-led problem solving and teamwork in the classroom. Large-scale analyses of SCALE-UP by Beichner (2011) have found it increases engagement among all students, enhances understanding of key concepts, reduces failure rates (33% down to 8% for Black students taking physics; Beichner, 2011), and closes attainment gaps in historically disadvantaged student cohorts. Working together on problems, using theory and empirical research alongside this more practical teaching style, allows staff to work collaboratively with students and reduces power differentials in the classroom while enhancing employability prospects by improving applied skills.

This collaborative approach leads students to work together, rather than to see themselves as being in competition with the rest of their cohort. It leads to interpersonal interactions between students who may not otherwise associate with each other. The social psychological literature is clear that such positive exposure (or 'contact') reduces prejudice and discrimination. This happens whether an interaction occurs in the physical world, online, or is simply imagined (Crisp et al., 2009), with the effect dating back to Gordon Allport's work in the mid-1950s. 

Another promising approach is mentoring, whereby more-experienced members of communities maintain relationships with less-experienced ones, offering support to enhance the success of the mentee. Such schemes can enhance both academic attainment and lower drop-out rates (Campbell & Campbell, 1997) and promote a sense of comfort in the university environment (Bordes & Arredondo, 2005), with peer-mentoring helping BAME students recognise their agency for success (Watson, Sealey-Ruiz, & Jackson, 2016). Not only this, but encouraging BAME engagement in mentoring schemes enhances the prospects of seeing these under-represented groups in positions of authority, which is known to encourage more members of such populations to strive for similar positions (Aish et al., 2018; Morgenroth et al., 2015) What constitutes best practice within mentoring schemes, however, appears to need more research.

Finally, there are many other ways that universities can 'do something' in the immediate term, including anonymised marking, timetable consolidation, academic writing support for entrants from BTEC backgrounds, and improving childcare provision. These are positive for all students, but might be expected to particularly benefit those students with the fewest historical privileges. These groups do not only include BAME individuals, but also women, mature students, and those from poorer socioeconomic backgrounds. Inequalities are reduced by eliminating any effects of stereotyping (in the case of anonymised marking and academic writing support; Malouff & Thornsteinsson, 2016) and removing attendance barriers (through consolidating timetables and supporting those with childcare issues; Markle, 2015). Some universities are already taking action in one or some of these areas, but there is much room for improvement. We would like to see movement on initiatives such as these to truly improve outcomes for historically disadvantaged students, rather than seeing universities expand existing (and seemingly ineffective) approaches to tackling this important social justice issue.

Beyond 'keeping up appearances'

In closing, it is necessary to be aware of the empirical social scientific data on the range of bias-reducing options available in higher education. Universities should be using this evidence base to affect real change for their students and staff, instead of falling prey to the emerging zeitgeist with minimal empirical support. This can be done by questioning our prior intuitive beliefs about the bases of racial inequalities in higher education, and dispassionately looking at the evidence about what works to balance the needs of all students and employees within the context of higher education.

As educators working in psychology, our focus in this article has been on improving the outcomes of our students. This does not negate the need to address inequalities for Black colleagues in terms of their career progression opportunities. We hope that this article, and the arguments made within it, will provide food for thought about how to best engage with the available evidence to affect real change in the progression of under-represented groups. Our current approaches look productive on their face, but favouring appearances over effective change in this way is simply not good enough.

- Dr Craig Harper is a Senior Lecturer in Psychology at Nottingham Trent University. [email protected]
- Harry Purser is a Senior Lecturer at Nottingham Trent University.

Editor's note: See also our range of statements, perspectives and links on 'standing against racism', with more to come in the September edition. Engage with the discussion on Twitter @psychmag

A 'Letter' version of this article is included in the September print edition.

References

Aish, N., Asare, P., & Miskioğlu, E. E. (2018, March). People like me: Providing relatable and realistic role models for underrepresented minorities in STEM to increase their motivation and likelihood of success. In 2018 IEEE Integrated STEM Education Conference (ISEC) (pp. 83-89). IEEE.

Allport, G. W. (1954). The nature of prejudice. Addison-Wesley

Beichner, R. J. (2011). The Student-Centred Active Learning Environment for Undergraduate Programs (SCALE-UP) project. Paper presented at the APS Annual Convention. Washington, DC.

Blanton, H., Jaccard, J., Klick, J., Mellers, B., Mitchell, G., & Tetlock, P. E. (2009). Strong claims and weak evidence: Reassessing the predictive validity of the IAT. Journal of Applied Psychology, 94(3), 567–582. https://doi.org/10.1037/a0014665

Bordes, V., & Arredondo, P. (2005). Mentoring and 1st-year Latina/o college students. Journal of Hispanic Higher Education, 4(2), 114–133. https://doi.org/10.1177/1538192704273855

Campbell, T. A., & Campbell, D. E. (1997). Faculty/student mentor program: Effects on academic performance and retention. Research in Higher Education, 38(6), 727–742. https://doi.org/10.1023/A:1024911904627

Chang, E. H., Milkman, K. L., Gromet, D. M., Rebele, R. W., Massey, C., Duckworth, A. L., & Grant, A. M. (2019). The mixed effects of online diversity training. Proceedings of the National Academy of Sciences, 116(16), 7778–7783. https://doi.org/10.1073/pnas.1816076116

Crisp, R. J., Stathi, S., Turner, R. N., & Husnu, S. (2009). Imagined intergroup contact: Theory, paradigm and practice. Social and Personality Psychology Compass, 3(1), 1–18. https://doi.org/10.1111/j.1751-9004.2008.00155.x

Dover, T. L., Kaiser, C. R., & Major, B. (2020). Mixed signals: The unintended effects of diversity initiatives. Social Issues and Policy Review, 14(1), 152–181. https://doi.org/10.1111/sipr.12059

Forscher, P. S., Lai, C. K., Axt, J. R., Ebersole, C. R., Herman, M., Devine, P. G., & Nosek, B. A. (2019). A meta-analysis of procedures to change implicit measures. Journal of Personality and Social Psychology, 117(3), 522–559. https://doi.org/10.1037/pspa0000160

Higher Education Statistics Agency (2020). Higher education staff statistics: UK, 2018/19. Available from https://www.hesa.ac.uk/news/23-01-2020/sb256-higher-education-staff-statistics

Howell, J. L., Redford, L., Pogge, G., & Ratliff, K. A. (2017). Defensive responding to IAT feedback. Social Cognition, 35(5), 520–562. https://doi.org/10.1521/soco.2017.35.5.520

Lai, C. K., Skinner, A. L., Cooley, E., Murrar, S., Brauer, M., Devos, T., Calanchini, J., Xiao, Y. J., Pedram, C., Marshburn, C. K., Simon, S., Blanchar, J. C., Joy-Gaba, J. A., Conway, J., Redford, L., Klein, R. A., Roussos, G., Schellhaas, F. M. H., Burns, M., & Nosek, B. A. (2016). Reducing implicit racial preferences: II. Intervention effectiveness across time. Journal of Experimental Psychology: General, 145(8), 1001–1016. https://doi.org/10.1037/xge0000179

Malouff, J. M., & Thornsteinsson, E. B. (2016). Bias in grading: A meta-analysis of experimental research findings. Australian Journal of Education, 60(3), 245–256. https://doi.org/10.1177/0004944116664618

Markle, G. (2015). Factors influencing persistence among nontraditional university students. Adult Education Quarterly, 65(3), 267–285. https://doi.org/10.1177/0741713615583085

Morgenroth, T., Ryan, M. K., & Peters, K. (2015). The motivational theory of role modeling: How role models influence role aspirants' goals. Review of General Psychology, 19(4), 465–483. https://doi.org/10.1037/gpr0000059

Office for Students (2019). Differences in student outcomes. Available from https://www.officeforstudents.org.uk/data-and-analysis/differences-in-student-outcomes/ethnicity/

Oswald, F. L., Mitchell, G., Blanton, H., Jaccard, J., & Tetlock, P. E. (2013). Predicting ethnic and racial discrimination: A meta-analysis of IAT criterion studies. Journal of Personality and Social Psychology, 105(2), 171–192. https://doi.org/10.1037/a0032734

Watson, W., Sealey-Ruiz, Y., & Jackson, I. (2016). Daring to care: The role of culturally relevant care in mentoring Black and Latino male high school students. Race Ethnicity and Education, 19(5), 980–1002. https://doi.org/10.1080/13613324.2014.911169