Carnegie Mellon Admission Levels Field by Eliminating Demonstrated Interest
admission.enrollment.cmu.edu
admission.enrollment.cmu.edu
The argument advanced seems to be that eliminating demonstrated interest makes the admissions process more equitable because wealthy parents will push their children into the activities CMU is looking for and will know how to demonstrate interest. The counter to this argument is that wealthy parents will always be able to game an adversarial process. Parents will just push their children to study for the SAT more and pay more for SAT test prep. Parents will just push their children to work more on grades and classwork instead of things the children actually want to do like robotics or programming.
Basically, I'm reading this as a diversity play, not a rich/poor play. Whether it will actually work or not is another question that's hard to answer without seeing their data. I can pretty much guarantee that on the male/female side, though, demonstrated interest massively favors male applicants. Female students just aren't encouraged to do (or maybe don't have as much interest in) the kinds of things that impress admissions committees in that category, regardless of what they may end up wanting to study by the time they graduate.
Is there a factual basis for this reading?
They want diverse students, which includes poorer students, but also includes underrepresented demographics (looks like male/female split is 55/45, and hispanic/latino and black numbers are small).
Contextual recruitment (which they say they're moving to), is the current industry best practice in this regard of looking at the issue across most of these factors, so not sure it's favouring one characteristic over another.
with the harvard asian personality thing, the nyc public schools shsat thing, and now this i wonder how is that people don't recognize all of this hand-wringing over affluence begetting affluence and "hacking" for what it is: band-aids on a broken system. we should decouple education (which is essential for being productive) from social selection. how? i don't know because all of this is so highly coupled that you'd need a phd in market design and non-linear optimization to even start unwinding the relationships.
edit: i'm getting downvoted. when microsoft/google/uber/etc. speak from their ethos does anyone believe them? when fb last year released chat for children and claimed to be interested in helping kids or whatever did anyone believe them? what's the difference here? CMU doesn't even operate under the guise of a public service - it is a private institution after all.
It seems straightforward: Lower the cost of education significantly. The reasons you cite are the reasons for the public universities in the US, AFAIK. The number one predictor of college education, IIRC, is parental wealth or income; education wasn't always this expensive.
Either way, education funding is just something that the US population is demonstrably not interested in. Very few people push for more taxes to pay for education, which is what it would take, leading to things like teachers needing to strike just to get low-class income. And outside of specific demographic groups we do not have a culture that encourages academic study from a young age like many other countries do. These are big socioeconomic problems that are going to take a generation or more to change even a little bit.
[1] http://www.thefiscaltimes.com/Articles/2012/11/30/Shocking-C...
That's not supported by the facts. A candidate for U.S. President who advocated funding free college education for families with incomes under ~$100K received the most votes for President. A candidate who advocated free college education for everyone arguably was third most popular (though we have no official measure of that).
> http://www.thefiscaltimes.com/...
The parent's cite is about the relationship between future earnings and tuition; it's not about public funding.
I see where the mixup is with that citation. Tuition is a direct proxy for college funding (think of it as student funding rather than public funding). I meant it to show that increasing funding for colleges does not equate with improving outcomes for most students.
Democrats may retake Congress in the fall.
> Tuition is a direct proxy for college funding (think of it as student funding rather than public funding). I meant it to show that increasing funding for colleges does not equate with improving outcomes for most students.
I see what you mean. AFAIK, tuition increases have been closely correlated with public funding cuts, so overall funding may not be increasing.
I have wondered though: Students pay X times as much as they did 20 years ago; are they getting something better? Unfortunately, that's not how market pricing works; as the U.S. has tried to expose everything to market forces (including people's health!), higher education prices seem to be based on supply and demand, and demand for something so essential is highly inelastic.
I'm trying to find total funding numbers for public universities and am coming up short on nicely formatted nation-wide data. However, it looks like in CA, UC per-student public funding dropped from $23,000 to $8,000 from the late 70s to 2017 [1]. In the same time period, tuition went up from $630 (in 1976 dollars) to $12,630 last year. I can't find comprehensive information on the third big component of funding - alumni donations, but at UC Berkeley they have been growing significantly year over year for at least 7 years [2]. States are very different in how they fund schools, though. For example, Texas has a sovereign wealth fund that generates a huge amount of money. In Massachusetts, tuition growth outpaced per-student funding cuts over the last 15 years - look at chart 7 in [3].
Anecdotally, it seems like total funding is going up with tuition, but the money is going towards things like student annexes, stadiums, ancillary faculty salaries, etc. that generally have little to no impact on education outcomes.
[1 worth reading - some interesting data] http://www.ppic.org/publication/higher-education-funding-in-... [2] http://www.dailycal.org/2016/07/17/uc-berkeley-receives-reco... [3] http://massbudget.org/report_window.php?loc=higher-education...
https://www.insidehighered.com/news/2017/07/24/new-study-att...
> money is going towards things like student annexes, stadiums, ancillary faculty salaries, etc. that generally have little to no impact on education outcomes
I fear (I don't know) that's just another claim by the right to support funding and tax cuts. That is their goal in every situation, AFAICT.
Jumping through some of the arbitrary hoops and doing the busywork of high school turned out not to be among my passions, so if I'd been evaluated purely based on GPA I'd have probably been passed over in favor of someone better at following rules and perhaps less of a creative and independent thinker.
I ended up having quite a high QPA at CMU. I was the same person, but CMU and I turned out to be far better matched than high school and I.
I'm sad to hear that the admissions process no longer has an appreciation for these kinds of good matches.
During orientation there's a 1-2 hour ceremony where you're effectively welcomed in to the college- think like a graduation but the inverse of it (the exact phrase is slipping my mind right now), where the president speaks along with a few students.
The one point that stuck out to me from that was the then-president, Subra Suresh, mentioning that they hadn't admitted a single student who had gotten a 2400 on the SAT, though some had applied. It was implied that they weren't looking for test scores, but rather for people who had potential beyond checking statistical boxes.
My main question from reading this is "if your applicants are consistently getting 700+ SATs by category, and you don't want them to demonstrate the things they do in their field that demonstrates their passion, then how do you differentiate them?"
I had the same thought, but there may be more to it... they unfortunately left us speculating as to what findings about their admissions process led them to this change.
Trying on a slightly fringe hypothesis: engineering has surpassed finance as the perceived get-insanely-rich discipline. So the top engineering schools are now more vulnerable to those who will pretend their "heart is in the work" but really just want to get on track to becoming Lord Voldemort (take your pick of real-world unscrupulous CEO). I have nothing against financial success being a motivator. But as a university, or as a manager for that matter, I'd want to filter out anyone for whom it's the only motivator.
Put another way: a demonstrated interest requirement selects for intense drive/motivation of any kind. Maybe CMU is looking for a specific kind, e.g. intellectual curiosity.
As an aside, agreed there will always be admissions gaming. But if the game is going to be broken, changing the rules periodically at least prevents the hacks from becoming too commoditized.
what's the demo of kids that can commit to a private college before knowing what potential scholarship/aid packages are actually on the table?
"Demonstrated interest" is a term of art in college admissions that refers to "demonstrated interest in attending a particular school". Things like "I interviewed with a CMU alum", "I sent letters to the admissions department about how much I <3 CMU", etc. (All of these are in TFA.)
It doesn't refer to "demonstrated interest in STEM." TFA is not talking about ignoring demonstrated passion in STEM.
Like any elite school, CMU is very far-ranging and international w/r/t student body; it’s fair to assume that a disproportionate amount of students traveling to Pittsburgh to interview, tour, and attend pre-college are from above-average means. I get the logic. I also (anecdotally) know at least a dozen students at my time there who were wholly unqualified to be there save for their family’s ability’s to pay their tuition in full, cash.
I have zero interest in litigating the value of diversity; if that had to be explained here then you’re too ideologically gone for either of our time spent to be useful. What I can say - and I suspect CMU agrees - is that education and opportunity is a great way to give underserved populations upward mobility, therefore minimizing outputs of family affluence and giving more opportunities to lower-income applicants will yield a higher overall impact. I’m sure they likely saw performance (and potentially even donor) differences between the two populations; it’s possible that the demonstrative interest population simply didn’t perform as well.
There is an argument in favor of this. It lets the poor kids connect with the rich kid's network.
To generalize, what you are so honestly saying is something that often obstructs change: People with power are vested in the current system: They are the successful outcomes (the unsuccessful don't have power), so they believe in the system and instinctively don't want to undermine their self-image as deserving (i.e., that they didn't have an unfair advantage). They also don't want to undermine their own power, but that doesn't apply as much here.
> I have zero interest in litigating the value of diversity; if that had to be explained here then you’re too ideologically gone for either of our time spent to be useful.
The other side believes they can persuade people and have succeeded at an unexpected level. People and opinions change; look at views of women in college and the workforce; even your views on diversity (which I share) were probably unheard of a few generations ago, but people worked hard to change that. What will we bequeath on the next generation?
I understand this, but it is not a good policy. In effect you will now be discriminating against good kids who worked their asses off to do extra cirriculars and prep for the SATs, etc.
Get rid of legacy preference for sure, find new ways to be more inclusive of underrepresented populations, but don't fix discrimination with more discrimination.
They're not talking about extra-curriculars, which might demonstrate interest in STEM or whatever, but don't count as "demonstrated interest" in CMU.
Honest question: why should diversity (as they use it) ever be a goal? Immutable characteristics of a person should not be a consideration in any application to life unless the application is explicitly for that (tall people in basketball for instance). Meritocracy should be the paramount consideration in all things, and possibly a small consideration of diversity of ideas and beliefs, but not diversity of skin color or genitalia.
If all of the most brilliant astrophysicist minds in the world happen to be Japanese women, why shouldn't they be preferred over all others in getting into prestigious colleges to study astrophysics? Why do we care if we have a certain number of European males in that situation? The right answer is that we don't care how many European males enter that field (which is true today but for different reasons that I am not trying to get into).
Which is far more likely the case than some hunky-doodle strawman meant to justify continued bias.
Immutable traits should never be considered, only merit.
There's no mention in the CMU process of being 'passed along'. In fact, its all about them standing on their own merit, and not their alumni-parent nor influential uncle.
I'm so glad to hear 'your bias' is right in line with the new CMU process!
Given the amount of free things like Khan Academy, MIT Courseware, etc. the distribution of access to resources to perform well is becoming more distributed than ever.
A poor motivated kid with good encouraging parents and access to only free resources is on a more level playing field with kids from a rich family than at any point in history.
And even if you did get out of those traps and spend the time on free stuff, it's not even going to come close to getting you on par with a tutor who knows what they're doing spending 1 on 1 time with you and parents willing to spend thousands of dollars on electronics, flying lessons, or whatever other enrichment activities support you. And where's the free equivalent of your dad getting you a high school internship at his prestigious company for kids who need to work minimum wage service jobs to help pay the bills?
"A more level playing field" is not even close to the same thing as "A level playing field" today, and it won't be for a long time. That's why diversity programs are still important.
Your main contention is that a kid with access to tutors, expensive electronics and flight training is going to get a far superior education to someone who is working with their parents to direct their own study. Those things aren't as big of a differentiator as they used to be in one's ability to obtain an education. Much of the pedagogical value a tutor can provide in sequencing topics is easy to find online--often written by the very people who would make the best tutors. The ability to ask questions and get answers is also easily available. Is it nice having someone decide exactly what you need to work on and learn. Sure, but in terms of access to educational resources, not having a tutor isn't a particularly big differentiator.
> How much time have you spent with poor families?
I grew up below the poverty line, so a reasonable amount of time.
Maybe the world's best surgeons are actually within a traditionally marginalised group, but their skill and talent isn't developed because they're denied access to quality education because their previous levels of education were of lower quality?
The best way of finding that is to have access to proper education to develop that talent. Proper education is currently a myth. Education is unfair.
If the best surgeon is out there, but has had every barrier to educational success thrown in front of them (inner city school, poor parents who can't pay for tuition to improve your grades, who can't help you access support for university applications, where there's no expectation or ambition for success, where teachers have no investment in seeing you succeed, where you see another group repeatedly, systematically preferenced because they got the birth lottery to have access to all of the above), is that the right outcome for society? To preference one group just because it's always performed reasonably as a result almost solely of their privileges?
To be clear, it's not PC culture. Noones saying an underrepresented group has to be shoehorned in to make up quotas - but seeing the context of candidates, to put weight to performance that's not just proportional, but outweighted to their background (even if it means the overall performance at the previous educational level was lower, the markers of their ability to outperform their "position" mean that given access to the top tier of education, they'd outperform).
For hundreds of years and most of recorded history, non-whites and females have been seen as inherently less capable and intellectual. Only in the past 20-40 years has Western society (as a whole) made explicit, respectable attempts at giving minorities credit for their capacity to lead and to learn.
I understand your point, and I don't think your opinion is of malice: I used to have the same one. And it is tough, because unlike some other societal issues, college admissions is a zero-sum game. Therefore, no matter how noble the idea, it seems pretty shitty that a smart, motivated and passionate white man would get passed over at an elite university in order to allow an equally passionate, motivated person of color who isn't as smart (on paper).
Diversity is useful for its own sake, and giving historically suppressed groups of people more opportunity is noble. How can we do it equitably? Perhaps use the government to better higher education instead of trillion dollar tax cuts for the uber-wealthy. But that wades further into politics.
So, if you want the cream to rise... take it from the bottom.
---
The relationship between school type and academic performance at medical school: a national, multi-cohort study https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5589012/
What is the effect of secondary (high) schooling on subsequent medical school performance? A national, UK-based, cohort study. https://www.ncbi.nlm.nih.gov/pubmed/29792300
Why is that far more likely the case? Do you then also believe there are far less men in nursing because of a lack of opportunity rather than a lack of interest?
However, we should NEVER consider these characteristics. Instead we should considered the merit of the candidate. I don't care if my surgeon is an African refugee who escaped a crazed warlord and has an incredible story. I care that they are one of the best possible people in the world to be performing this surgery because not everyone is equipped to perform it and I don't want to die because they aren't able to handle this position.
This is exactly the reason. Right there. You just referenced it. And yet you're avoiding it. See why we have a problem right here?
So, I can see why a university, depending on which studies it's basing its decisions upon, would want to directly work on the diversity of its students, in order to improve the quality of the global education its giving (I don't think one need studies to be convinced that your education depend both on professors and fellow students, among other things).
Whether it's a good decision or not to increase diversity, I wouldn't judge rationally (even though my gut would say it's good, based on something like "mono genes crops die but wild crops live" things...), but it's certainly an impactful one !
>If all of the most brilliant astrophysicist minds in the world happen to be Japanese women, why shouldn't they be preferred over all others in getting into prestigious colleges to study astrophysics? Why do we care if we have a certain number of European males in that situation?
The fallacy with assumptions like these is that it presupposes a completely level starting point for all demographics/ethnicities.
If everyone had access to the exact same resources throughout the entirety of their developmental lifespan, then yes, no one would rightfully care whether or not $genderNationality was predisposed to be above-average at a given discipline.
That is just not how reality plays out though. More often than not, certain groups are predisposed to excel because they simply won the socioeconomic lottery. To draw on your hypothetical example, if Japanese women happen to dominate the field of astrophysics, is it because they're predisposed to it? Or is it because Ugandan men and Mongolian women simply don't have the same access to the educational systems and resources that give them a decided advantage in preparing for a rigorous, top-10 education in such a discipline?
I think a lot of times people look at efforts like this and try to oversimplify it to "oh look they just arbitrarily think they need more of a certain gender or ethnicity".
It's far more complex than that, and we need to make an honest effort to understand why some groups are underrepresented. I'm willing to be it has less to do with pre-disposition, and more to do with inherent inequality in an adversarial and unevenly distributed socioeconomic system.
Additionally, you're casting an individual who happens to have certain characteristics as a member of a group of people who share those characteristics (and more often than not it's a concentration on a very particular characteristic as opposed to the individual as a whole). "We need more black women in the student body" is not a valid reason to take on a specific black woman as a student. Of course, dismissing that woman based on those characteristics is also ridiculous.
Like you've said, this is a very complex topic that definitely requires further research. The pity is that operational regulations (and in a few cases even laws) are being changed before this research has had a chance to take place.
Wouldn't it be best to start at the root of the problem then, rather than merely patronizing?
If someone does not meet the high standards of an "elite" institution but are admitted anyway, aren't you setting them up to fail?
You're not going to do an extracurricular project when you have to work weekends to support your family. You're not going to the university library to read research material when you don't have access to a car. Joining that after-school robotics team is impossible when you have to watch your little sister because both your parents are working and can't afford child care. Yes, you could work even harder to get around these obstacles, but there's no recognition of "level of effort" in meritocracy, the only thing that matters is results.
What CMU is proposing won't fix everything (per nsnick above, "wealthy parents will always be able to game an adversarial process"), but it's a step in the right direction.
I tend to think that diversity should not be a "goal" except perhaps in the case of political representation. However, diversity is generally a good thing and if there is a lack of diversity in some field then this should be seen as a potential symptom of something being wrong. People just need to make sure that they correctly identify and fix the underlying problem rather than just "fix" the symptom.
This is socioeconomic factors, this is high school value-add, this is location, this is previous family members attending an HE institution, this is performance in the face of adversity.
Diversity isn't a box-ticking exercise to make the front of the brochure have different sorts of people on it, but recognising that there are a wealth of different barriers and paths to "equality of opportunity".
To boil it down to "only the best people who are in the position to be best at it should be able to do it" completely ignores the fact that there are groups of people (through circumstance, or through no fault of their own birth) who would never have access to the same opportunity to succeed. To not only accept that, but to make some allowances (and in HE, the amount we're talking about is exceptionally minor in the vast majority of cases) so that outweighted performance proportional to background (contextual) rather than absolute performance is the factor you're looking at to consider giving someone the opportunity, seems innately fair to me.
But how does that in any way go along with shunning "supplementary submission of materials, including resumes, research abstracts, writing samples, multimedia demonstrations of talents, and maker portfolios." How is work (most of which, like a writing sample or resume, requires no resources or affluence to create) less equal than academic records which are literally determined by what neighborhood one lives in? Essays aren't bad in theory, but it's only a matter of time before entire books are written by "admissions counselors" on exactly what influential people to name drop and what "shared values" with the school to highlight. Beyond a basic capability (prerequisites) and a lack of contrary evidence (ex. failing classes repeatedly), what could possibly be a better discriminator of student success then the interest they demonstrate through actual self-driven initiative?
How much "self-driven initiative" is actually self-driven though when it comes to a formal application process, and not influenced by someone in an advisory role? E.g. a parent who's been through the application process, or a careers advisor in a school (if you're lucky enough to have access to either)?
I think this is a step back from the best solution to the problem of finding a good college/student fit.
EDIT: Try not to downvote without at least giving a reason.
Much like the elite high schools here in New York, CMU’s entrance requirements have changed now that the wrong kind of people are the majority.
I don’t know how to feel about that.
The above indicates very high demand for Carnegie Mellon's educational services (along with other universities like CMU). Perhaps the obvious solution is to increase supply - if so many are so willing and apparently able to study at that level (the difference between the last admitted and first rejected is, AFAIK, almost undetectable), let's give them the opportunity!
Let's start with an open mind; I know many of the 'answers' already jumping to people's fingertips, but let's challenge those assumptions and the status quo. Nothing in the universe requires that there should be N spots available at CMU-level schools. Why not 2N or 10N or 100N? My impression is that the availability of quality college education boomed in the mid-20th century, especially in the U.S. Why not now, with such high demonstrated demand? At least public universities could be expanded.
It raises many questions: What do students and parents perceive is better about CMU than the next school down their list? Is their perception accurate? Can other schools be changed to match CMU in some or all regards?
I imagine (but don't want to assume) that the faculty at CMU are better than the next best school, but is that really so? Is there a material difference, other than prestige? And how important is that - all faculty (and most PhD students) at any school know infinitely more than almost any freshman, and few undergraduates take advantage of the faculty resources until later in college. Perhaps the solution is that admission to CMU-level schools and access to their faculty shouldn't happen until a student demonstrates the ability and interest in their first year or two.
like i wrote elsewhere in this thread the cognitive dissonance here comes from the idea that CMU is about educating. it's not. it's about generating profits. just like porshe/rolex/debeers could easily produce more product but doesn't in order to artificially deflate supply (in order to be able to give the impression of scarcity in order to be able to charge more) every elite institution does the same. for what it actually looks like when a school is honest about the material value of its degree just take a look at the online masters in CS from GT.