Academic support and the BAME attainment gap: Using data to challenge assumptions

Lucy Panesar, Educational Developer (Diversity and Inclusion), Teaching and Learning Exchange, University of the Arts London


Within higher education, multiple factors are said to be causing a gap between Black, Asian and minority ethnic (BAME) and White students. Whilst there is the will to close it, some underlying assumptions amongst university staff maintain the idea of a BAME student skills deficit that tends to point towards academic support as a form of salvation. This article explains how a data driven approach effectively challenged assumptions about BAME student engagement with academic support, by discussing a quantitative study of students at London College of Fashion (UAL), and highlighting issues connected with using institutional data to research ethnic inequalities.


academic support; BAME; attainment; institutional data; monitoring engagement; ethnic categorisation

Despite entering with the same qualifications, Black, Asian and minority ethnic (BAME) students are leaving university with lower outcomes than their White peers. Multiple factors are said to be causing this attainment gap, from the micro (student), through the meso (institution) to the macro (society) (Mountford-Zimdars et al, 2015, p.ii-iii). Whilst there is the will within institutions to close the gap, there are underlying assumptions amongst staff that maintain the idea that there is a BAME student skills deficit, which leads to a reliance upon academic support as a means of overcoming this perceived difference. This article explains how I have used a data driven approach to effectively challenge assumptions about BAME student engagement with academic support at London College of Fashion (LCF, UAL) and reflects on issues around using institutional data to research ethnic inequalities, on both a personal and political level.

A report published by the Higher Education Funding Council for England, investigating the possible causes for differential attainment, identifies profitable interactions with staff as a 'key variable' in student success (Mountford-Zimdars et al, 2015, p.iii). Other studies reveal that ethnic minority students feel less satisfied with their experience of Higher Education than White students and less supported in independent learning (Havergal, 2016), rating family and peer support higher than tutor support (Ferguson and Scruton, 2015, p.44). Although universities offer additional tutor support through library and academic support services, in many cases it is perceived that those students who might profit from such services do not use them. Some academic support teams have been tackling this with interventions aimed at engaging BAME students with their services; the University of Coventry piloted the use of roving, ethnically diverse study support assistants, who succeeded in gaining popularity with more BAME and international students than with White students (Lawrence and White, 2010, p.57-58).

When discussing the BAME attainment gap in the summer of 2015, there was an assumption amongst my colleagues working in academic support at LCF, that BAME students could be making better use of the service to reinforce their independent study and that interventions, like those at the University of Coventry, could help. Whilst this assumption stemmed from a noble belief in the efficacy of academic support in raising attainment (supported by anecdotal evidence of student success after regular use of the service), it also indicated a belief about BAME students having deficits which can be remedied with additional support (Alexander and Arday, 2015, p.21). I was uncomfortable with this notion and in agreement with Pam Tatlow from +milion, who states that attainment differentials ‘cannot be explained by a “deficit” model linked to the BME students themselves' (cited in Alexander and Arday, 2015, p.10). Engagement with academic support may be an effective means for individuals to advance academically, but students’ lack of engagement with the service could not possibly explain the persistent and, at that time, increasing gap between BAME students and their White peers at LCF.

As the newest addition to the academic support team, who was at this time also half way through an MA in Creative Arts Education ‐ and the only person in the team ticking the BAME box ‐ it felt both natural and pragmatic for me to undertake some research around the ethnic status of students using academic support. In line with my discomfort with this perception of a ‘deficit’, the research aimed to test the assumption that BAME students were using academic support less than their White peers. Since the data to determine this was not available to the team at that time, I thought I would have to generate it myself and prepared myself for a dark winter lost in mind-boggling datasets. However, it did not take me long to discover a number of underused institutional data sources that would serve to answer the question. LCF academic support had been inputting attendance data to UAL’s student record system since 2012, using an Access database to produce reports from this information. As this was the same system used to record students' biographical information at enrollment, including ethnicity, it seemed plausible to generate a database showing academic support attendance alongside ethnicity. I just had to ask the right people in the right way.

In the first instance I emailed the UAL Registry with my request, explaining how this data was needed to investigate the BAME attainment gap and that the research would be submitted to another university as part of the assessment for the MA I was completing. This sparked a series of emails escalating to the level of executive management, which established conditions to ensure that students' personal data would be rightly protected within and beyond the institution. In the end two conditions were agreed. The first was that the university awarding my MA relinquish ownership of the research. The second ensured that any possibility of identifying individuals within the research was minimised by representing the broader category of ‘BAME’ rather than specific ethnicities. Despite the fact that I was able to comply with both conditions and appreciated the legal obligation to protect student data, fulfilling the latter condition caused me a degree of existential discomfort, not only because I would be reinforcing the socially constructed binary categories of BAME and White, which in my opinion are reductive and divisive, but also because I would find myself once again sitting awkwardly on a fence.

The sense of reduction that accompanied this initial research request is connected to my own background and sense of heritage. I was brought up in a White British household and community, but tick the box for ‘Mixed ‐ White and Asian’ as my estranged father is from India. Whilst I am proud of my dual heritages, I feel that I am neither one nor the other. When with White people I am the ethnic minority and when with Black and Asian people I am the White person. Although the box I tick seems to me to be misrepresentative ‐ in that it mixes a colour with a continent ‐ at least it signifies my middling position, existing in the middle of the binary spectrum. But with the broader White/BAME categorisation, I am automatically placed on the BAME end of the spectrum for the 50 per cent of Asian in me, and that is where all the assumptions seem to start.

As I reflected on this and the whole process of ethnic categorisation, I started to feel more inclined to challenge these categories rather than assumptions about attendance, which I did not have time to research effectively during the remaining portion of the MA. I had to find a way to stay on-track with the research I had proposed. It was whilst reading about how to conduct institutional research, that I came across some encouraging advice from Smith. Her 2008 publication Using Secondary Data in Educational and Social Research includes examples of how institutional data has been used to investigate educational inequality and discusses the issues of data protection and categorisation arising from this. When contextualising ethnic categorisation, Smith acknowledges the natural and inevitable fluidity of ethnic categories along with the problems this indefiniteness raises for those researching ethnicity (personally, practically and politically), explaining how this fluidity 'reflects the real processes that are occurring in contemporary Britain' (2008, p.324). Reading this insight gave me a better understanding of how categories in higher education have changed ‐ and continue to change ‐ to reflect these contextual processes. I also gained an appreciation of how these categories have enabled us to identify ethnic inequalities in student attainment in the first place. If I wanted to investigate ethnic equality in academic support and effectively communicate within the institution, then I had to use the same language of categories, whether I liked them or not.

Having come to terms with these problematic categories, I was able to meet the second data protection condition I had been set and was then presented with an Access database to generate the necessary data in the form of an Excel spreadsheet. In the resulting spreadsheets produced, myself and colleagues in LCF academic support were able to see ‐ for the first time ‐ the frequency of students’ academic support attendance alongside the additional data that had been requested for further equality monitoring, including ethnicity, gender, age and disability. Subsequent stages required some manual processing, as I first filtered data down to my desired sample (a school within LCF), then secondly as I reduced the various ethnic category codes down into the two broader categories of BAME and White.

Internal LCF reports had indicated that the BAME attainment gap was particularly wide in one particular school, so data was sampled from this school to find out whether their BAME students had been attending academic support less than their White counterparts. Surprisingly, the answer was the opposite: the data revealed that between 2012 and 2015 BAME Home students in the selected school had been attending academic support at a proportionally higher rate than White Home students: not less but more. This was not a cause for me to celebrate, as it raised even more questions, demanding further research to find out why BAME Home students were attending more during this time, how this affected their attainment and whether specific ethnic groups were attending more or less, as well as what that might reveal about racial discrimination. Before I was able to reflect on these questions, another big surprise came my way. At the start of 2016, just after finishing my data analysis, the latest attainment report was published and revealed that the LCF BAME attainment gap in 2014/15 had dramatically reduced, meaning that LCF now had the smallest ‘gap’ of all UAL colleges (UAL, 2016, p.7). This inevitably lead to more questions and speculations as to whether the disproportionately high attendance of BAME students to academic support in the previous years might have contributed to higher outcomes and attainment when they graduated: a question that would take a team of researchers to answer!

Despite having the feeling that my small piece of quantitative research had opened a can of slippery worms, it did successfully challenge the assumption that BAME students have been under using academic support. This outcome has validated my decision to begin with a data driven approach and highlights the importance of basing decisions on what interventions to make (to close the attainment gap) on accurate data rather than assumptions. The research indicates the importance and very real demand for such data to be periodically collected for all schools.

This demand reflects recommendations from the Equality Challenge Unit, outlining the importance of collecting accurate data and monitoring protected characteristics in relation to support services (2014, p.1). In addition, it reinforces the potential demands of the proposed Teaching Excellence Framework (TEF), which has recently been launched at UAL. Once implemented, TEF will draw upon core metrics that indicate the satisfaction, retention and employment of students (DBIS, 2016, p.47) including the most disadvantaged (p.11), further emphasising the need for more sophisticated student record systems. Academic support are currently the only department to use UAL’s student record system in monitoring attendance and equality. This method of data collection does raise issues to do with data protection and categorisation, yet this study has shown that the system can be used to gain a more accurate understanding of who our students are and how they engage with support. Furthermore, since completing this analysis I have learnt that the same system also records student attainment on each unit, which offers the possibility of analysing correlations between a student’s ethnicity, attendance and attainment.

In writing this paper it was my intention to express to colleagues the importance of having ready access to ‘big data’, in order to gain a more accurate understanding of the BAME student experience and tackle the inequalities impacting on attainment. In doing so, I have also critically reflected upon my own feelings about being categorised BAME in an academic context. I end this article feeling both exposed and challenged in a way that I hope, leads to more meaningful connections and to more effective research into tackling ethnic inequalities.


Alexander, C. and Arday, J. (eds.) (2015) Aiming higher: race, inequality and diversity in the academy. London: Runnymede. Available at: (Accessed: 22 November 2016).

Department for Business, Innovation and Skills (2016) Success as a Knowledge Economy: Teaching Excellence, Student Mobility and Social Choice (Cm. 9258). Available at: (Accessed: 22 November 2016).

Equality Challenge Unit (2014) Embedding equality in student services. Available at: (Accessed: 22 November 2016).

Ferguson, B. and Scruton, J. (2015) ‘How do support mechanisms for adult Black and minority ethnic students affect retention, progression and attainment?’, Race Equality Teaching, 33(2), pp. 43‐46.

Havergal, C. (2016) ‘Ethnic minority students less satisfied with university experience’, Times Higher Education, 9 June. Available at: (Accessed: 22 November 2016).

Lawrence, L. and White, S. (2010) ‘Ethnicity and study skills: active intervention in the library setting’, SCONUL Focus 50. Available at: (Accessed: 22 2016).

Mountford-Zimdars, A., Sabri, D., Moore, J., Sanders, J., Jones, J. and Higham, L. (2015) Causes of differences in student outcomes. Available at:,2014/Content/Pubs/Independentresearch/2015/Causes,of,differences,in,student,outcomes/HEFCE2015_diffout.pdf (Accessed: 22 November 2016).

Smith, E. (2008) Using secondary data in educational and social research. Maidenhead: Open University Press.

University of the Arts London (2016) Student Attainment Report Undergraduate 2014/15. Internal UAL Report. Unpublished.


Lucy Panesar joined the UAL Teaching and Learning Exchange in January 2017 as an Educational Developer (Diversity and Inclusion), after working for almost two years as an Academic Support Lecturer at the London College of Fashion. Before that she taught at the University of Creative Arts, where she first began to research cultural diversity and ethnic inequality in response to data highlighting the BAME attainment gap. Lucy’s interest in data relates to her work as a live artist, in which she has interrogated methods used to quantify complex social issues in the guise of her corporate alter-ego Felicity Mukherjee.