Bar exams: Research highlights attainment gap between white and ethnic minority wannabe barristers
As Oxbridge grads clean up across the board
Reformed bar exams still put ethnic minority students at a disadvantage, Bar Standards Board (BSB) research shows.
Changes to the Bar Course’s centralised assessments in 2016 failed to close the attainment gap between white and non-white candidates.
The report also shows that going to Oxbridge or another Russell Group uni is a significant predictor of bar exam performance.
These differences are after controlling for other variables, such as how well someone did on their degree.
BSB number-crunchers looked at the bar exams in civil litigation, criminal litigation and professional ethics. Between 2011 and 2015, these all involved a mix of multiple choice and problem questions. From 2016 to 2019, they changed to multiple choice only for the two litigation exams and problem questions only for ethics.
Previous research had found, under the old format, ethnic minority bar candidates scored 4.7 points lower than white students. The new format seems to have done little to improve that, with the report stating:
“Those from Asian/Asian British, Black/Black British, Mixed/Multiple ethnic backgrounds, and from other ethnic backgrounds were all predicted to do worse on the assessments than White students on each centralised assessment, even when controlling for other variables such as prior academic attainment.”
The 2016 format change “did not appear to lead to a consistent change in differential outcomes”.
This is not, the report cautions, down to non-white people being inherently worse at exams. “Ethnicity per se is most probably not the effective variable affecting students’ success on the BPTC”, the statisticians write. “Instead, it is a proxy for other factors correlated with ethnicity that are not controlled for in the analyses that have been undertaken; these are likely to relate to socio-economic status and psychosocial-cultural experience (including family and other support networks), and differing behaviour towards those of different ethnicities.”
Meanwhile, Oxbridge types tend to do better across the board. Graduates of Oxford and Cambridge scored 14.4 points higher in civil litigation, 12.3 points in criminal and 10.1 points in professional ethics (again controlling for other variables).
Graduating from a non-Oxbridge Russell Group uni was also a predictor of higher exam scores, although to a lower extent.
For all the latest commercial awareness info, news and careers advice:
Sign up to the Legal Cheek Newsletter
18 Comments
Anonymous
Taking a degree score into account will not help substantially control the experiment, because degree standards are different from university to university; bachelor degree classifications are roughly equally distributed at most universities, even though the entry standards very wildly differ. Clearly, the average ability of Chichester students does not equal that of Oxonians at the end of their degrees. The reason that Oxbridge graduates do so well on the test compared to the rest is obviously that they are on average more intelligent. If we compare black and white Oxbridge graduates’ (who attained firsts) average bar exam results, the difference in outcome will be far less than the that between black and white examinees generally.
Anonymous
There are plenty of Black and Asian Oxbridge graduates with Firsts who never get pupillage.
These extra tests are a distraction from what’s actually going on.
Capable Black and Asian candidates simply aren’t being picked for pupillage by barristers. End of.
Anon
First rule of internet comments, anyone that says “end of” has not said anything worth reading.
Wow
Astounding that you seem to care more about words than the real fact a Black barrister could earn less than a White barrister for exactly the same work.
Anon
The research did control for university attended though. The relationship between results and ethnicity was in addition to the relationships seen between university attended and results, and degree classification and results.
Poppit
This is data that is likely to misused. The key is in the sentence identifying socio-economic status as a factor. The likely difference in outcomes is mainly explained by socio-economic factors not ethnic background and the research ought to look at the relative performance of candidates by socio-economic background and ideally socio-economic background and ethnicity. Presenting the data only as a basic ethnicity-based tick box report will drive ethnicity based approaches in regulation and positive discrimination which harms the prospects of those from lower socio-economic backgrounds whatever their ethnicity. Good times to be from a wealthy family and from a minority ethnic group, good times.
Anon
Socio-economic factors are a key to explaining the differences, sure. But I would say culture (for which ethnicity is, while not the same, an effective proxy) is also a key factor.
If you look at school results based on ethnicity and eligibility for free school lunch (the closest proxy for socio-economic status which is publicly available, albeit admittedly still a crude one), you can see that ethnicity (or culture) is strongly correlated with certain results. For example, Chinese students eligible for free school lunches perform better on average than any other ethnicity, whether or not eligible for school lunches. Black Carribean students perform worse in school even when controlled for school lunch eligibilty.
Marking everything down to socio-economic factors and ignoring the strong effects of culture just isn’t accurate.
https://www.ethnicity-facts-figures.service.gov.uk/education-skills-and-training/11-to-16-years-old/gcse-results-attainment-8-for-children-aged-14-to-16-key-stage-4/latest#by-ethnicity-and-eligibility-for-free-school-meals
Anon
The data set you linked to shows both the biggest gap based on socio-economic factors was among white students and also that disadvantaged white students performed much worse than both Asian students, disadvantaged and not disadvantaged and black students, disadvantaged and not disadvantaged. Other than white disadvantaged students, the relative performance by ethnicity data were all broadly similar by socio-economic background, with the exception of the higher performance of Chinese students. The data you refer to indicate that rather than focusing on ethnicity focus should be on socio-economic issues as that is where the clear discriminatory processes are embedded in the educational system.
Anon
The reply’s the first sentence agreed that socio-economic factors are indeed one of the key factors. It’s agreed that socio-economic factors are where the focus should be on.
I think the point is that there are statistically meaningful differences which can’t be explained away by socio-economic factors, such as the high performance of Chinese students or the low performance of Black Carribean students.
The results aren’t ‘broadly similar by socio-economic background’ – there are some discrepancies which look like they’d be a couple of standard deviations apart.
Anon
The statistics in question show the impact of socioeconomic factors are much greater than ethnicity. So while there appears to be ethnicity effects too, the data would indicate that focusing on socioeconomic factors is more pressing, particularly in a regulatory world where the focus is almost exclusively on gender and ethnicity, to the virtually exclusion of socio-economic issues.
Anonymous
You do understand that you can be poor and from an ethnic minority group too? Race and socioeconomic factors are closely linked.
If you are a member of minority who grew up on a council estate, you may be perceived by some pupillage committees as being less able to join in the Stewards Enclosure/skiing/Club Suite/interior design/WSET/London restaurant/Mulberry bag/Royal Opera House/private maternity nurse/choosing the right boarding school/Pall Mall club/pedigree dog/country house conversations that almost ALL barristers have.
The focus should be on decision-makers’ snobbery and prejudice towards candidates, remembering that ethnic minorities are certainly affected by this too.
Anonymous
This is absolute nonsense, a barrister’s chambers does not take in a pupil for small talk – they are able to see smart, hard-working people, and are not here to mess around.
How much more evidence do you need?
Black and Asian barristers earn less than their White counterparts and are FOUR times more likely to experience bullying.
Report after report (like this one), year after year say exactly the same thing –
https://www.barcouncil.org.uk/uploads/assets/d821c952-ec38-41b2-a41ebeea362b28e5/Race-at-the-Bar-Report-2021.pdf
They aren’t making it up. They aren’t being ‘over sensitive’ or ‘imagining’ things. It is systematic abuse and mistreatment.
Asian lawyer
As an Asian lawyer working at a MC firm I can say I’m frequently subjected to borderline racist remarks by my white colleagues (like calling me names, making fun of my cultural background, etc.). I have to say the issue is often neglected by senior practitioners and it’s severity and impact on the rules of law must not be understated.
I do hope things would change one day though. Lots of bullies out there, but rarely reported by the media.
Anon
The earnings data can be explained almost completely by socio-economic factors. A report that does not take socioeconomic explanations into account is useless and this report accepts that it did not take this into account.
The report states:
“Anecdotal experience suggests socio-economic background is also likely to be a factor, although we do not at present have the data to categorically state this.”;
“The intersection with sex and educational background (as an approximate proxy for socio-economic background) is significant.”; and
“The impact of intersectionality should be properly analysed. Any barriers to data collection and analysis should be identified and removed.”
So all the citation does is show there is some evidence of difference of treatment but there is no data to show ethnicity is a significant part of that difference.
The rest of the “evidence” came from self-reporting roundtables, which means it is not evidence of any credible weight. If you ask people to join a roundtable on bullying or discrimination you will end up with participants who want to report bullying or discrimination.
WTF?
How exactly do you propose to get evidence if people don’t open their mouths and ‘self report’???
This is literally how courts and the police work – people speaking in front of others and saying what they saw.
Does it really matter whether minorities write about their own experiences, sit for an interview or have someone else publish what they say if they all say broadly the same thing about having bad experiences within a professional setting?
Is every single actor, footballer, medic, executive, teacher or waitress who ‘self-reports’ racism not to be trusted???
Anonymous
You missed the point. These reports always provide statistics of rates of issues which are utter tripe and distorted by the self-selection factor. Those commissioning the surveys know that a headline press release “67% of [black]/[female]/[insert cause] lawyers report [X]” that this figure is embedded in onward reporting and then churned out without question in later articles to advocate regulatory changes. I’ve seen it first hand when third party commercial organisations offer to run survey processes and they all but expressly guarantee a favourable outcome in advance by the selected process. It is not accuracy of the individual self-reporting that is the biggest problem, it is the alleged rates of reported behaviours. Saying that, there is another problem which is most of these processes allow those reporting to self-define conduct subject to a complaint so there is a lack of consistency there too.
P.N.S. Phagharty
We need to break this down further to get to the root of it.
Applicants from certain ethnic minorities attain higher performance than white British students eg Indian and Chinese.
There must be something else to consider other than white privilege, surely?
Comments are closed.