Which schools did you turn down for Berkeley?

sakky · May 3, 2011, 10:27pm

Yet I would say that within those statements, you’ve embedded your own counterargument. Berkeley students are not ‘normal’ students with mediocre preparation by any stretch of the imagination. The vast majority of Californians cannot even apply to Berkeley because they’re not even UC-eligible, and even of those that can and do apply, the vast majority (~75%) are rejected. In fact, the entire purpose of the University of California system is not to educate most Californians, and never has been. Only the top 12.5% of California students are even UC-eligible, and even of those that are, most of them won’t get into Berkeley.

Hence, since Berkeley is already largely ‘closed’ to the overwhelming majority of Californians anyway, is it even really so outrageous to increase the selectivity further to match Stanford? You say that such a policy would result in many prospective students being turned away and forced to attend a lesser school, but that already happens now. Frankly, I don’t see that much of a difference between a 25% admit rate and a 20% admit rate. Either way, you’re rejecting the overwhelming majority of your applicants.

The only entity within the California higher education system that can truly be said to be serving all/most Californians are the open-admission community colleges. Even the CalStates, by design, only serve the minority of state residents who are eligible.

phantasmagoric · May 4, 2011, 1:54am

I think increasing Berkeley’s selectivity would harm its dominance in social mobility. It provides the highest number of low-income students with the best opportunities a school can offer (UCLA has a comparable low-income student population but doesn’t have the opportunities that Berkeley does).

I think it’s largely because of the creation of Silicon Valley (alumni had stepped up during the 1930s). That’s what put Stanford at the top in the decades to come.

But saying that Stanford was ‘in Berkeley’s shadow’ and ‘nothing more than a good regional private’ before that is an overstatement. For example, “the first Stanford student” Herbert Hoover became president in 1928, which boosted Stanford’s prestige. Back then, there wasn’t as much comparison among schools, and the differences that exist now were much smaller back then. Stanford also started out with some $400-500 million (adjusting for inflation) and had set out to replicate Harvard, which it had done by that time–again, the quality of schools was not so defined, nor did schools like Harvard enjoy huge advantages. Back then, Harvard and Berkeley would have been equal.

m3csl1994 · May 4, 2011, 2:09am

^Phanta, We’ll have to agree to disagree here. Stanford was indeed a good regional school until after World War II. It was after Stanford got some large federal grants for defense related research, and the creation of the Stanford Research Park (and the leadership of Fred Terman) that things started to roll for Stanford. Until 1951, the admit rate at Stanford was 91%. Look it up.

sakky · May 4, 2011, 2:24am

I’m not sure that’s true, and in fact, I might even argue the opposite: Berkeley may actually decrease social mobility, relative to the other available choices of those low-income students. That’s because Berkeley has a rather nasty habit of taking those students and then flunking them out and thereby marring their academic records for life. In this day and age, in order to garner most decent jobs, employers expect you to have a degree. It’s therefore better to graduate from a CalState or a lower UC than to flunk out of Berkeley.

Heck, those students would have been better off not having gone to any college at all rather than go to Berkeley (and flunk out) because at least then their academic records would still be pristine, for then they could still apply to many top schools through freshman-admissions. With their academic record trashed, they will find it unnecessarily difficult to be admitted (as a transfer) to any decent college. Every college adcom demands to know whether you had previously enrolled at another college, and if so, what happened there. Far better that your answer be that you had never enrolled in college at all, rather than that you had but flunked out.

Even those students who graduated - but barely so - from Berkeley would probably have been better off elsewhere. Let’s be honest, if you graduated with less than a 2.5 GPA, you probably did not enjoy a particularly fulfilling experience. You’re probably not going to be competitive for the better jobs nor for any decent graduate schools. You probably would have been better off at a school that was more suited to your talents.

Certainly I agree that some low-income students perform very well at Berkeley and therefore maximize their social mobility. But others perform poorly, and when you calculate the net summation of all gains and losses in social mobility, it’s not clear that you would obtain an overall positive figure.

m3csl1994 · May 4, 2011, 2:40am

Sakky, I think I basically agree with you here.

Singh2010, I see some of your points. However, we have some clear differences of opinion.

First of all, UC in Kerr’s Master Plan was intended to educate the top 1/8 of the state’s population. Remember, this is when the state was MUCH smaller. There’s no hard and fast rule about how many students Berkeley has to take.

So let’s look at the numbers. Should Berkeley be more “democratic” and educate more students than Stanford? Sure, why not. So Stanford has 7,000 undergrads and Berkeley has 25,000. That’s a massive gulf. One thing this state needs more than anything else is top notch, somewhat personal college education - Stanford alone can’t and shouldn’t be the only choice there. But with 3.5X higher enrollment than Stanford, many prospective students and their parents in California are deterred, and end up leaving the state for other private schools.

You make it seem like Berkeley is a school of second choice. I don’t think it should be. And I think it gets plenty of top-notch kids (who says the Stanford or any other ad com is perfect? Plenty of kids with awesome high school records burn out; while there are many others who get their second wind only in college). However, it could easily stand to be 30% smaller than it is today. The top 1/3 at Berkeley is pretty damn awesome, the middle 1/3 is very good, and the bottom 1/3 is often questionable. If you dropped that bottom 1/3, how much better would things be? Think about it - undergrad enrollment would still stand at 17 or 18,000 kids - much more than Stanford, still, but a little more manageable. A more community oriented student body, and with better stats (higher US News ranking), and I’d argue with a greater sense of loyalty to the school - and with better per capita results in various placements (jobs, grad and prof schools, prestigious scholarships, etc). That’s what Berkeley desperately needs. It does it at the grad level. Why not bring some of that selectivity to the undergraduate ranks?

There needs to be a balance between providing a very good education for more students, but also not diluting quality too much to the point where no one benefits: not the top students, and not the weakest ones. While I agree the idea of a public university is a noble one, 1) no one says that public can’t be elite (look at IIT in India or the University of Tokyo in Japan) 2) Berkeley is being actively divested by the state of California. It should look after itself now, and the best private universities should provide a directional template.

In the best case, Berkeley can be a hybrid public/private that takes the best of both worlds: providing an education second to none for a slightly larger student body, but also not killing itself by spreading resources too thin and graduating class after class of disgruntled alumni.

ucbalumnus · May 4, 2011, 3:12am

But how easy is it to determine the bottom 1/3 from information available to admissions committees? As schools get more selective at the top end, the differences in academic qualifications that are visible to admissions committees become minute to the point that a single seemingly random aspect can swing the decision for some applicant.

Berkeley may not yet be at that point, but it may be getting close. About 57% of freshman admits have a high school GPA (weighted) of 4.20 or higher, and another 30% have a high school GPA of 4.00 to 4.19.

The 4.20+ high school GPA group has average Berkeley GPAs of about 0.1-0.2 point higher than the 4.00-4.19 high school GPA group. They also have a slightly better retention rate.

Students who entered as freshmen in 2003 (most recent for which graduation data is listed, probably due to use of 6 year graduation rates):
4.00-4.19 high school GPA: 92.2% graduated, average GPA 3.33
4.20+ high school GPA: 94.4% graduated, average GPA 3.48

Freshmen who entered in 2008:
4.00-4.19 high school GPA: 97.0% retained, average GPA 3.20
4.20+ high school GPA: 97.6% retained, average GPA 3.39

Source: [University</a> of California: StatFinder](<a href=“http://statfinder.ucop.edu%5DUniversity”>http://statfinder.ucop.edu) custom tables.

sakky · May 4, 2011, 11:00am

I think it would be elementary, given that Berkeley is sitting on a comprehensive registrar database of the past performances of thousands of past students. Berkeley also has a world class statistics department. Surely the school could marshal a few of the top statistics professors to devise a statistical predictive model that would analyze the admissions records of all of those past students to collate which attributes seem to be highly correlated with poor performance (i.e. flunking out of Berkeley). Berkeley should then simply admit fewer future applicants possessing those attributes.

Outrageous and unfair, you might argue? Well, that’s what happens now in the insurance industry. Auto insurance firms provide policies at different rates - or may choose not to even offer you insurance at all - depending on where you live, what type of car you drive (i.e. a sportscar vs. a minivan), your driving record, your demographics, and a host of other data because actuarial tables have determined from past data that certain people are likely to be more costly to insure than others. Similarly, homeowners insurance is more expensive in locations where there are many burglaries.

Keep in mind that nobody has a ‘right’ to auto insurance. If you’re a 16-year-old male with a poor driving record who lives in a congested neighborhood and driving a Corvette, you probably won’t find anybody to insure you, and if you do, you will be charged nosebleed premiums. Similarly, nobody has the ‘right’ to attend Berkeley. The top 12.5% of Californians might have the ‘right’ to attend the University of California according to the Master Plan, but they don’t have a right to be admitted to any particular campus within that system. Students are perfectly free to attend Riverside or Merced, and indeed, the whole point of why the latter was built in the first place was to catch the overspill. Like I said, Berkeley already rejects 75% of its applicants. What’s the difference between that and rejecting 80-85% of the applicants? Either way, the vast majority of applicants are prevented from attending.

I think the easiest and least controversial reform is to simply determine what attributes seem to be highly predictive regarding Berkeley students who actually flunk out, and then simply admit far fewer of those students in the future. What purpose is served by admitting those students, only to later flunk them out and tarnish their academic records? Berkeley loses, those students lose…everybody loses. Everybody would have been better off if they had never been admitted at all.

Would anybody like to make the counterargument that Berkeley should continue to admit students who will flunk out?

bluebayou · May 4, 2011, 11:10am

Huh? Such as?

I concur with Sakky.

There is no reason necessarily to shrink Cal’s size. Instead of social-engineering the bottom quartile, perhaps Cal should just start admitting the best and brightest nation-wide and raise the standards. So what if Merced loses a few of its Regents Scholars?

ucbalumnus · May 4, 2011, 11:44am

If those attributes are visible on the application. Not all such attributes, like poor ability to handle the less structured university life versus more structured high school life, are going to be visible on the application.

What does this have to do with anything?

No one is (in public), and the non-graduation rate is not particularly high other than the very low end of admits (good luck trying to eliminate the revenue producing athletic admits, etc. that have strong political backing).

cansei · May 4, 2011, 1:29pm

At my school 3 people turned down Berkeley for USC and they are going into business…I lold.

UCBChemEGrad · May 4, 2011, 1:34pm

^ USC Marshall is the safer bet. But, Haas > Marshall.

ucbalumnus · May 4, 2011, 1:47pm

For what it is worth, graduation rates (@N refers to graduation after N years):

Berkeley:

Entering 2003: 66.7%@4, 88.0%@5, 90.8%@6
Entering 2004: 69.3%@4, 89.3%@5
Entering 2005: 71.4%@4

UCLA:

Entering 2003: 64.6%@4, 86.6%@5, 89.4%@6
Entering 2004: 67.3%@4, 87.3%@5
Entering 2005: 66.9%@4

For AAU members, including many top public universities, here is a handy chart for 6 year graduation rates for the entering 2002 class: <a href=“http://www.ir.ufl.edu/nat_rankings/students/gradrate.pdf[/url]”>http://www.ir.ufl.edu/nat_rankings/students/gradrate.pdf</a> . Note that Berkeley and UCLA are #2 and #3 in this chart.

MIT:

Entering 2003: 91%@6

Stanford:

Entering 2003: 92.1%@5
Entering 2004: 92.2%@5
Entering 2005: 92.0%@5

Yale:

Entering 2003: 98%@6
Entering 2004: 96%@6

USC:

Entering 2003: 88%@6
Entering 2004: 89%@6

Many more universities can be found here: [Sortable</a> Table: Graduation Rates Over Time, by Institution - Students - The Chronicle of Higher Education](<a href=“Sortable Table: Graduation Rates Over Time, by Institution”>Sortable Table: Graduation Rates Over Time, by Institution)

ballpointpen · May 4, 2011, 1:48pm

I turned down Cornell, Northwestern, USC, and Michigan for Berkeley (intending to major in EECS). i’m OOS and reading this thread, it sounds like you guys would have chosen differently if you were in my situation haha. I would have paid roughly the same in terms of tuition for each of those schools, probably more for Berkeley as it seems like the tuition will go up dramatically in the next five years. what do you guys think of my decision?

sakky · May 4, 2011, 2:40pm

So then you simply find correlated proxies for the unobserved variables. This is a basic and well-established process that statisticians deal with every day.

For example, maybe drivers with red cars are more likely to have accidents compared to drivers with white cars. Maybe drivers with (unobserved) riskier driving profiles tend to prefer red cars. Or maybe red cars are simply more difficult for other drivers to see and therefore avoid. Or maybe red paint makes cars inherently unstable. From the perspective of the insurance company, who knows and who cares? For the purposes of calculating insurance premium, all that matters is that, for whatever reason, red cars seem to be more dangerous than are white cars.

I’m preempting any counterargument that some might propose - and have in the past - that to reduce admissions is outrageous and unfair because it ‘constrains opportunity’. Berkeley already constrains opportunity by rejecting the vast majority of its applicants. I don’t think an 80-85% rejection rate is much more outrageous or unfair than the current 75% reject rate.

The only higher educational institution who can truly be said to not be constraining opportunity are the open-admission community colleges, because they truly will take everybody. Only a small fraction of state residents are even eligible to study at any campus of the University of California whatsoever, much less at Berkeley.

How many revenue-producing athletic admits are there? Football and men’s basketball are the only two revenue producing sports, and they only admit maybe 25-35 recruited athletes every year. Even assuming all of them flunk out (which they obviously do not), there are still far more than 25-35 students who flunk out every year.

UCBChemEGrad · May 4, 2011, 2:40pm

Ballpoint, to put your mind at ease a bit, I doubt OOS tuition will be increased dramatically. It is already very high and in order to be competitive with other high cost privates, Berkeley has less room to increase the OOS tuition. Besides, they’re looking to attract more OOS students and won’t do this by increasing tuition.

In-state tuition has been a bargain. Unfortunately, given the budget cuts, in-state tuition is more likely to increase.

You made a great choice. Berkeley EECS is one of the best in the world. EECS members on this site (and their parents) have not complained about Cal… Berkeley offers by far the strongest program in EECS out of your choices. Proximity to Silicon Valley is also helpful for your future career.

ucbalumnus · May 4, 2011, 3:02pm

That presumes that there is a “smoking gun” type of observable characteristic among applicants* that is both a good predictor of flunking out and can reduce the size of the admissions class by the amount that people are referring to (e.g. 1/3).

Berkeley’s non-graduation rate is not enormously high (about the same as the more selective MIT, lower than all of the top publics other than Virginia, and lower than USC).

Also not all non-graduations are due to flunking out – some depart for financial or medical reasons, or transfer out, or to start a job (e.g. athletes turning pro, “college dropouts starting successful companies” (who would go back to school if their companies failed), etc.).

The whole premise that “Berkeley is letting lots of students in so that they can flunk out in large numbers” is flawed.

sakky · May 4, 2011, 3:25pm

There doesn’t need to be a smoking gun. It can (and likely will) be a collection of observable variables that combined will indicate that somebody is likely to flunk out.

In fact, Berkeley - and the entire UC/CSU system - does this now by incorporating high school GPA and test scores into the admissions process. Presumably those variables are correlated with your predictive success in college, otherwise why use them at all? Why doesn’t Berkeley just admit everybody with 2.0’s and terrible test scores?

With enough data - and Berkeley clearly has the records and admissions profiles of thousands upon thousands of prior students as observations - you can construct highly precise predictive models.

MIT (and all other tech schools) are obviously hampered by the harsher grading schemes within technical majors. That is why any predictive model should be stratified by major, or at least by college (i.e. L&S vs. CoE), but that’s trivial to perform.

And we’re not talking about them. We’re talking about students who have performed poorly, as defined below.

If the numbers are truly small, then there should be no controversy about eliminating them, right? After all, we wouldn’t be talking about many students anyway.

On the other hand, I have already argued that anybody who is graduating with less than a 2.5 is performing poorly and would have been better off elsewhere. If your grades are that poor, then clearly Berkeley was not the best place for you.

Would anybody like to make the counterargument that Berkeley should continue to admit students who end up with GPA’s or less than 2.5? Or should those students have gone to another school?

crowslayer91 · May 4, 2011, 5:06pm

^well, someone has to get the C’s for curves to make sense.

ucbalumnus · May 4, 2011, 6:25pm

Yes, they do use high school grades and test scores. In other words, they are already incorporating the observable characteristics that are what most people believe have the highest predictive value in whether a student will be successful at the university level. But as selectivity increases, the distinctiveness of the students becomes less – if you have to select 1,000 admits out of 2,000 students with the maximum possible GPA available at their high schools (based on straight A grades with all possible honors/AP courses), then you will see the limits of these characteristics that work very well at lower levels of selectivity.

Do you believe that there are other additional characteristics or combinations that are likely to be highly predictive (as opposed to very small changes, like 5% versus 7% non-graduation, meaning that even the “worse” group still has a 93% chance of graduation in this example)? If so, which ones are likely candidates? Or are you just speculating/assuming that there exists such a thing for the sake of argument, but leaving the determination of that to someone else (much like politicians claiming to want to balance the budget, but leaving the actual spending cuts and/or tax increases to someone else)?

Can either Berkeley, or the students themselves, predict with any degree of certainty what GPA they will get?

In addition, if some means of reliably removing the bottom third or fourth of the students (as in reliably predicting low GPAs at university) were found, then someone else would end up filling the bottom of the grading curves. But then if you advocate more grade inflation to eliminate low GPAs, that can be done without admissions changes.

m3csl1994 · May 4, 2011, 8:55pm

ucbalumnus, you sure are looking at admissions using a very one-dimensional approach. GPA isn’t everything; nor should it be.

Berkeley has a massive range in its 25%/75% SAT scores. The bottom feeders in terms of the SAT should be eliminated. Additionally, the subject SAT tests should be REQUIRED (these were proven in earlier UC studies to be more closely tied to college grade point). And the process should rely more heavily on teacher recommendations, which should also be a requirement.

With all of these in place, GPA notwithstanding, I think it would be very easy to whittle down class size by 1/3, and I think it would make Berkeley overall a much more attractive place for top students while still serving the state of California (I’d argue more efficiently).

Thread		Replies	Views
So you're going to Berkeley next year, but which schools did you turn down? University of California - Berkeley	119	237	April 17, 2021
Which schools did you REJECT to go to Cal? University of California - Berkeley uc-berkeley	49	164	April 16, 2021
What are you turning down for Princeton? Princeton University	17	74	April 17, 2021
Turning down Berkeley for... University of California - Berkeley	145	933	April 17, 2021
Which colleges did you turn down to go to UCSC? University of California - Santa Cruz	28	217	April 17, 2021

Which schools did you turn down for Berkeley?

POPULAR STATES

SEARCH SAT SCORES

SEARCH ACT SCORES

SEARCH GPA’S

Which schools did you turn down for Berkeley?

Related topics

POPULAR STATES

SEARCH SAT SCORES

SEARCH ACT SCORES

SEARCH GPA’S

CONNECT WITH US