<p>I find that freshman admits blame transfers for a lot of things. Just an observation.</p>
<p>But really, I think that Berkeley and UCLA don't perform well in the rankings game because they don't play it like WUSTL or Stanford do. It's kinda sad, really.</p>
<p>Matt,</p>
<p>I hate to say it, but graduate level social science work is a lot of times just slogging through data as well...</p>
<p>
[quote]
Sure there’s is no evidence to suggest transfers are in any way superior, but there is also none to suggest they perform any worse. Therefore is fallacious to suggest Berkeley should be worse in the USNews rankings based only on the number of transfers. (Kinda making a giant circle back the OPs original topic)
[/quote]
</p>
<p>I'm never said that Berkeley should be worse in the rankings. </p>
<p>I am simply countering the contention that Berkeley ought to be better solely because of SAT measurement problems. Some people have pointed out that Berkeley only reports combined SAT scores, and other schools report individual scores, so that Berkeley artificially suffers as a result. My response to that is that Berkeley should then recompuate the SAT scores of ALL of its students, including its large pool of transfers, and then see what the "true" average score really is. </p>
<p>It's like a student wanting his exam to be rescored, but only those particular questions that he got wrong. The more fair thing to happen is for his ENTIRE EXAM to be rescored, such that he may gain points on questions of which he was inaccurately marked wrong, but he may also actually lose points on questions on which he was inaccurately marked correct, such that his overall test score might increase or decrease. </p>
<p>The point is, you can't raise one test methodology problem without raising ALL of the problems.</p>
<p>"The point is, you can't raise one test methodology problem without raising ALL of the problems."</p>
<p>No doubt. There are loopholes that allows lesser schools to gain spots in USNews rankings as most of us are aware. But, as they add more relevant criteria, we get a more accurate picture of the schools while minimizing the discrepancies.</p>
<p>"I hate to say it, but graduate level social science work is a lot of times just slogging through data as well..."</p>
<p>But anyone can plod through data. I realy think creativity/imagination plays a greater part in award winning research than most people think.</p>
<p>Well...yes and no. I'd say that modern poli sci (especially the more quantitative stuff) is about 10-15% creativity (good thesis and observation) about 10% good math skills, and the rest is just...legwork. I had a great idea for a paper as an undergrad that never came to fruition because I simply didn't have the patience to go through miles of Japanese data...in Japanese.</p>
<p>
[quote]
For example, take EE117 and Physics 110. Aren't they basically the same thing - electromagnetic fields/waves? Or take Math 118 (wavelets and signal processing). Honestly, isn't that really a class more appropriate for EE? Isn't CS 174 (Combinatorics and Discrete Probability) really just a math class? CS 170 and 172 are also arguably math classes. In fact, Alistair Sinclair, who teaches CS 170/172/174, sometimes publishes his articles in mathematics journals.
[/quote]
</p>
<p>I think the reason why is because the simple fact is that EE117 and Physics 110A/B aren't the same course. In fact, I doubt you could swap professors and have the same quality of teaching in each course. Although they generally cover similar material, some that overlaps, they emphasize different aspects of that material.</p>
<p>I'll give you a lower division example, since I don't have sufficient personal knowledge of physics 110A versus EE117. Math 55 and CS70 are both discrete math. They are technically interchangeable for the discrete math requirement for EECS majors. However, in CS70 we take a lot of the discrete math ideas and apply them to specific CS-related ideas, such as encryption. We write pseudocode algorithms for some topics in order to emphasize how they relate to CS.</p>
<p>Perhaps there are Math 55 professors that can code and do know enough CS theory to teach this course. It may be reasonable to confer with professors for courses in high enough demand to ascertain whether they could in fact teach a course outside their field. But ideally, I would rather have someone who has studied CS extensively to teach me CS theory than a math professor, because the CS professor has been vetted on this material while the math professor hasn't (even though he may know it).</p>
<p>My point is, CS and math are close, and physics and EE are close, but not close enough that I'd say I'd want professors overlapping regularly. My EE105 and EE140 professors could easily be designing analog or digital circuits for a living at the highest levels. Perhaps a physics professor knows how transistors work from a physics perspective, but without significant circuit design experience, I simply don't trust that s/he would give the same quality of instruction in a circuit design course.</p>
<p>I do agree that more flexibility is a good thing. The ability to react and respond quickly to the needs of the student community is a good thing. I've just never seen a case where this has been a critical issue, so I personally don't consider it very important relative to other things.</p>
<p>
[quote]
We could offer a number of things to sweeten the pot. Most notable would be something that guarantees admission to graduate school, such as a BS/MD program with UCSF, or a BA/JD with Boalt.
[/quote]
</p>
<p>I like that idea. Again, I personally don't see it as a big deal, but if it were on the table I'd support it.</p>
<p>
[quote]
I think Berkeley should try to compete for this talent. For example, for the honors students, why not give them a private desk somewhere on campus - or at least a locker? I would have loved to have had a place on campus to use as a temporary dropoff point instead of always having to carry all my stuff with me. Why not give them, say, a parking spot (which would be HUGE at Berkeley).
[/quote]
</p>
<p>Okay...but seriously, attracting students with parking lots and/or desks or lockers is kind of silly IMO. I guess the real difference between our viewpoints is that it seems like you take the administrative viewpoint. Yes it helps Berkeley look better to have better students, and giving them incentives is one way to get better students. Then you give a list of incentives.</p>
<p>For me, I don't care if we have better students. I want Berkeley to maximize its ability to take any student (namely, me) and give that student a great education. Actually, I think Berkeley would be best to avoid trivial incentives such as lockers and parking spaces and put more resources into better education. Attract students through better teaching and educational resources, not through parking spaces.</p>
<p>
[quote]
The fact that transfer students have a higher Berkeley GPA than the freshman-admits by itself means nothing if they don't take the same classes, and in particular, if the transfers don't have to take the weeders.
[/quote]
</p>
<p>To be fair, I know that in EECS most transfers have to go through EE20N, EE40, CS61A/B/C. I know they aren't Math 1A/B in terms of weeder status, but transfers don't get out of everything. It should also be noted that the number of weeder courses a student must take is generally small. I agree the statistics are misleading, but if we assume an average (for example) EECS student takes 4 "weeders" (Math 1A/B, Physics 7A/B) out of 32 courses (4 per semester assumption), we'd be talking about 1/8 of the GPA depending on weeders, which isn't an overwhelming amount (though significant).</p>
<p>Perhaps a simple random sample of SAT scores with a response level percentage could bring the scores into perspective. A school could have a 1400 student average (old SAT) but if data is only available for 40% of students that would knock the school down a rung compared to a similar school with a 90% response rating. </p>
<p>Of course, that would probably take more time than USNews is willing to spend.</p>
<p>
[quote]
Attract students through better teaching and educational resources, not through parking spaces.
[/quote]
</p>
<p>Unfortunately, as you may or may not agree, at the top strata of universities, the difference between professors and resources somewhat wears thin. It starts becoming a game of perquisites, name-brand, and $$$.</p>
<p>
[quote]
My point is, CS and math are close, and physics and EE are close, but not close enough that I'd say I'd want professors overlapping regularly. My EE105 and EE140 professors could easily be designing analog or digital circuits for a living at the highest levels. Perhaps a physics professor knows how transistors work from a physics perspective, but without significant circuit design experience, I simply don't trust that s/he would give the same quality of instruction in a circuit design course.
[/quote]
</p>
<p>Well, let me put it to you this way. I freely admit that this idea of mine is not entirely original. In fact, it is PRECISELY what MIT does. Many of the CS classes at MIT ARE math classes in the sense that the same class is actually cross-listed in both departments and thus can be counted in either major.</p>
<p>For example, MIT's EECS classes are tagged with the '6' prefix. Math classes are tagged with the '18' prefix. Yet note the following crosslistings.</p>
<p>6.042J - Mathematics for Computer Science -18.026J
6.045J - Automata, Computability and Complexity - 18.400J
6.046J - Introduction to Algorithms - 18.410J
6.337J - Introduction to Numerical Methods - 18.335J
6.338J - Parallel Computing - 18.337J
6.840J - Theory of Computation - 18.404J
6.841J - Advanced Complexity Theory - 18.405J
6.852J - Distributed Algorithms - 18.437J
6.854J - Advanced Algorithms - 18.415J
6.856J - Randomized Algorithms - 18.416J
6.875J - Crytography and Cryptanalysis - 18.425J
6.876J - Advanced Topics in Cryptography - 18.426J</p>
<p>To give you one simple example, Sipser, the guy who teaches 6.840J (Theory of Computation) is actually not an EECS professor at all. He's a professor in the Math department. Goemans, who teaches 6.841J is also from the Math department. </p>
<p>And that's just those EECS classes that are cross-listed with math. I haven't even started talking about those that are cross-listed with other subjects like physics. For example, I know that the MIT EECS nonlinear optics and plasma physics courses are fully cross-listed physics courses. </p>
<p>The point is, MIT seems to mix EECS with math and physics to the point where plenty of classes are actually cross-listed, yet I don't hear anybody complaining that EECS students aren't getting the proper kind of instruction. I've never heard of anybody take 6.840J and complain that he didn't learn what he needed to learn about computation theory because the class was taught by a math professor. I hear them complain about how difficult the class is, but not that they aren't getting proper instruction.</p>
<p>The point is, MIT cross-lists classes all the time, and nobody seems to think it's a problem. So if MIT can do it, why can't Berkeley? </p>
<p>
[quote]
I do agree that more flexibility is a good thing. The ability to react and respond quickly to the needs of the student community is a good thing. I've just never seen a case where this has been a critical issue, so I personally don't consider it very important relative to other things.
[/quote]
</p>
<p>I can't believe you don't see it. Once again, I would point out - all those people who wanted to get into CS or EECS but didn't get in. More flexibility may ultimately mean more spots for these students. </p>
<p>
[quote]
Okay...but seriously, attracting students with parking lots and/or desks or lockers is kind of silly IMO. I guess the real difference between our viewpoints is that it seems like you take the administrative viewpoint. Yes it helps Berkeley look better to have better students, and giving them incentives is one way to get better students. Then you give a list of incentives.</p>
<p>For me, I don't care if we have better students. I want Berkeley to maximize its ability to take any student (namely, me) and give that student a great education. Actually, I think Berkeley would be best to avoid trivial incentives such as lockers and parking spaces and put more resources into better education. Attract students through better teaching and educational resources, not through parking spaces.
[/quote]
</p>
<p>Desks and parking spots are cheap thrills. There's nothing wrong with cheap thrills - it's better than nothing and can serve as the starting point to greater reforms. The real problem seems to be administrative institutional culture, and the way you change culture is that you start with quick wins (i.e. cheap thrills) and use them to build momentum. </p>
<p>I also disagree with your notion you shouldn't care about the quality of the students. You should care, because better students means a better education. That's because education is a social phenomenom. When the students around you are smarter and work harder, you tend to learn more. When students around you are lazy and untalented, you tend to learn less. </p>
<p>
[quote]
It should also be noted that the number of weeder courses a student must take is generally small.
[/quote]
</p>
<p>But only if you actually graduate. Keep in mind that plenty of students actually flunk out. For these students, the weeders may represent most of the classes they ever took at Berkeley (before they flunked out), and are almost always the reason for why they flunked out. They flunked out because they did poorly in the weeders.</p>
<p>"You should care, because better students means a better education."</p>
<p>I think the biggest difference between you and me (I don't mean that pejoratively), is I don’t think that better SAT score equates to a better student.</p>
<p>
[quote]
Look, my point is this. I don't see any evidence to believe that USNews is systematically biased against Berkeley. After all, if USNews was really biased, then you would expect that Berkeley's USNews undergrad AND grad rankings to be low.
[/quote]
</p>
<p>US News picks arbitrary metrics to evaluate colleges. I don't see how you could possibly exclude bias in choosing such metrics. First you have to decide which metrics to use, then how much weight to give each, then go about measuring the values of those metrics. Since choosing and weighting metrics is entirely subjective, US News is biased in some way (either positive or negative) in every one of its ratings. Perhaps that means it is biased against Berkeley undergrad and for Berkeley grad (I'm not saying this is true, just a possibility).</p>
<p>If I gave you a metric which stated that the college with the most sunshine got 5 points on my scale of 20, I wouldn't necessarily be biased against any university. "You can improve how much sunshine you get--just buy land in Texas and relocate!" When you pick an arbitrary metric, there will always be bias, period. See, US News wouldn't be biased if they labeled their rankings "America's Colleges That Score Best On These Metrics". They don't, though. They label it "America's Best Colleges".</p>
<p>
[quote]
It is as if US News and the Berkeley critics on this board believe that a wall surrounds undergrad engineering and business and the extraordinary grad programs at Berkeley, and little or nothing reaches beyond this wall to enhance the experience of the great majority of Cal undergrads. That’s ridiculous.
[/quote]
</p>
<p>Uh, actually, there is a wall. Not everybody who wants to major in engineering or business gets to major in them. Haas only admits 50% of the continuing students who apply. And that's just those that apply. Plenty of other Berkeley students don't even apply because they know they won't get in. For example, if you are pulling a 2.5 in the prereqs, you know you're not getting into Haas. </p>
<p>So the resources of engineering and Haas are not really available to all of the students at Berkeley. What is that, if not basically a "wall"? </p>
<p>
[quote]
the public university best able to compete with the elite privates Berkeley is hurt most by this bias. One indication of the ludicrous and internally inconsistent quality of the US News rankings is to look at the only two undergraduate areas ranked by US News, engineering and business. I have not examined this year’s issue, but for years Berkeley undergrad engineering and business were both ranked by US News in the top five nationally, and several years in the top three, while at the same time Berkeley’s overall undergraduate ranking ranged from 20-27.
[/quote]
</p>
<p>The answer to that is simple - because engineering and Haas really are better than the overall undergraduate program. Those 2 programs do things that I wish Berkeley as a whole would do to improve itself - namely be more selective, have a better student-faculty ratio, have tighter ties to corporate recruiting, offer BS/MS programs (in the case of some of the engineering majors) etc.</p>
<p>Let's take Haas as an example. Like I said, Haas only admits 50% of the continuing students that apply. I would furthermore guess that for every one student who applies to Haas and doesn't get in, there probably is another student who wants to get in, but doesn't even apply because he knows that he can't get in. Like I said, if you have a 2.5 in Haas pre-reqs, you know you're not going to get in. Hence, I would say that the "true" admission rate to Haas is more like 25% in the sense that only 25% of those students who want to get into Haas actually get in. </p>
<p>Imagine how much better would be if Berkeley only had the top 25% of its current undergrad student body. I think Berkeley would now give HYPSM a very serious challenge. </p>
<p>Now, let me make things clear. I am not saying that I advocate that Berkeley get rid of 75% of its undergrads. I know somebody out there is going to accuse me of this, so let me make it crystal-clear that I do not advocate this. </p>
<p>I am simply explaining the major reason why the Haas BS program is so much more highly regarded than the undergrad program as a whole. Haas simply has fhigher average student quality than Berkeley as a whole does. That's why Haas is ranked higher.</p>
<p>
[quote]
I think the biggest difference between you and me (I don't mean that pejoratively), is I don’t think that better SAT score equates to a better student.
[/quote]
</p>
<p>And when did I say that a better SAT score necessarily equated to a better student? I never said that. I think that Shiboing Boing may have said that, but I didn't, and if he said it, you should take it up with him. </p>
<p>I am simply saying that there is a concept of a 'better student', and Berkeley would be better if it attempted to upgrade its student body. That doesn't necessarily mean only getting people with top SAT scores. I don't want to study with somebody who has a great SAT score but who is downright lazy, for example.</p>
<p>
[quote]
The point is, MIT cross-lists classes all the time, and nobody seems to think it's a problem. So if MIT can do it, why can't Berkeley?
[/quote]
</p>
<p>Because they aren't the same courses! We DO have cross-listed courses, just not as many as MIT does. I don't see that as a problem if the courses are, in fact, different. If you have two classes in different departments that have identical curricula, then I agree, cross list them. But we already do that! We just do have as many of those as MIT.</p>
<p>
[quote]
I can't believe you don't see it. Once again, I would point out - all those people who wanted to get into CS or EECS but didn't get in. More flexibility may ultimately mean more spots for these students.
[/quote]
</p>
<p>I'm sorry, I don't see it as critical that students have to compete to get into a major. Critical would be "the EECS department has seen a massive shortage in <something> that the department needs to <supply buy="" hire=""> more of now, mid-semester".</supply></something></p>
<p>
[quote]
I also disagree with your notion you shouldn't care about the quality of the students. You should care, because better students means a better education. That's because education is a social phenomenom. When the students around you are smarter and work harder, you tend to learn more. When students around you are lazy and untalented, you tend to learn less.
[/quote]
</p>
<p>We know where we stand here, so that's that. Pure disagreement.</p>
<p>
[quote]
But only if you actually graduate. Keep in mind that plenty of students actually flunk out. For these students, the weeders may represent most of the classes they ever took at Berkeley (before they flunked out), and are almost always the reason for why they flunked out. They flunked out because they did poorly in the weeders.
[/quote]
</p>
<p>Average GPA of all students is what the statistic said. Doesn't include those that aren't enrolled (i.e. those that dropped out). This fact doesn't affect the validity of that statistic. Again, I'm not saying this is an apples-to-apples comparison, as the statistic is flawed. But maybe more like apples-to-pears--closer than apples-to-oranges, okay?</p>
<p>
[quote]
Uh, actually, there is a wall. Not everybody who wants to major in engineering or business gets to major in them. Haas only admits 50% of the continuing students who apply. And that's just those that apply. Plenty of other Berkeley students don't even apply because they know they won't get in. For example, if you are pulling a 2.5 in the prereqs, you know you're not getting into Haas.
[/quote]
</p>
<p>Yes, no wall! Let everyone in!</p>
<p>
[quote]
The answer to that is simple - because engineering and Haas really are better than the overall undergraduate program. Those 2 programs do things that I wish Berkeley as a whole would do to improve itself - namely be more selective, have a better student-faculty ratio, have tighter ties to corporate recruiting, offer BS/MS programs (in the case of some of the engineering majors) etc.
[/quote]
</p>
<p>More selectivity! Don't let in the slackers!</p>
<p>You can't have your cake and eat it, too.</p>
<p>
[quote]
US News picks arbitrary metrics to evaluate colleges. I don't see how you could possibly exclude bias in choosing such metrics. First you have to decide which metrics to use, then how much weight to give each, then go about measuring the values of those metrics. Since choosing and weighting metrics is entirely subjective, US News is biased in some way (either positive or negative) in every one of its ratings. Perhaps that means it is biased against Berkeley undergrad and for Berkeley grad (I'm not saying this is true, just a possibility).
[/quote]
</p>
<p>But the other possibility is that USNews may actually be biased in FAVOR of undergrad and AGAINST grad. I am not saying that I think this is true, but I am saying that just because a metric is biased does not mean that you are being discriminated against. You may actually be getting favored. </p>
<p>For example, I think there are certain USnews metrics that I think that Berkeley actually benefits from. For example, the whole metric of what percntage of students graduated from the top 10% of its class. That just artificially inflates the value of students who happen to come from a weak state high school system (and I do believe that California's high school system is rather weak), and happens to hurt those schools who draw lots of students from the elite prep schools. For example, Harvard matriculates a lot of students from Philips Exeter, including many students who are not in the top 10% at Exeter, so Harvard actually gets hurt in USNews by bringing in so many Exeter kids. But the truth is, a kid who graduated in the top 20% of Exeter is probably better than somebody who graduated in the top 10% of some mediocre high school. </p>
<p>But anyway, the point is not to say that I think that Berkeley is overrated. I didn't say that. What I am saying is that evidence of bias is not evidence of 'bias against'.</p>
<p>
[quote]
Because they aren't the same courses! We DO have cross-listed courses, just not as many as MIT does. I don't see that as a problem if the courses are, in fact, different. If you have two classes in different departments that have identical curricula, then I agree, cross list them. But we already do that! We just do have as many of those as MIT.
[/quote]
</p>
<p>You say they aren't the same courses. But that's my point - why shouldn't they be the same courses? At MIT, they're the same. So why can't they be the same at Berkeley? </p>
<p>
[quote]
I'm sorry, I don't see it as critical that students have to compete to get into a major. Critical would be "the EECS department has seen a massive shortage in <something> that the department needs to <supply buy="" hire=""> more of now, mid-semester".
[/quote]
</supply></something></p>
<p>Of course you wouldn't think it's a big deal. You're already in. Try telling it to somebody who just learned that he didn't get into CS or EECS that it's not a big deal. Careful, he might punch you in the mouth. It's easy to think that famine is unimportant as long as you have food. </p>
<p>
[quote]
Average GPA of all students is what the statistic said. Doesn't include those that aren't enrolled (i.e. those that dropped out). This fact doesn't affect the validity of that statistic. Again, I'm not saying this is an apples-to-apples comparison, as the statistic is flawed. But maybe more like apples-to-pears--closer than apples-to-oranges, okay?
[/quote]
</p>
<p>Huh? I can't believe you don't see the critical validity of the statistic. After all, students usually don't flunk out immediately. They almost always spend a semester or two on academic probation before finally flunking out. Hence, their bad grades are still counted because they are still officially enrolled students for the time that they are on probation. </p>
<p>Furthermore, flunking them out doesn't really "solve" the statistical problem. That's because as they are flunking out, another group of students right behind them is going onto probation, on their way to flunking out. Hence, at any given moment in time, you have a bunch of students who are on their way to flunking out, but haven't had it happen yet. Hence, their grades are still on the books, and are still affecting the overall GPA's of the freshman-admits.</p>
<p>Those students who went to CC and did poorly don't even get to transfer to Berkeley, so they're automatically "cut off" from the statistics. Yet those students who come to Berkeley as freshman and do poorly are counted in the statistics. That's a statistical anomaly.</p>
<p>
[quote]
Yes, no wall! Let everyone in!
[/quote]
</p>
<p>
[quote]
More selectivity! Don't let in the slackers!</p>
<p>You can't have your cake and eat it, too.
[/quote]
</p>
<p>How am I trying to have my cake and eat it too? </p>
<p>Berkeley Parent asked why Haas and engineering can be ranked higher than the Berkeley undergrad program in general. I gave the reasons. Those 2 programs operate under significantly different rules, and that's why they get better rankings.</p>
<p>I would say that, if anything, it is my detractors that want to have their cake and eat it too. They want Berkeley to rise in the rankings. But they're not willing to countenance the steps needed to do so. Like increasing selectivity.</p>
<p>
[quote]
establishing a program that not only creates a sort of caste at the school (which offends many people, it seems)
[/quote]
</p>
<p>As far as I can tell, this is really why it has not been done.</p>
<p>
[quote]
Of course you wouldn't think it's a big deal. You're already in. Try telling it to somebody who just learned that he didn't get into CS or EECS that it's not a big deal. Careful, he might punch you in the mouth. It's easy to think that famine is unimportant as long as you have food.
[/quote]
</p>
<p>And how about all those you want to reject in the first place? I'm not saying you're wrong in wanting to do it, or that I disagree (i'm not sure), but is it not that same?</p>