The Student Room Group

Ucas to enforce 'name-blind' applications to tackle racial bias

Scroll to see replies

Original post by Observatory



Aptitude is a large factor, probably a very large factor, in selection for university places in this country, especially outside Oxbridge. If we suppose that aptitude is the only factor in selection then we do not have a perfect model of the real world but we do have a very good approximation that makes clear predictions that diverge from the racial prejudice hypothesis. This permits the two theories to be compared with data.

My suspicion - though I haven't done the work to prove it - is that the aptitude differences model will explain the data almost perfectly and that the racial prejudice model will fail everywhere its predictions diverge from those of the aptitude differences model.



I disagree. University admissions are based on attainment with a mild corrective for guesses of the extent to which attainment does not reflect aptitude.

Virtually all those guesses are one way; the perception that an applicant is brighter than his or her exam results suggest because either he or she went to a poor school or the fact that receipt of free school meals or living in a particular area indicates personal obstacles to success.

Universities rarely correct for exceptional quantity or quality of teaching (save the few courses that disregard resit marks) but then complain about the over-prepped student who is incapable of independent thought.

Of course the reference and the personal statement should be indicators of aptitude but universities (a) do not trust the honesty of referees who are, after all, salesmen for their charges; and (b) do not trust the integrity of personal statements.

My own view is that the best selection tool remains the interview. That is why it is the selection tool of choice in the open market. I think few interviewers are consciously prejudiced. I think Scotgrot's point is an important one. Prejudice, whether conscious or unconscious, is usually depersonalised.

I think the government is fairly confident about the name effect in university admissions because it has conducted impressive research on the name effect in CVs for employment. One of course may not correlate to the other.

I am not aware of any recent UK research on race and job interviewing.

Although slightly off topic, I have always worried about the discriminatory effect of non "ist" interviewing. If you calve out large parts, and perhaps the most interesting parts, of the lives of certain of your interviewees as forbidden territory, do you put them at a disadvantage?
I think this policy could be a good thing, if it helps to prevent unconscious bias, but then you have to question whether universities that what to interview candidates, should also have to do so in an anonymous manner, i.e. over the phone and without using names etc, because I doubt making just the initial part of the university application anonymous will have any significant improvements on the pre-existing biases that exist.
Reply 62
I didn't know university lecturers were racist.
Original post by PrincessZara
I think its a great idea. Why should we keep things as they are if they are discriminating people, don't you want to stop that ?


We do not know they are discriminating. You have just made Observatory's point for him. If you infer discrimination exists from steps that eliminate the possibility of discrimination, that is a reason for not taking those steps.

Original post by sdotd
But this wouldn't change anything at all. They will find other reasons to choose a candidate over another


They have to find reasons to choose one candidate over another. That is their job. The question is whether they should have access to information that could only be used illegitimately in that selection process?
Original post by YellowWallpaper
Not all unis conduct interviews for their courses.


Really?

I'm guessing they aren't very good unis then.
Original post by Josb
I didn't know university lecturers were racist.


Very few of them will have any overt racial prejudice. However, they, like you and I, will come with cultural baggage.
Original post by DiddyDec
Really?

I'm guessing they aren't very good unis then.


On the contrary, except for Oxbridge, medicine, dentistry, vet med, nursing, midwifery and teaching only a tiny proportion of students are interviewed.
Reply 67
Original post by nulli tertius
Very few of them will have any overt racial prejudice. However, they, like you and I, will come with cultural baggage.


Most lecturers would consider positively applicants from poorer backgrounds. By making applications blind, they will only judge on grades and PS, for which white rich kids do better.

This is the perfect example of a stupid law implemented under the pressure of vocal minority groups that does more harm than good.
Original post by nulli tertius
On the contrary, except for Oxbridge, medicine, dentistry, vet med, nursing, midwifery and teaching only a tiny proportion of students are interviewed.


I got interviewed at every uni I applied for to do surveying. I assumed it would be standard practice.

Posted from TSR Mobile
Original post by Josb
Most lecturers would consider positively applicants from poorer backgrounds. By making applications blind, they will only judge on grades and PS, for which white rich kids do better.

This is the perfect example of a stupid law implemented under the pressure of vocal minority groups that does more harm than good.


They are not judging on poorer backgrounds though. If they wish to judge on poorer backgrounds, universities buy data and flag applicants living in poor areas and the UCAS form identifies applicants who are or have been on free school meals. If that is insufficient, UCAS can collect more parental financial data if it wishes.

What you are suggesting is that academics should select on their prejudices regarding the wealth of the holders of particular names.
this is great news for my friends Lashonda, Lafonda, Tiffany, Porscha, Mercedes, Rudy, Trevor, Ashley, Britney, Chloe, Jamal, and anyone with the names of footballers/their kids
Original post by selfteaching
this is great news for my friends Lashonda, Lafonda, Tiffany, Porscha, Mercedes, Rudy, Trevor, Ashley, Britney, Chloe, Jamal, and anyone with the names of footballers/their kids


With the exception of Chloe (7745) these really weren't popular names in 1997

No Lashondas, 24 Portias, 451 Tiffanys, 47 Mercedes, 358 Brittanys (plus about another 20 who couldn't spell) 200 female Ashleys (but there were 857
Ashleighs plus about another 100 with spelling variants). Only Chloe and Ashleigh were in the top 100 names in England and Wales



9 Rudys, 33 Trevors 1660 male Ashleys (41 very confused male Ashleighs) and 116 Jamals. Again only Ashley made the top 100

I think you were thinking of the 3640 Shannons or the 2812 Courtneys. There were 5631 Ryans and 4890 Connors.
Original post by nulli tertius
I disagree. University admissions are based on attainment with a mild corrective for guesses of the extent to which attainment does not reflect aptitude.
I am not sure you do disagree. You seem in fact to be agreeing that universities are attempting to measure aptitude and to do this they take something that is strongly correlated with aptitude, attainment, and then try to improve their guess by further corrections. I agree that this probably doesn't result in an exact correlation between the students universities select and those applicants with the highest aptitude, but it will result in a high correlation between the two.

Virtually all those guesses are one way; the perception that an applicant is brighter than his or her exam results suggest because either he or she went to a poor school or the fact that receipt of free school meals or living in a particular area indicates personal obstacles to success.

Universities rarely correct for exceptional quantity or quality of teaching (save the few courses that disregard resit marks) but then complain about the over-prepped student who is incapable of independent thought.

This is a fair point but I do not think it is important because university admissions are effectively norm-referenced. It does not matter if you meet an absolute standard, it matters if you are better than the other candidates. Anchoring the norm to the performance of the best prepared students does not necessarily give the best prepared students any advantage in admissions, beyond their any pre-existing aptitude advantage.

Of course the reference and the personal statement should be indicators of aptitude but universities (a) do not trust the honesty of referees who are, after all, salesmen for their charges; and (b) do not trust the integrity of personal statements.

My own view is that the best selection tool remains the interview. That is why it is the selection tool of choice in the open market. I think few interviewers are consciously prejudiced. I think Scotgrot's point is an important one. Prejudice, whether conscious or unconscious, is usually depersonalised.

You might be right that the interview is a better tool but it is not the tool that is used. The fact is that the vast majority of university places in this country are awarded primarily on the basis of A level results. The government is commenting on alleged problems with the admissions system that exists, not problems with a hypothetical alternative system.

I think the government is fairly confident about the name effect in university admissions because it has conducted impressive research on the name effect in CVs for employment. One of course may not correlate to the other.

I am not aware of any recent UK research on race and job interviewing.

I have heard of this research. Has it been corrected for pre-existing aptitude? For instance what I have heard is that CVs with muslim names receive fewer call-backs, but most muslims in the UK come from low IQ groups. They do not make a sweeping claim about all racial groups and all non-British names. Do employers underselect people with Chinese names? Do they overselect people with Chinese names? What plausible racial prejudice theory could explain that?
(edited 8 years ago)
Original post by nulli tertius
With the exception of Chloe (7745) these really weren't popular names in 1997

No Lashondas, 24 Portias, 451 Tiffanys, 47 Mercedes, 358 Brittanys (plus about another 20 who couldn't spell) 200 female Ashleys (but there were 857
Ashleighs plus about another 100 with spelling variants). Only Chloe and Ashleigh were in the top 100 names in England and Wales



9 Rudys, 33 Trevors 1660 male Ashleys (41 very confused male Ashleighs) and 116 Jamals. Again only Ashley made the top 100

I think you were thinking of the 3640 Shannons or the 2812 Courtneys. There were 5631 Ryans and 4890 Connors.


haha, i was actually gonna include ryan and connor, but i didn't want to offend anyone on here that has those names,

well played though : D
Original post by Observatory



I have heard of this research. Has it been corrected for pre-existing aptitude? For instance what I have heard is that CVs with muslim names receive fewer call-backs, but most muslims in the UK come from low IQ groups. They do not make a sweeping claim about all racial groups and all non-British names. Do employers underselect people with Chinese names? Do they overselect people with Chinese names? What plausible racial prejudice theory could explain that?


It was done with fake CVs and was statistically valid. Interestingly there was a greater level of discrimination (though reported as not statistically significant) against the Chinese than those with South Asian Islamic names.

www.natcen.ac.uk/media/20541/test-for-racial-discrimination.pdf

Similar research in France has found the same result.
Reply 75
Good. This will hopefully help reduce discrimination (both positive and negative).
Original post by nulli tertius
I think you were thinking of the 3640 Shannons or the 2812 Courtneys. There were 5631 Ryans and 4890 Connors.


Are they egregiously unclassy names? Real shame, I like all of them :frown:
Original post by Observatory
I am not sure you do disagree. You seem in fact to be agreeing that universities are attempting to measure aptitude and to do this they take something that is strongly correlated with aptitude, attainment, and then try to improve their guess by further corrections. I agree that this probably doesn't result in an exact correlation between the students universities select and those applicants with the highest aptitude, but it will result in a high correlation between the two.


No, I think they are for the most part measuring attainment. Admission decisions generally boil down to the strength of the weakest subject rather than aptitudes for the subject intended to be studied.



This is a fair point but I do not think it is important because university admissions are effectively norm-referenced. It does not matter if you meet an absolute standard, it matters if you are better than the other candidates. Anchoring the norm to the performance of the best prepared students does not necessarily give the best prepared students any advantage in admissions, beyond their any pre-existing aptitude advantage.


That doesn't work. You have three candidates, X, Y and Z. They do an aptitude test that the university never sees. Candidate X comes top. Candidate Y comes next. Unfortunately Candidate Z's results are corrupted.

All three sit terminal exams with a view to going to university. Candidates X and Y get offers of A. Candidate Z is offered a B. There are two places and the right number of offers is made on past predictions to fill two places.

Candidate Y has been brilliantly prepared for his exams and he gets an A. Candidate X has been less well prepared and gets a B. Candidate Z goes to Bash Street Comp and has an out of work druggie for his single parent. He gets a B

Candidates Y and Z meet their offers and they go to university. Candidate X builds mud huts in Africa for a year,

However on the basis of aptitude, the wrong decision has been made.

We don't know how bright Z really is. The university guesses that he is one grade brighter than the results he will get.

If the university's guess is correct, the right result is that X and Z go to university. If the university is wrong and Z is less than one grade brighter than than his exam result; X and Y should go university. Under no circumstances should the pair who go to university be Y and Z but that is what happens if you merely norm reference to the best prepared candidate.
Original post by scrotgrot
Are they egregiously unclassy names? Real shame, I like all of them :frown:


My prejudice (and I am saying that university admissions should not be able to indulge their prejudices) is that they are.

They are doing what names do and they are going out of fashion. Last year there were 989 Ryans and 841 Connors.

There were only 77 Shannons and 109 Courtneys

The girl's name that really needs saving is Tracy. There were only 7 last year.

The Marchioness of Worcester needs to do more press features

http://www.dailymail.co.uk/home/you/article-2601719/This-little-piggy-went-Marchioness.html
Original post by sdotd
But this wouldn't change anything at all. They will find other reasons to choose a candidate over another


So we should let this slip without even trying.

Quick Reply

Latest