Students’ understanding and support for anti‐racism in universities

Critical Race Theory (CRT) suggests psychology’s contribution to racism takes various forms. Abstractly, racism is promoted through psychology’s flawed theoretical conceptualization as an individualized, inevitable occurrence. Concretely, it occurs because psychology is one of the most popular reasons students come to university and Black Asian and Minoritized Ethnic (BAME) students report racist harassment and less access to formal support whilst there. This racism and student‐proposed anti‐racist recommendations are often ignored. Concretely assessing what racism students face, assessing how students understand racism, and demonstrating student support for anti‐racist recommendations, are CRT‐informed methods of challenging university racism. White (n = 213) and BAME (n = 182) UK students were asked about their estimation of racism, any positive action and discrimination experienced, and their access to university support. Participants were also randomized into multiple conditions where five anti‐racist recommendations were proposed (by Professors N. Patel, R. Smith, or no one). Participant consensus was found in high racism estimations, in benefiting from similar positive actions and in accessing four types of university support. However, White students underestimated racism more so, received less discrimination, and reported more access to three university support types. Almost all participants supported the recommendations regardless of proposer. These results suggest the implementation of anti‐racist recommendations converges with university’s interests as student stakeholder support them. Psychologists in universities can advocate for these recommendations and take other anti‐racist actions.

Misunderstanding racism can be most prominently seen in a recent nationally representative survey (Mohdin, 2018) that found between 52% and 61% of White British participants reported BAME people faced the same amount or less discrimination in various areas of UK life compared with White people (for similar research in Australia see Leviston, Dandy, & Jetten, 2021). Three examples suggest a flawed understanding of racism specifically circulates in universities (Henriques;Reed, Thompson, Brannick, & Sacco, 1991). First, when researchers (Stevenson et al., 2019) asked universities if they had any anti-racist positive action policies in place, five responded they had, despite the policies being more likely to benefit White (working class) students. The second example refers to the predominance of racism research that solely assesses the interpersonal experiences of BAME students rather than any material and structural barriers they face (Kanter et al., 2017). The third example refers to universities' colour-blind practices of failing to even monitor (let alone redress) BAME students' access to established support (e.g., mitigation), collectively leaving BAME students (and staff) without redress to the structural disadvantage they contend with (Mahony & Weiner, 2020;Stevenson et al., 2019). A final example comes from research (Cabrera, 2014;Kanter et al., 2017;Spanierman et al., 2008) finding some white university participants downplayed racism as an individualized system that they too, could be victims of, despite nonetheless holding racist views themselves.
Psychology's flawed conceptualization of racism circulates within universities and CRT reminds us this is a significant barrier to anti-racism (Delgado & Stefancic, 2017;Gillborn, 2019;P. S. Salter & Adams, 2013;Tate, 1997;Yosso, 2013). CRT urges researchers not only to acknowledge racism, as a structural system embedded across multiple institutions, but also to challenge it (Delgado & Stefancic, 2017;Sol orzano & Yosso, 2002;Tate, 1997). To challenge racism, CRT's offers two key way forward. First, CRT advocates we name and concretely monitor racism so as to eschew colour-blind resistance (Delgado & Stefancic, 2017;Gillborn, 2019;Tate, 1997;Yosso, 2013). Specifically, CRT advocates encourage BAME people to counter the neoliberal and deficit narratives applied to them for example by naming the discrimination they have experienced (Crenshaw, 2006;Sol orzano & Yosso, 2002;Yosso, 2013). Crenshaw also advocates for people to identify the structural advantages that might have benefitted them if ever narrating their success (Kimberl e Crenshaw Showreel, 2012). There is therefore a need to assess White and BAME student's university support received, and their experience of positive actions and discrimination encountered.
Second, CRT points to the concept of interest convergence to help combat racism. This is the notion anti-racism occurs only when White powerholders see this as within their own interests (Bell, 1995;Delgado & Stefancic, 2017;L opez, 2003;Park & Liu, 2014). Whilst university interests are complex and not homogenous across the sector, a case could be made that implementing anti-racist steps is within university interests. Specifically, if it can be demonstrated that students, as the largest stakeholders within these public organizations, support them (NUS & Universities UK, 2019). Especially, given these recommendations were originally proposed by students (NUS & Universities UK, 2019;NUS, 2012). As such, attending to concrete anti-racist recommendations directly proposed by students may be a feasible first step in challenging university racism.
The National Union of Students (NUS) is arguably the single largest student consultation body that exists in the United Kingdom. Helpfully, it has made the following recommendations to counter university racism: [Universities should]: Ensure all marking is anonymous (Marking), Ensure curriculums represent people and issues from around the world (Diversification), Offer specific mentoring, workshops & training to BAME students (Mentoring), and Offer specific bursaries to BAME students on courses that are under-represented (Bursaries; NUS, 2012;NUS & Universities UK, 2019). Initial research (NUS, 2015, p. 2) also suggests a final recommendation is also needed: Offering contextual admissions to BAME students (Admissions).
The above anti-racism recommendations are based on well-grounded and comprehensive investigations into racism (McDuff, Tatam, Beacock, & Ross, 2018;NUS & Universities UK, 2019;NUS, 2012) and many recommendations are not new. For example, as evidenced by long standing challenges to curriculum whiteness (Du Bois, 1935;L opez, 2003;Tate, 1997) and advocacy for 'race' based affirmative action (Aguirre, 2000;L opez, 2003;Savas, 2014). Nonetheless they have so far, largely, been ignored (NUS & Universities UK, 2019; Tackling the 'BPOC' Attainment Gap in UK Universities, 2018). This reflects a broader trend to ignore anti-racist calls. For example, British MP David Lammy (2020) recently counted 201 anti-racist recommendations made since 2017 that the UK Government has ignored (despite originally commissioning the investigations that produced the recommendations itself). The recommendations arose from investigations into racism in various British institutions including the workplace, the criminal justiceand immigration-systems. Understanding resistance to such calls is necessary. Gardner and Ryan (2020) offer evidence that the standpoint of the proposer is again relevant to anti-racism. The researchers found that when participants (75% White) read a workplace diversity proposal they were more likely to support it if their hypothetical co-worker was White compared with Black.
The present study Psychology's flawed conceptualizations of racism circulates within universities and, according to CRT (Delgado & Stefancic, 2017), impedes anti-racism. Anti-racism in universities is more likely to be implemented if support from students is demonstrated according to CRT's interest convergence principle (Park & Liu, 2014). Addressing how university students understand racism, eschewing colour-blindness by assessing student's positive actions and discrimination encountered and assessing student support for antiracism (whilst contextualising this by assessing standpoints) are CRT informed steps that 'make good' on psychology's anti-racist commitments. This study therefore aims to answer the following questions: (1) How do UK students understand racism? (2) What positive actions, discrimination and university support have UK students experienced? and (3) Do students support university anti-racist recommendations and does this differ according to the proposer's 'race'?

Method
Design This mixed-methods, cross-sectional, survey follows the 'QuantCrit' approach to research (Garcia, L opez, & V elez, 2018;Gillborn et al., 2018) which acknowledges, in particular, the limitations of quantitative methods in capturing the impact of intersectional racism (Gillborn et al., 2018) but argues if research foregrounds BAME people's experiences, focuses on social justice, critically interprets data as one proxy of experience among others and adds 'voice' by including qualitative data it can be valuable.

Procedure
Prolific (the research management company) was utilized in June 2020, to invite eligible participants (current UK university students) to complete the survey. Prolific is a company that connects potential participants to researchers and focuses on paying participants a minimum wage in contrast to other recruitment platforms (Peer et al., 2017). Like many forms of practical recruitment strategies (including in-person recruitment), recruiting from this platform will have disadvantages and advantages. Disadvantages might include systematic and slight differences in responses between participants across various online platforms including Prolific (on the topic of food knowledge; Armstrong et al., 2021) and higher naivety specifically for Prolific participants (Peer et al., 2017). Potential advantages might include producing more honest response from more diverse participants compared with other platforms (Peer et al., 2017). Steps were taken to mitigate recruitment bias, including having clear, pre-screening criteria, emphasizing incentives are allocated regardless of response and including a clear attention check question to filter out spam or ineligible participation.
A pre-determined number of study spaces (N = 400) were purchased and participants who indicated they were currently in the United Kingdom and were current students were invited (N = 7,531) to participate on a first come first served basis. One survey was sent to participants indicating their ethnicity was White and another to those indicating their ethnicity was Black, Asian, or other minoritized ethnicity [those indicating their ethnicity was 'other' (but unspecified) were not invited]. Fifty-four (19 BAME and 35 White) submissions timed out where participants took a place but did not initiate the survey. These places were reallocated to other participants. Recruitment completed within 10 hours. Within the survey, participants were randomized to 1 of 3 anti-racist recommendations conditions. The conditions were identical except those in the two experimental conditions were presented a 'recommendation vignette' by a Professor Nadiya Patel or Professor Rebecca Smith in the following format: A researcher, Professor Nadiya Patel, has studied university processes and made a list of recommendations to reduce racism. Please indicate the recommendations you think should take place. Those in the control condition were presented with the recommendations anonymously (i.e., "A list of recommendations to reduce racism have been made. . ."). After the survey, participants were debriefed and compensated with credits.

Definition of racism
Participants were asked to select which of the following two definitions, provided by DiAngelo (Dastagir, 2018, paras. 11-13) were more accurate: (1) "Racism is an individual who consciously does not like people based on race, and intentionally seeks to hurt them" (Individual Racism Definition) and (2) "Racism is a group's collective bias backed by legal authority and institutional control" (Institutional Racism Definition). Participants were able to input their own free-text response as well by selecting 'Other'.

Racism estimation
Eight questions taken from Mohdin's (2018) nationally representative racism survey were used to assess participant's estimation of racism. Participants were asked to indicate the level of discrimination for BAME people regarding (1) Access to University, (2) Access to Good Schools, (3) In TV Shows or Films? (4) Access to Finance? (5) In the Workplace? (6) In the News? (7) Access to Jobs? Participants were asked to answer on a 3-point Likert response ranging from À1 (Less) to 0 (No difference) to +1 (Greater). Participants were also asked to respond on a 3-point Likert scale ranging from 0 (None at all), to 3 (A great deal) for the eighth item: (8) To what extent they thought racism was present in UK society today? After excluding unsure responses, scores across the 8 items were averaged with higher, positive, scores indicating greater perceived racism.

Racism accuracy
Nine items were constructed to assess participants' specific estimation of racism in higher education and wider society. Each item was constructed based on a publicly available source demonstrating relatively recent, numerical, inequalities between White British and BAME British people (full sources are provided in Table 3). Participants were asked to estimate: (1) What percentage of hate crimes were recorded as race-related in the United Kingdom in 2017? (2) How much more likely (out of 10) a Black person had of being stopped and searched relative to a White person? (3) What percentage of BAME people will be invited to a job interview relative to a White person? (4) How much more likely, were innocent Black people wrongfully convicted compared to innocent White people on a 10-point scale from 0 (equal likelihood), to 10 (ten or more times more likely) and finally the percentage of (5) White, (6a) BAME and (7b) Black students who were awarded a 1st or 2:1 grade in Higher Education, respectively. Scores on item 5 (White attainment) were subtracted from items 6 (BAME attainment) and 7 (Black attainment) to calculate participants' estimation of a BAME-(6b) and Black-attainment gap (7b), respectively. Participants' responses were subtracted from the actual inequality data to form Racism Accuracy scores for items (1-4 and items 6b and 7b). Finally, these item accuracy scores were converted to percentages and then averaged. Scores ranged from À100 to +100. Negative scores indicated an under-estimation of racism, scores close to 0 indicated accurate estimations, and positive scores indicated an over-estimation of racism.

Positive action and discrimination
Participants were asked about any positive action and discrimination encountered during their education and employment. For positive actions, participants were presented with the following: Kimberl e Crenshaw argues people should identify the positive and negative discrimination they may have faced in their lives. The author (GJ, 30; a gay, white man who grew up in foster care) can think of 4 examples of positive action he has received relevant to his education experience (and later career): (1) a contextual admission (where the university lowered the entry tariff), (2) state financial support to continue in education at A Levels (called the Education Maintenance Allowance), (3) state support towards living costs for university, and (5) state support to buy furniture and equipment when living independently. For discrimination, participants were presented with a similar version of the above except about the author's own experience of discrimination (under the homophobic school legislation: Section 28). For both questions, participants were asked if they had experienced positive actions and discrimination (Yes, No, and Unsure), and if they were comfortable sharing this, what specifically these were (using open-ended responses).

University support
Participants were asked to indicate their agreement with the following seven statements designed to assess students' access to key areas of university support: (1) I have been or will be on a placement that helps my degree/career (Placement), (2) I feel able to gain mitigation, suspension or other form of support for assessments during university (Assessment Support), (3) I feel I can access university mental health support (Mental Support), (4) I feel I can access university financial support (Financial Support), (5) I feel able to share my perspective with lecturers and staff (Share Perspective), (6) I have been or will be given positive references from university staff for future employers (Employment Reference), and (7) I feel university staff treat me equally to other students (Equal Treatment). Participants were asked to respond on a 5-point scale ranging from À2 (Strongly disagree) to 0 (Neither) to +2 (Strongly agree). Responses were averaged across the statements with higher and positive scores indicating more university support and lower, negative, scores indicating less support.

Open-ended responses
Participants were asked if there was anything else, they would like to add, and if they could not think of anything, to 'Please write down a question they thought should have been included in the survey'.
Anti-racist university recommendation support, ranking, and favourability Participants read 1 of the above 'recommendation vignettes' and were then presented with the five anti-racist recommendations: Marking, Diversification, Mentoring, Admissions, and Bursaries. Participants were then asked to rank the recommendations that should take place in priority order of importance from 1 (Most important) to 6 (Less Important), leaving blank any recommendations that should not take place. Participants were able to input their own free-text response as well by selecting 'Other'. Responses were collapsed into three categories (1) More important (1st, 2nd, and 3rd ranking), (2) Less important (4th, 5th, and 6th ranking), and (3) Not selected (Blank responses). Participants were also asked to rank collectively how (1) Useful, (2) Appropriate, and (3) Important these recommendations were from 0 (lowest) to 10 (highest). Scores across these latter three questions were averaged with higher scores indicating greater recommendation favourability.

Anti-racism university combatting
Participants were asked to what extent they thought universities should combat racism by selecting one of the following responses: (0) Universities should not combat racism (e.g., because it is not their job), (1) Universities should combat racism that occurs in Higher Education (HE), (2) Universities should combat racism that occurs in HE & Employment, and (3) Universities should combat racism that occurs in HE & Employment & Society. Scores were averaged with higher scores indicating that universities should combat racism to a greater extent.

Results
Data screening and baseline equivalence of conditions Missing data were minimal (the highest number of missing values was n = 12 (3%) on the penultimate 'Equal treatment' question) and appeared to be random (the Little MCAR test was non-significant; v 2 = 673.18 (625), p = .088). Five participants were excluded given they appeared to be significant multivariate outliers (and nonsensical responses; p's < .05). Finally, White participants did not differ to BAME participants in their study duration, demographics or distribution across conditions (p's > .08).
Analytic strategy Descriptives, frequencies, and comparative statistics from other sources are presented in Tables 1-4. Inferential statistics are presented in the following text from three betweensubjects MANOVAs and two chi-squares where space permits. For these analyses, the more robust Pillai's test and Bonferroni corrected p-values were interpreted (Field, 2009). Confidence intervals of the mean differences between the groups are also reported (equal variances unassumed). Partial eta-squared and Cramer's V statistics served as effect sizes.

Racism extent and racism accuracy
More participants selected the Individual definition of racism (50%) compared with the Institutional (42%). The remaining 8% selected Other which tended to be a combination of both definitions. Most participants estimated BAME discrimination in the seven different areas of UK life as Greater (54%-71%) followed by No difference (14%-25%) and Lesser (7%-10%). Participants differed to Mohdin's (2018) by rating Greater discrimination more frequently. However, participants in both studies were similar in the pattern of their estimations across the 7 UK areas of life. Most participants estimated the extent of racism present in UK society generally as A great deal (n = 213, 54.5%) which was higher than Mohdin's (2018;20%). See Table 1 for full frequencies across both studies. On average, participants underestimated the extent of racism, across the Racism Accuracy items (e.g., Black attainment), by À7% (see Table 2).

Open-ended responses
A total of 202 participants (52%) responded to this question with some leaving multiple responses (n = 229). In line with the study aims, these responses were approximately categorized into three patterns below [except 12 (5%) which were about survey edits such as spelling corrections] indicating support or issues with anti-racism (either as it was conceived in the study or more broadly). Further subcategories were also produced. Example responses to these open-ended questions are available in Table 5. 1. Responses indicating explicit support for anti-racism (n = 50) (1a) Twenty-eight participants indicated they supported anti-racism (e.g., by noting 'Black Lives Matter'), were keen to learn more about anti-racism and were looking forward to receiving the survey results. (1b) Twelve suggested anti-racist recommendations they wished would take place (such as curriculum diversification, university outreach at younger ages, more BAME leaders). (1c) Six indicated general support for positive actions but concern it could be misinterpreted as 'reverse racism', could lead to perceptions of tokenism or may arise without proper research backing it up. (1d) Finally, some participants indicated the problematic homogenization of people into the category BAME  (3a) 'Yes, I detest the separation of people by their skin colour. Asking anecdotal questions will surely skew the results'.
(1b) 'More specific ways on how racism can be tackled in HE'.
(2e) whether students have experienced discrimination from other students (3a) 'I can't help feel a palpable politic bias from these questions. I worry that university students can't escape identity politics, and surveys like this one are an example of this'.
(1c) 'I think more research like this should be done. My course.
. .have just sent an email saying what they plan to do but ideally it should be backed by research'.
(2h) 'Do you think you have ever been racist towards anyone? Would you be able to describe this situation?' (3b) 'As a BAME individual, I will always be grateful that my parents chose this country to move to, rather than any other. I strongly feel almost every society globally has an inherent disliking for any individual whose appearance is diferent to the natives, but that this prejudice is very small in this country compared to others'. noting it obscured the important differences and forms of racism that affected those within this group (n = 4).
3. Responses indicating explicit or implicit issues with anti-racism (n = 14) (3a) This included participants who perceived political bias where the research was believed to highlight racism where it did not exist (n = 2). (3b) Eight participants also indicated that they believed the United Kingdom to largely be a meritocracy, to harbour no/ minimal racism, and racism to be an inherent feature of human nature in any case. (3c) Finally, four BAME participants indicated they experienced no or minimal racism in the United Kingdom or their university.

Anti-Racist higher education recommendations and combatting
Most participants indicated they believed the recommendations should take place (87%-98%) and rated them collectively as highly favourable (M = 7.57, SD = 1.94). Just 16 (4.3%) participants rated the recommendation's favourability as low (≤3). In order of most important, participants ranked the recommendations as follows: Diversification, Marking, Bursaries, Mentoring, Admissions, and Other. Full descriptives are provided in Table 4. Most participants indicated universities should combat racism broadly in Higher Education, Employment, and Society (81%). Twenty participants made other anti-racist recommendations including having robust racist harassment policies (n = 6), removal of any racist students or staff (n = 3), bursaries based on low income/ and 'race' for students (n = 3), more BAME staff (n = 3), mentoring (n = 2), anonymising 'race' in admissions (n = 1), offering anti-racist mental-health support (n = 1), and education about anti-racism (n = 1). Finally, five participants indicated universities should not take anti-racist actions and should instead 'treat everyone equally' (n = 5).
A 6 9 3 9 5 chi square assessed whether participants (White-Nadiya, White-Rebecca, White-Control, BAME-Nadiya, BAME-Rebecca, BAME-Control) differed in their importance ranking (Higher, Lower, and Not) on any of the recommendations (Marking, Diversification, Mentoring, Admissions, and Bursaries). No difference between the groups, in any condition, on any of the recommendations were found (p's >.05). Finally, there was not a significant between-subjects multivariate main effect across the two groups meaning neither White nor BAME participants differed in their Favourability or Combatting responses in any condition (p's >.05).

Discussion
This study sought to assess how university students understood racism, what support and positive action they access and whether they support anti-racist recommendations in response to recent anti-racist commitments from mainstream psychology organizational bodies (APA, 2021; BPS, 2020) and informed by CRT (Delgado & Stefancic, 2017;Salter & Adams, 2013).

Racist estimation and accuracy
Half the participants selected the individual definition of racism. Participants were more accepting of racism's occurrence compared with those in the UK survey (Mohdin, 2018). Furthermore, participants were very accurate in their estimation of the BAME-and Blackattainment gaps. Nonetheless, there was a slight under-estimation across the racism items by around À7% and for some items (e.g., racist hate crimes) this increased to an underestimation of À20%. White students slightly under-estimated racism more so than BAME. This supports previous research findings (Howarth, 2009) and CRT (Delgado & Stefancic, 2017) in highlighting that those who have lived experience of racism are more knowledgeable about it. Nonetheless, the difference was slight, and participants estimated more racism than other samples (Mohdin, 2018). These findings can be taken in two ways, first that many students (especially BAME students) appear to be aware of racism, especially in its varied forms. It may be that Black Lives Matter protests in the Summer of 2020 (Dave et al., 2020) increased student anti-racist understanding, and further, that this knowledge is underestimated by universities. Alternatively, despite an awareness of racism, the individualization-and an underestimation-of it is still a problem. CRT proponents (Crenshaw, 2006;Delgado & Stefancic, 2017) and others (Dastagir, 2018;Reed, 2008;Salter & Adams, 2013) have criticized psychology's promotion of an individualized definition for implying that any group can experience racism and implying that individual solutions will suffice. Furthermore, on average 9% of participants reported they were unsure about levels of discrimination in some areas of UK life and reported the least racist discrimination occurred through financial access (as did participants in Mohdin, 2018). In addition, a small minority of the open-ended responses (6%) did not indicate support for anti-racism; sometimes reaffirming colour-blindness or the meritocracy of society. Some of these responses echoed the 'majoritarian minority story' (Sol orzano & Yosso, 2002), where BAME people feel particularly compelled to defend the status quo and emphasize society is a meritocracy. These findings perhaps speaks to the failure to see racism as a materially unjust system, one that overlaps heavily with class, where BAME people are more likely to earn less, have less inherited wealth and less 'liquid' wealth (Khan & Shaheen, 2017;Ladson-Billings & Tate, 1995;Tippett, Jones-DeWeever, Rockeymoore, Hamilton, & Darity, 2014).
Pragmatically, the need to counter colour-blindness and other conceptualizations that leave racism as a minimal, singular system is underscored (Delgado & Stefancic, 2017;Gillborn & Ladson-Billings, 2010). As Salter and Haugen (2017, p. 124) note: 'the psychology of racism is best understood as the reproduction, maintenance, and internalization of preexisting, historically derived, systemic racial dynamics regardless of individual-level racial animus'. This supports one recommendation in particular: decolonization of the curriculum (Du Bois, 1935;NUS & Universities UK, 2019). Psychology educators can contribute significantly to this. Even in the current accrediting guidance, there is space for curricula to include content on structural racism, e.g., through the BPS's subtopics that relate to racism: 'diversity', 'social constructionism', 'identity' and 'culture' (BPS, 2017, p. 11) or through the more explicit guidance of the APA which notes racism, among other discriminations, is 'highly relevant to teaching about diversity across all five learning goals of the undergraduate psychology curriculum' (APA, 2013(APA, , p. 13, 2019. Indeed, a wealth of resources exist to facilitate this (Fairchild, 2017;Kurtis ß & Adams, 2015;Owusu-Bempah & Howitt, 2000;Salter et al., 2018;Sonn, 2008;Wilbraham, 2016). For example, Schmidt (2019) outlines psychology teaching examples about the violence, impact, and ubiquity of Canadian colonization on human behaviour.
University support and discrimination Between 33% and 45% of all students did not agree they received or could access the seven different types of university support. Furthermore, BAME students reported being less able to access mitigation, extensions, and other forms of assessment support. BAME students also reported significantly more discrimination (than White students). This was largely racism, however, other discrimination types were also reported (e.g., sexism) emphasizing discrimination's intersectional nature (Delgado & Stefancic, 2017). Sixteen percent of BAME students reported experiencing racism in the NUS research (NUS & Universities UK, 2019), whilst in the present study, 50% reported experiencing any kind of discrimination. These results eschew university's colour-blindness by adding to the concrete body of evidence that BAME students are disadvantaged within universities in the United Kingdom (Brown & Jones, 2013;Bunce et al., 2019;McDuff et al., 2018;NUS & Universities UK, 2019;NUS, 2012) and elsewhere (Savas, 2014;Sol orzano & Yosso, 2002). Given universities typically have established support systems already in place (including assessment and mental health support), widening the accessibility of these to all students, especially BAME students, can be an immediate and significant anti-racist gain.
Furthermore, in this study more BAME students felt unable to: share their perspectives with-and receive equal treatment from-lecturers concurring with previous research conducted almost 10 years ago (NUS, 2012). The continuing need for university staff (who tend to be White; Mahony & Weiner, 2020) to retrain in their interactions and treatment of BAME students is again underscored (NUS & Universities UK, 2019;Owusu-Kwarteng, 2020;Solorzano, Ceja, & Yosso, 2000). Furthermore, assessing staff's support of positive actions, as fellow university stakeholders is also important.

Anti-Racist recommendations in higher education
Overall, students were united in supporting the five university anti-racist recommendations including those specifically targeted to BAME students. Encouragingly, participants did not differ in their support of the recommendations by the 'race' of the proposer, thus contradicting Gardner and Ryan's findings (2020) 2 . Participants also made other recommendations including the need to implement robust, zero-tolerance, racist harassment processes (the need for which is backed up by other research; for example, Equalities & Human Rights Commission, 2019).
Most of the open-ended responses (89%) also indicated support for anti-racism too typically encouraging more anti-racist research and actions. These findings add to the growing body of evidence showing anti-racist positive actions are supported by students (NUS & Universities UK, 2019;NUS, 2012) and show anti-racism converges with university management's interests given support for it arises from a key stakeholder (Bell, 1995;Delgado & Stefancic, 2017;Savas, 2014). Encouragingly, the concept of interest convergence, suggests that White students may have supported the recommendations because they can see that these serve their own interests (because anti-racism means studying at a more just institution and/or because some policies may be perceived as directly benefiting white students too, e.g., anonymized marking; Salter & Adams, 2013).
Some might argue these anti-racist recommendations do not go as far as CRT requires, given racism is institutionally embedded (Delgado & Stefancic, 2017), however, they may be important first steps in a wider programme of positive action challenging university racism. Positive action in UK universities tends to be minimal with most offering widening participation schemes only (although for exceptions to this see: Stevenson et al., 2019). These schemes typically attend to students' geographic location and care experience but not racism (which operates 'through' such factors anyway: Gorard, Boliver, Siddiqui, & Banerjee, 2019;Ladson-Billings & Tate, 1995;Warikoo & Allen, 2019). Unsurprisingly then, most participants' positive actions likely arose from these widening participation schemes. There are three reasons why positive actions should be 'race' based. First because class-based positive actions alone will not undo racism (e.g., given it is more likely to reach White working-class students numerically; Khan & Shaheen, 2017). This is supported by the lack of access reported to some areas of university support among BAME participants. Second, because the recommendations, which were proposed by students (e.g., NUS & Universities UK, 2019), were supported by almost all student participants in this study including those recommendations that targeted BAME students only. Finally, anti-racist positive actions may benefit White people as much as they would BAME students (e.g., by better preparing them to work in a diverse workforce; Maxwell & Garcia, 2019;Park & Liu, 2014).
The implementation of anti-racist positive actions requires nuance to avoid the frequent charges of tokenism or reverse racism (Aguirre, 2000;L opez, 2003;Savas, 2014). As Peterson and Rudgers (2017, para. 16) note: 'We recognize that the arguments in support of affirmative action are complicated. They are high-definition messages in a low-definition world'. Furthermore, positive action will need to be defended; as evidenced by the attacks on US anti-racist contextual admissions (Maxwell & Garcia, 2019;Park & Liu, 2014;Tran, 2019;Warikoo & Allen, 2019). Fortunately, there is expert guidance on how to implement anti-racist positive actions (NUS & Universities UK, 2019; Stevenson et al., 2019) and empirical evidence of their benefits (e.g., Clayton, 2012;Maxwell & Garcia, 2019;Peterson & Rudgers, 2017) that can be utilized. Psychologists, within and outside of Higher Education, can advocate here.

Limitations
The sample was recruited via Prolific for a study about views on racism in exchange for financial credits. Thus although the sample was more removed from traditional samples recruited (e.g., through an author's single institution: Bunce et al., 2019;Kanter et al., 2017;Owusu-Kwarteng, 2020;Spanierman et al., 2008) and participation occurred anonymously where incentives were awarded regardless of response, participants who may have held more colour blind or racist views may have avoided participating in the survey, potentially leading to a more anti-racist student view than actually exists in universities. A further limitation is the homogenization of BAME students, including international and home students, which obscures the very real inequalities within these groups (NUS & Universities UK, 2019). The sample here (N = 395) is a fraction of the UK student population (around 2.4 million; Universities UK, n.d) and was over representative of BAME students (48% in this sample vs. 23% in the UK university population; UK Government, 2020). Finally, whilst these survey items were constructed for purpose and based on easily comparable previous research (following the QuantCrit approach; Garcia et al., 2018;Gillborn et al., 2018), they had limitations such as not being broad enough to capture racism in its many varied forms. Thus, there is a danger that the results favour a white status quo (Gillborn et al., 2018) by overstating how informed students are about racism and underplaying how distressing BAME students' university experiences can be. Qualitative work should be attended to those which captures these experiences (e.g., NUS & Universities UK, 2019;Owusu-Kwarteng, 2020;Solorzano et al., 2000;Yosso, 2013). Further research should attend to the questions and topics participants suggested in the open-ended responses (see also Table 5). These include the need to explore racism among university peer groups and staff, to assess racism witnessed (and any proxy impact this might have) and to assess racism's intersections. These are fruitful further research directions provided by key university stakeholders who have insight into university operations: students.

Conclusion
This study sought to understand university students' support of key anti-racist recommendations, students' understanding of racism and experiences of positive actions, discrimination and university support, informed by key CRT tenets (Crenshaw, 2006;Delgado & Stefancic, 2017;Garcia et al., 2018). Findings revealed high support for the anti-racist recommendations across students regardless of the proposer's 'race'. Participants also reported similar experiences of positive discrimination, relatively informed understandings of racism and access to four sources of student support. White students reported less discrimination, more access to three sources of university support and were slightly more inaccurate about racism relative to BAME students. In addition, evidence of colour-blindness, under-estimation of racism and 'majoritarian stories' that support the status quo was also found (Sol orzano & Yosso, 2002;Spanierman et al., 2008). Ultimately, it is within universities' own interests to implement these student-stakeholder supported recommendations, especially given these steps can begin to undo the higher rates of discrimination and areas of lowered university support BAME students experience (a well-borne finding from university research: Equalities & Human Rights Commission, 2019; NUS & Universities UK, 2019; Solorzano et al., 2000). Psychologists, particularly those within universities, can advocate for the top-down implementation of anti-racist steps, can teach racism as a structural system in their curricula, and can help widen the accessibility of university support systems to all students, particularly BAME students. These steps will help to realize psychology's commitment to anti-racism (APA, 2021; BPS, 2020) and help to produce an anti-racist, CRT-informed, education for the benefit of all (Gillborn & Ladson-Billings, 2010;Ladson-Billings & Tate, 1995;Schmidt, 2019). Student support for anti-racism 339