<p>Then they shouldn’t be admitted at those schools either. </p>
<p>
</p>
<p>Certainly I agree that such a system would not be perfect. But neither is the current system. The real question is, whether the proposed system better than the current system.</p>
<p>I would submit that it would be. Sure, I agree with you that the proposed system would be slow to react to changes. But compare that to the current system that seems to be non-reactive to changes. Which is worse? </p>
<p>
</p>
<p>You just (subtly) changed the terms of debate. I would submit that somebody who would have earned a 2.0-2.5 at Berkeley would indeed be far worse off than if he had earned a 3.5-4.0 at UCM (although not a 2.0-2.5 at UCM). Why? Simple - confidence. Let’s face it - much of success in life is due to confidence. How confident are you really going to be if you barely graduated from Berkeley? Probably not very. You were being threatened with failure at every step, so even when you do make it to graduation, you don’t exactly have the best mindset at hand - certainly not ready to ‘conquer the world’. You would have been better off attending a school that made you feel that you really could surmount every obstacle. </p>
<p>The other problem is that, however difficult the obstacles may be towards a Berkeley bachelor’s degree, people don’t really seem to care. This isn’t like a PhD. Employers don’t really seem to care about the difficulty of a Berkeley bachelor’s degree. If you perform poorly, nobody appreciates how difficult it may have beem. All they’ll see is that you earned mediocre grades as a college student.</p>
<p>Let’s also keep in mind that many (probably most) decent employers enact GPA cutoffs - usually around 3.0, and on occasion higher than that. If you have a GPA less than that cutoff, they won’t care why. All they will see is that your cutoff is lower than that cutoff, and then throw away your resume accordingly. What that means (sadly) is that, for those particular employers, somebody with a 3.1 from San Jose State is indeed better off than somebody with a 2.9 from Berkeley. Sad but true. The problems of Berkeley therefore fall disproportionately upon those students who earn poor grades.</p>
<p>Well, I think they would if they were more reflective of college work. The AP exams can be pitifully easy to do well on (e.g. the BC exam, where people who have far, far below a good understanding of calculus manage 5’s quite routinely). </p>
<p>
</p>
<p>Well right now, is there a proposed system? There only is a suggestion to become more selective, and other things, but no means to increasing selectivity has been proposed that I know of. We can all agree that different schools are selective for very different reasons. </p>
<p>I think we can all agree that it would be neither better not worse, but impossible to become more selective before we have a plan how to do so. I can tell you that it’s certainly not as simple as Berkeley being easy to get into – some very qualified people get rejected, but conjecturally probably for bad reasons (e.g. not inflating their GPAs to cater to the numbers-based admission).</p>
<p>Basically I want to know what “it” means. I mean, there are some ways of increasing selectivity I can think of that would not do anything to make underqualified students go elsewhere, and simply make the school a crapshoot. I don’t think we need to be anywhere as close to selective as some of the top privates to ensure things, because frankly what they do is make it near impossible to get in, with often little rhyme or reason why one person gets in over another known to anyone but a small admissions body. </p>
<p>I think Cal’s philosophy in admitting students is good, i.e. “if you seem pretty good, you have the chance to come,” but the standard for “pretty good” needs to be raised, not changed.</p>
That is a lie. It is much better to learn the pain of failure than go through life with the illusion of success. Of the students at Cal, some of the lowest GPA are those who were valedictorian at their high school or graduated 4.0 at their community college. Those students who were used to destroying the competition at much easier institutions aren’t prepared for the rigor and endurance necessary to do well at Cal. </p>
<p>There is a fine line between confidence and bravado</p>
<p>Please. Given all the cockiness and social pressure to go to the top colleges people aren’t going to think they’re all that for graduating well at a lower college.
High school is different, since you don’t usually hear about many other high schools or have as much choice in which one you go to.
Also, subjective evidence thumbs down. Coming from a noob school does not mean you’re a noob.</p>
<p>You say that it is better to learn through the pain of failure than go through life with the illusion of success. Are you sure? Then perhaps you could explain the notable success of graduates of HYPS. Let’s face it - almost all of those students were stars in high school, yet few will ever experience much ‘failure’ at those colleges, in which it’s nearly impossible to actually receive a failing grade. {You might not get top grades, but it’s practically impossible to actually flunk out.} Students emerge from those schools brimming with confidence - or bravado, if that’s the term you want to use - which clearly doesn’t seem to hurt them, as evidenced by the boatloads of HYPS grads heading to top employers and grad schools. </p>
<p>Which leads to the other problem. Even if it really were true that people learn better through the pain of failure rather than the illusion of success, grad schools and top employers clearly don’t seem to agree. They’re not going to take you if you don’t have top grades, even from an easier school. An argument that your 2.0 Berkeley GPA proves that you learned from the pain of failure and are therefore a better candidate than others who have high GPA’s is going to fall on deaf ears. Heck, you probably won’t even be allowed to present that argument as your application will be tossed as soon as the adcom or hiring committee sees your GPA. Sad but true. </p>
<p>That all highlights the real problem. If I were to be magically invited to tryouts with the Boston Red Sox and fail miserably, nobody cares. Nobody would even know. Only I would know that I was a failure in baseball, and so I’d be able to move on with my life with a clean slate. But bad grades become part of your permanent academic record with which you will be judged by grad schools and employers for the rest of your life. </p>
<p>What I believe ought to happen - and what I have proposed on other threads - is that schools should simply cancel the grades of those who don’t graduate due to poor performance. If somebody isn’t going to earn a Berkeley degree anyway, who cares what his Berkeley grades were? Let him go to another school with a clean slate if he wants (and if he prefers to maintain those grades for transfer credit purposes, he could do that: it’s his choice). Similarly, just because I failed miserably at the Red Sox tryouts doesn’t mean that my permanent record is forever besmirched in the eyes of future employers or grad schools. But if schools aren’t willing to cancel those poor grades, then they should simply not admit those students who would predictably earn poor grades in the first place.</p>
<p>It’s better than graduating at the bottom of your class at Berkeley, or heck, not even being able to graduate at all. It’s better to graduate from San Jose State than flunk out of Berkeley, and that is empirically provable. As we all surely know, many employers require, or at least strongly prefer, hiring college graduates, often times with GPA cutoffs (usually around 3.0). If you don’t meet that cutoff - or if you don’t even have a degree at all because you flunked out - they’re not going to care why. All they will care about is that you didn’t meet the hiring criteria. Sad but true.</p>
<p>I don’t have a proposal because I don’t have the data. But Berkeley does. It would be a simple statistical data modeling exercise. Take all of the student registrar data of the last, say, 10-15 years and look for admissions ‘variables’ that seem to predict whether a student is going to encounter academic difficulty (i.e. land on academic probation or otherwise earn poor grades). Then simply admit fewer applicants with those same variables in the future. Given enough data - which Berkeley surely has - you could even come up with a dynamic model that predicts with reasonable accuracy the likelihood that a given applicant will encounter academic difficulty if he matriculates. You could then establish certain common sense rules: i.e., anybody who has a greater than 90% of landing on probation will not be admitted. Granted, such a system would not be perfect, as one could criticize this rule or that. But it would be a clear improvement over the current system in which students are often times admitted without much rhyme or reason. </p>
<p>The advancement of statistical tools combined with ever-cheaper computer power and free open-source software means that schools can easily develop statistical models to guide their admissions decisions. The characteristic of ‘not being relegated to future probation/earn poor grades’ should be a highly desirable - perhaps one of the most desirable - traits of any applicant. Again, I don’t presume to know what the data says. But I do know that the data exists and can be leveraged.</p>
<p>Besides, to my detractors, let me turn the question around. Can anybody come up with a reasonable argument as to why Berkeley should not use that data in the manner described? What exactly would be the rationale for Berkeley to not have future admissions decisions guided by the performance of past students, in an effort to have fewer future students who will perform poorly?</p>
This is a pure educated conjecture on my part:
The folks at Berkeley have done extensive statistical modeling but they don’t like the results, which suggest an even less diverse student population than the current one produced by holistic approach.</p>
<p>I’m in accord with this, and I agree not everything that could be done is being done. I guess I was more worried about irrationally drastic class size decrease proposal, but it seems like you’re advocating something much more systematic, and I’m in favor of that.</p>
<p>They already do this. Admissions try to predict college success using a variety of measures such as high school GPA, SAT scores, and passion/drive through personal essays. What you’re really suggesting (I’m assuming this based on your past posts) is to take this even further and use data such as graduation rates of students from certain regions or certain high schools, so that if admits from a certain high school graduates more, Berkeley should admit more students from that high school. Or, if Berkeley sees that students who took biology in their senior years tend to graduate college in higher numbers, Berkeley should admit more students who took biology senior year.</p>
<p>I see two problems with this: first, this system would be very slow to react to major changes. I already provided one example: a high school receives funding/donations, revamps its school and curriculum, adds AP courses, hires new teachers, and dramatically improves its quality in a short period. Students from this high school would be unfairly judged as poor candidates for admission based on their high school’s low-performing, and irrelevant history. Under the current system no such bias would exist, as each student is judged based on the merits of his accomplishments instead of statistics that may or may not be outdated.</p>
<p>Another example: I believe a few years ago the statewide testing was changed so that 10th graders were tested on the life sciences. Because of this, my old high school (and many others in the state I’m sure) had to change their whole science curriculum around because biology used to be a senior course, but now they had to have freshmen take it to prepare them for the statewide exams. Had Berkeley had a statistic where they were biased to admitting students who took biology in their junior/senior years, this would cause an unfair bias across many high schools because of a change in the system. Again, these biases do not exist under the current system.</p>
<p>My second concern: should we place too much emphasis on college success? Should we endeavor to admit only those who are most likely to do well, i.e. earn a 3.0+? Because if we were to do that, certainly other factors would suffer. The student body would be a lot less diverse. Certain bright students from poorer areas, who bring their own perspectives and are valuable to classroom/out-of-class discussions, would not get in. There could be an overwhelming number of engineering students and not enough in certain other majors. Our sports teams would likely be much worse off. The male-female ratio might become one-sided (look to MIT and Caltech as examples; in fact MIT has put a lot of emphasis into admitting more female students). This is a large motivation behind Berkeley’s current holistic admissions approach: because there are many factors in play, not just how well an admit would potentially do in college.</p>
<p>Now, I certainly see the merits to your proposal, and I think it might be better than the current system, but I’m not convinced that it definitely is better than the current system.</p>
<p>
</p>
<p>I was simply debating whether students who do poorly at Berkeley (get 2.0-2.5) would be better off at another college (say, UC Merced), and I was arguing that they might not be better off somewhere else. I don’t think this is changing the debate. There are some students who just end up with a terrible work ethic, or get into drugs, or have family problems, and they would do poorly anywhere. Also, if you are struggling to graduate Berkeley, it’s far from a guarantee that you’d be an academic superstar at a place at UC Merced; you might get a 3.5-4.0, but you might also do poorly at UC Merced.</p>
<p>But even if I were to accept your assumptions: it’s not entirely clear that someone who earned a 2.0-2.5 from Berkeley would be less confident or less successful in life than someone who earned a 3.5-4.0 from Merced or a Cal State. Graduating from a prestigious university like UC Berkeley is a confidence boost in and of itself. For many people / companies / fields, all that really matters is getting the degree, and grades are secondary. Hey, C’s get degrees, right? The fact that George W. Bush or John Kerry got straight C’s at Yale certainly didn’t seem to hurt their confidence or their success later in life.</p>
<p>Now, I have given this some thought and here is my current proposal:</p>
<p>Keep the number of admits steady, and attempt to increase the graduation rate by means of grade inflation. This is similar to what the elite privates are doing. More specifically, by inflating grades in certain majors. As it is now, certain majors have highly inflated grades while others (i.e. engineering) do not. For most students who are struggling to graduate, the reason is because they are in a major that is too competitive, grades too harshly, or their professors are terrible at teaching, the class is poorly structured, and there is little or no support for those who need help.</p>
<p>As it stands, the best course of action for your average engineer (and this is what I recommend to others) is this: spread out your required classes, and fill the rest of your schedule with filler classes: classes in other departments with very little work and easy grading. That way, the other classes will artificially raise your overall GPA, while you have more time to work on your engineering classes. Many employers only look at overall GPA and don’t bother to look at your transcripts. Now, wouldn’t these students be better prepared for the industry by taking more engineering classes? Probably, but under the current system many would struggle to graduate and end up with a terrible GPA.</p>
<p>So, first, inflate grades across the board so that all departments are fairly equal, and rarely if ever assign failing grades. On top of this, we need to make it easier for the students to succeed in Berkeley. There are several ways to do this:</p>
<ol>
<li><p>Have better professors. Take the CS department for example. There is an honor society ran by students that make professor evaluations public. If you look at the history of various CS classes, the professors who received terrible evaluations did not come back to teach in future semesters, while the professors who received good evaluations continued to teach the class often. Another way to increase teaching quality is to separate the researchers from the lecturers. I know of a few classes that are taught by dedicated lecturers who do no research and basically have “tenure” to teach that class every semester. These are great ideas that should be expanded further and to every department.</p></li>
<li><p>Have better support for struggling students. For example, encourage classes to set up study groups for students in class, have TAs hold review sessions every week, force professor to make notes available. I’ve noticed a trend where most classes now have previous midterms/finals available for the students to study from; this is a step in the right direction.</p></li>
<li><p>Integrate departments better. As it stands now, many departments operate independently, which also leads to disparate grading systems like I have mentioned earlier. Allowing sharing of resources across departments, and shifting resources between departments (I know you’ve suggested this before), would benefit all departments across the board.</p></li>
</ol>
<p>If Berkeley can do this, if it can make it easier for students to do well, then grade inflation should not be a problem, in fact it should be the natural result of such a change. This way, Berkeley could keep its current admissions system and not have to flunk out students and effectively punishing them for going to Berkeley.</p>
<p>Because what’s the worst thing to happen? Berkeley graduates some students who are weaker than others. The intangible benefits from having a diverse student body should more than offset this. Those who don’t do well will simply graduate with poor grades, but (just about) everyone would graduate.</p>
<p>
</p>
<p>This wouldn’t even be an issue under my proposal because no one would fail out of Berkeley. That’s even better than having your grades wiped clean and start somewhere else.</p>
<p>First off, I think I can safely rely on the fact that changes at the individual high school level are also similarly slow in coming. Let’s be perfectly honest: most high schools barely change, if at all. A bad high school in one year is still probably going to be bad the next year. To take your example, even if a high school does dramatically change its curriculum, let’s face it, most of the teachers are still going to be the same, as it is exceedingly difficult for any school to terminate a large number of teachers quickly. The student culture is still largely going to be the same. </p>
<p>But having said that, I think you’re greatly discounting the responsiveness of a statistical model to new variables. For example, if a formerly bad school suddenly produces a large cohort of AP-course-taking students, then that fact alone may be enough to outweigh the negative influence of the overall school effect (depending on how strongly the ‘AP-course’ variable predicts the chances of avoiding probation, which is ultimately an empirical question.) Students who study under newly hired teachers at a formerly bad school could be coded as such, if that information were available. </p>
<p>
</p>
<p>And again, you greatly discount the power and flexibility of statistical models to handle precisely those types of exogenous shocks. Statewide changes to high school curricula would surely be well-known by Berkeley, and proper adjustments to the model could be made accordingly. For example, the model could easily accommodate your specific concerns by simply resetting any variables that biased the results in favor of admitting students who took biology in junior/senior year. Surely more sophisticated statistical techniques would be available, depending on the available data. </p>
<p>
</p>
<p>And yet the underlying problem is that the price is paid by those students who perform poorly, in the name of this purported ‘diversity’. This is not a game: you’re playing with people’s lives. Somebody who flunks out of Berkeley will have that fact marked on his academic record for the rest of his life. </p>
<p>Besides, I’m not sure that I agree that my policy would actually decrease diversity - in fact, it might increase it, if diversity is measured as some sort of normalized variance over a vector of personal characteristics. For example, it is entirely possible that the students who tend to perform poorly also tend to have highly correlated personal traits, and excluding them would actually serve to increase overall student diversity. But again, this is ultimately an empirical question. </p>
<p>
</p>
<p>Ideally, those who would perform poorly at Merced would not be admitted to Merced either. Perhaps Berkeley could share its admissions data model with Merced. </p>
<p>But all I can do here is deal with Berkeley. If other schools wish to make admissions mistakes, that’s their prerogative. </p>
<p>
</p>
<p>Given their family connections, Bush and Kerry would have been successful no matter what they did. Besides, Bush and Kerry went to college at a time when grades had not yet inflated due to efforts by professors to allow their students to dodge the Vietnam War draft. It’s perfectly fine for your confidence to receive C’s when plenty of students around you are also receiving C’s. </p>
<p>But I don’t think too many future Bush’s and Kerry’s are coming to Berkeley for undergrad. {Let’s face it, such a person would probably still get into HYPS today, and that person will almost certainly choose those schools over Berkeley.} </p>
<p>
</p>
<p>As I said before, the clearest example would be derived from the policies run by grad schools and, especially, the better employers who employ GPA cutoffs (usually 3.0). If you have a 2.0 from Berkeley, you’re not going to be hired by those companies, no matter what confidence you may have acquired simply by graduating from Berkeley or however much you may have learned from your pain. They won’t care. All they’ll see is that you didn’t meet their GPA cutoff, however arbitrary it may be.</p>
<p>Ha! Not to be overly territorial, but I think many of those were actually my ideas, or at least modified versions of my ideas. Sure, do them all. </p>
<p>But I don’t see why those ideas are incompatible with this new proposal. </p>
<p>
</p>
<p>And we’re still left with the problem of some students graduating with poor grades. As I said before, to be barely surviving with less than a 2.5 is a rough existence. Unless you happen to be extremely well connected as Bush and Kerry were, and living in an age when a C was still considered a respectable grade as they were, you’re going to find that obtaining a job from any decent employer will be tough sledding. Graduate school is definitely out. I again also fear greatly for your personal confidence. For example, one of my brother’s great criticisms of Caltech, and mine of MIT, is that the schools produce a large cohort of graduates who, frankly, are psychologically traumatized. 4 years of earning mediocre grades will do that to anybody, even if you do actually manage to graduate.</p>
<p>So let’s say there are students who study under new teachers with good track records at a low-income area high school. Now, should we view this as, the school is getting better so we should admit more of these students? What if the new teachers had good track records only because they used to teach at high-achieving high schools, and we should really admit fewer of these students because the high school still has poor resources? This all seems highly subjective and I’m not sure that statistical mining can account for all the different factors.</p>
<p>
</p>
<p>But again, how would they know which variables to reset? What if this new shift in curriculum causes most students to take biology their freshman year, except for a certain group of students who want to take AP biology which is offered their senior year, opting for physics their freshman year, so all the strongest students are still those who take biology their senior year? (this is currently happening at my old high school) Should we reset the variable then? But then at other high schools, the only people who take biology senior year are those who failed it and need to retake it before they can graduate? It’s unclear what to do in response to such a change. And what if the changes aren’t statewide, but restrict to a much smaller region? Would we even notice the change in time to reset our variables? This might be too much trouble that admissions doesn’t want to deal with.</p>
<p>
</p>
<p>I think we only have to look to MIT and Caltech to see that this, unfortunately, isn’t true. It’s no secret that the other elite privates make enormous efforts to admit a diverse class and admit URMs with lower stats, surely this wouldn’t be needed if all we had to do is simply admit the highest achieving students?</p>
<p>
</p>
<p>Oh, I’m not claiming that I thought up of these ideas. I know you brought up grade inflation before and I was iffy about the idea, but now I think it’s really the way to go. I’m not saying they’re incompatible with the new proposal, just that I’m not entirely sold on it. It would depend a lot on the implementation.</p>
<p>
</p>
<p>Well, if we try to foster an environment with quality teaching, sufficient out-of-class support, a fair and lenient grading system, then I don’t think that will be a big problem. Something like what Harvard or Stanford has. I mean, if you’re getting straight C’s at Harvard, you probably deserve it, and grad school is probably not for you.</p>
<p>I think judging a student by their high school is a slippery slope. While in a perfectly implemented model (which it almost certainly will not be), it may allow a better prediction of success, it changes the current standard (at least in the UCs) that a student is judged individually on their accomplishments, and not by the school they’ve come from (something they have little influence over). Somehow this seems ethically questionable to me (kind of like Gattaca where you’re judged by your DNA).</p>
<p>
</p>
<p>I think part of vicissitudes’ plan is to do grade inflation at the level of HYPS, applied evenly to all departments. This will virtually eliminate students “flunking out” of Berkeley. For most technical departments, such as Engineering, the Sciences, etc., this will mean a big increase in the average GPA. For humanities, I expect to see very little change or even perhaps deflation from certain departments where the average GPA is currently ridiculous. Yes, a given GPA will mean less than it used to, but I don’t think most people ask the question: “How hard was it to get X GPA at this college” anyway.</p>
<p>As it stands, many departments in Berkeley have average GPA requirements that are extremely low. For example, the average grade given in a lower division class in CS is 2.7. This is a requirement for professors and they can get in trouble if they do not adhere to it. No matter how selective your admissions process, this basically means the lower 10-20% of the class is going to get a C or below. Ignoring humanities for a minute, this means 10% of the class would flunk out, no matter how smart they are. This is clearly unacceptable, not just for flunking out, but in giving a GPA that compares accurately to other institutions, and needs to be fixed. Once it has, I think most of the issues sakky has been raising, from confidence to students flunking out, would be solved.</p>
<p>Attempting to predict a student’s academic performance using “variables” that have nothing to do with academics is blatant prejudice. Qualities possessed by a group to which I belong have nothing to do with my capabilities as an individual.</p>
<p>This sort of prejudice is supposed to have lead to affirmative action at prestigious private universities. They defended the perceived disadvantage Asians have in college admission by stating that Asians tend to underperform after starting University, perhaps because they put too much importance on getting in.</p>
<p>BTW, the lack of grade inflation in science and engineering disciplines applies to all colleges, not just Berkeley. If you think you are at a disadvantage compared to the Ivy Leagues or that you got a bad grade because of grade deflation, you are wrong.</p>
<p>I am quite skeptical of this. I think the science and engineering disciplines are hard at many schools, but there’s a big difference between “not grade inflated” and “grade deflated.”</p>