<p>It’s more an attempt to end a debate that is going nowhere, as you have proven unable not only to support your claims, but even to admit that the lack of available evidence precludes you from deducing what you keep trying to do. You haven’t presented anything beyond guesses (guessing what Stanford’s GPA is now based on 20-year-old data, guessing at how well the CS department’s grade distribution correlates with the university’s, etc.). You cite the fact that both draw from the top 10%, as if that matters at all (there are meaningful differences in the top 10%; Berkeley is drawing most heavily on California public schools, which are notoriously lower-quality; less than half the incoming students submitted high school rank at Stanford, though Berkeley refuses to fill that field out on its CDS because it likes to trumpet having x% in the top 10%, a meaningless statistic). And what does this mean about the lower-tier UCs that have nearly the same proportion of students in the top 10%? Since they have lower average GPAs (e.g. UCI is .3 points below Berkeley), I guess that means their grading is harder than Berkeley’s. ;)</p>
<p>Your other point, on SAT, is irrelevant, because I wasn’t basing my argument on the predictive ability of SAT or any single statistic, but rather on selectivity in general. I have admitted, multiple times, that I do not have the data on this (i.e. statistics on the various factors that decide admissions, like # ECs and such), and neither do you. Since we can’t really quantify selectivity (that’s what makes elite college admissions uncertain, after all), we have to go on our own assumptions. Without the data to prove it (which I’ve said many, many times), I am surmising that the difference in selectivity accounts for most if not all differences in grade distribution. Without the data to prove it (which you refuse to admit), you are surmising that the difference in selectivity cannot account for most/all differences in grade distribution.</p>
<p>Neither of us can prove our case (your top 10% point doesn’t sway me, just as my point that Berkeley admits 5x as many students, and loses most of its top admits to elite private schools, doesn’t sway you). We both have statistics that suggest our respective point is right. See how this post doesn’t advance the debate? It’s because neither you nor I have any new point to make, and neither of us has any new evidence to present. I wanted to spare you and everyone else the nitty-gritty tearing apart of your recent unsupported argument, and instead try to get you to agree to disagree and move on. In the process you might admit, at least, that this argument is “unwinnable,” and agree to end the debate nicely. But instead you seem to insist that I address all the illogical pitfalls of every inane argument of yours.</p>
<p>Tell you what: if one day you happen upon interesting statistics and you can construct a cogent argument with this data, feel free to send me a PM. But I have a feeling I’ll be waiting a long time. ;)</p>
<p>Given the vast resources at Stanford’s disposal, it’s very hard for me to sympathize with this position. Seems to me that Stanford could easily afford to employ a ‘dream team’ of the very best social researchers and statisticians in the world to enforce whatever ‘ethnic diversity goals’ it has to meet without having to explicitly use race as a standalone variable.</p>
<p>Allow me to propose a investigatory research design. First, carefully and numerically define what you (or Stanford) means when they say that the standalone race variable is used only “very little” (for example, does “very little” mean that being a URM, with all other covariates fixed, mean that your admissions odds would double? Or increase by 50%?) Then, run a two-part admissions process where Stanford would: (1) Run the current admissions process as it stands right now, but with all racial information expunged. But after (preliminary) admissions decisions are made, at that point: (2) Now you release racial information to the adcom and then carefully document exactly how many changes the admissions committee would like to make, now that it has full access to the race variable. Then compare that number of admissions changes to the prior claim that the race variable is used only ‘very little’. </p>
<p>{Now, granted, the admissions process would indeed be lengthened. But again, given Stanford’s vast resources, it’s hard to believe that Stanford couldn’t hire additional staff or pay overtime to incentivize adcom officers properly. If necessary, hire new adcom officers on a contract basis, which means that they can be terminated as soon as the project is finished.} Then Stanford ought to publish that information publicly: the claimed value of the ‘very little’ effect of the race variable, and the calculated true value of that variable. To enforce anonymity, Stanford could agree to publish that information only years later, and with the specific admissions year in question omitted. </p>
<p>
</p>
<p>I just proposed a potential solution above. </p>
<p>But moreover, if I am indeed posing more solutions than answers, it is because I don’t have access to the data. On the other hand, Stanford does have access to the data and therefore could provide the answers - but deliberately chooses not to do so. </p>
<p>
</p>
<p>I actually had already said that that AA doesn’t qualify as ‘fundamental denigration’, for like I said: it is far more serious. At the end of the day, denigration, [url=<a href=“define:denigrate - Google Search”>define:denigrate - Google Search]defined[/url</a>] to be criticism or disparagement, however irritating it may be, doesn’t really matter. But admissions processes are a far more serious affair. Like I said, adcoms are playing games with people’s lives by providing certain opportunities to certain people while denying those same opportunities to others, and it behooves all of us to ensure that such a process is conducted justly (however you choose to define ‘just’).</p>
<p>As an analogy, if I verbally racially disparage a URM, however morally reprehensible that behavior may be, I haven’t really hurt them. They’re still free to pursue their livelihoods to the fullest. But if I, as a cop, decide to arrest a URM under the guise of racial profiling - stemming from what even the most left-wing sociologist would concede is a statistically robust correlation between race and crime - I have now actually truly impinged on his life. For the rest of his life, he will have to report his arrest on any job application that asks about it, regardless of whether he beat the charges or not. Many employers will therefore likely choose never to hire him simply due to his arrest record.</p>
<p>None of that is to say that AA-based admissions is necessarily wrong. I’m simply demonstrating that the stakes involved in AA-based admissions are far more weighty than a matter of mere ‘disparagement’. Stanford isn’t merely verbally insulting somebody based on race, they’re actually making admissions decisions - and therefore influencing future career opportunities - based on race. </p>
<p>
</p>
<p>Actually, I’m afraid to have to tell you that they are indeed mathematically symmetrical and therefore equivalent.</p>
<p>It’s all depends on your countries culture towards upwards mobility:</p>
<p>If the majority of the people want a system where upward mobility is maximized, then using socioeconomic status as a factor in admissions is one small step in that direction since there is a correlation between SAT scores and income (or so I’ve heard)</p>
<p>If the majority of people want a system based on pure merit alone, disregarding the fact that we all have different socioeconomic starting points which may affect the likelihood of certain groups of people upward mobilizing at high rates, then we’ll do that system instead.</p>
<p>I personally don’t give a damn since I’m a ORM who got into a decent college and I’m happy where I am since I know in the long run, any “affirmative action” (assuming it’s even noticeable) which is employed won’t permanently screw my life over, and if that system ends up choosing some kid who grew up with less access to resources over me, then that’s great for him and sorta blows for me, but it wasn’t like my socioeconomic status decreased.</p>
<p>If there ever was a situation where some URM got into say Harvard over me because of some diversity policy (which btw doesn’t make any sense as an argument since it’s impossible to prove and it’s probably you’re own lacking ability that resulted in your not getting in, and here I’m assuming I’m intelligent enough to get in), then who cares? If I’m smart I’ll probably find decent resources at some other university and work my way up from there. Life isn’t fair, but we all adapt and move on.</p>
<p>If the tools at harvard allowed that kid to gain some skills that makes him/her valuable, then society as a whole just gained since my socioeconomic status either stagnated or only rose slightly (rounding down to a worst case), while his or her grew insanely, which means a net benefit to society greater than if he or she was rejected and I got in.</p>
<p>Both of you could keep arguing ad infinitum and boast your deft debating abilities, but it’s pretty evident with some of the anecdotal quoting and refuting (which in some instances seems to circle around), that neither of you seem to have the clear solution to the issue, yet enjoy picking each other apart more for the sake of a perceived victory as opposed to actually reaching a solution.</p>
<p>Diivio, I believe that I actually proposed one solution: compare a race-blinded vs. race-unblinded admissions process and see how much that the two differ. Then, at the very least, we could finally quantify how much race matters as a distinct variable. </p>
<p>
</p>
<p>But here you’re now assuming that race and socioeconomic status are correlated whereas the above discussion hinges upon the residual component of race entirely orthogonal to socioeconomic status, or at least the observable and measurable factors of socioeconomic status. </p>
<p>Indeed, as a practical matter, numerous studies have demonstrated that if anything, the practical use of race within real-world AA schemes actually tends to be negatively correlated with socioeconomic status, in that it is precisely upper-class URM’s who benefit the most from AA at the expense of poorer whites and Asians. Let’s face it, there aren’t exactly a lot of poverty-stricken inner-city URM’s going to Stanford or Harvard- the vast majority of them come from wealthy families. Does society derive a net benefit when Harvard admits the son of a wealthy black doctor or businessman over a poor Asian from Chinatown? </p>
<p>
</p>
<p>Then by that argument, you should have no historical objection to the segregated system of only a handful of generations ago where many URM’s, primarily blacks, were outright prohibited from numerous universities in numerous states of the nation as a matter of formal policy. That was, after all, what the vast majority of the voters in those particular states wanted. Clearly the vast majority within those states had little interest in a meritocratic system, but rather were far more interested in enforcing racial supremacy and privilege. Even the majority of voters outside those states had little desire for change which is precisely why nothing did change for many decades. But hey, that’s what the majority of people at the time wanted (or at least what the majority did not deem worthy of changing), right?</p>
<p>The problem here is that these arguments are exaggeration, and these words overstep what the logic of these arguments actually implies. </p>
<p>You say that my claims are based on “20 year old guesses.” This is an exaggeration, because they are in fact based on data from 2001. I can furthermore provide logical reasons as to why the CS department’s distribution would follow the university average (an average is the sum of its constituents divided by the number of constituents, so it’s mathematically obvious that there is a positive relationship between the school average and the department average), while this argument hyperbolically dismisses it by saying it’s a “guess”.</p>
<p>Even in the UC Irvine example the facts are wrong. UCI is 0.17 points below Berkeley. Assuming you used gradeinflation.com, this is what they have to say about UCI’s GPA:
The problem here we are talking about the difference in GPA at Stanford and UC Berkeley. This correlates to academic intelligence. In other words, GPA has to do with how well someone works academically and how well someone tests. The only way to quantify this difference is through SAT scores and HS GPA (there is a good reason why colleges emphasize this as the #1 factor for admissions). </p>
<p>“Selectivity” (the word you use) has nothing to do with this. “Selectivity” is only relevant insofar as it is a proxy for the average difference in intellect. “Selectivity” (in terms of admit rate) is influenced by factors like desirability of campus weather, presence of sports teams, how good a college is at marketing, etc. A school like Caltech has ~13% admit rate, yet I would not deign to say that Caltech engineers are any academically smarter/dumber than Harvard students.
I think you meant to say that I am “surmising that the difference in [intelligence] cannot account for most/all differences in grade distribution.”</p>
<p>
</p>
<p>Very well, let us say for a minute that the top 10% point is not relevant (something for which you bring no evidence other than to say that Berkeley does not report/collect it). How exactly is your point about how Berkeley admits 5x more students relevant? Berkeley has 168% the number of applications that Stanford has. Berkeley’s admit rate is thus only 3x that of Stanford. That seems like a big gap, right? The problem with the “selectivity” statistic is that we are dealing with such small numbers. Harvard has an admit rate of 5.9%, Stanford has one at 6.6%. Stanford, therefore, is 12% less selective. Can you thus conclude that Harvard students are 12% smarter than Stanford students? Again, you also try to draw a correlation between yield and intellect, never mind that there are numerous confounding variables such as… prestige, etc. Harvard’s yield was, at 81%, 11% higher than Stanford’s 73% yield. Can you thus conclude that Harvard students are 11% smarter than Stanford students? Obviously not. </p>
<p>Academic statistics like SAT scores are the purest and most untainted way of measuring any sort of difference in intellect, NOT selectivity.
And my point was that, using SATs as a proxy for academic intelligence (which is the only falsifiable way of assessing academic intelligence and its relationship to GPAs - this is a necessary condition for any sort of empirical rigor), most studies only show a moderate correlation between SATs and college grades. Thus, given that Stanford’s SAT score average is only 135 points higher, and given that SATs are the best indicator of academic intelligence, and given that SAT scores are themselves only a slight predictor of college GPA, it is clear that academic intelligence cannot fully account for the difference in average college GPA. (SAT score difference is 6.6%, GPA difference is 9.2%, which would normally seem like SAT does account for GPA, if it weren’t for the studies I brought up showing that only a fraction of the SAT difference can explain the GPA difference).</p>
<p>
You make this claim that you have statistics to back your point and that we have thus a situation of competing statistics. This is not true. There have been virtually no statistics (statistics require actual numbers) brought up on your part, while I have brought up numerous statistics, in addition to bringing up statistics as refutations, and in addition to careful analysis of the points presented.</p>
<p>
Talking about how you are able to tear apart someone’s logic, and then proceeding to… not tear down someone’s logic (even if a reason is provided) is not a very convincing way of going about things.</p>
<p>
Resorting to snarky comments like “I have a feeling I’ll be waiting a long time” really does not help your arguments.</p>
<p>yeah Sakky, you’re right, obviously I wasn’t saying the majority opinion is the CORRECT or JUST opinion, as numerous sociological and psychological studies have shown that “majority” human behavior favors obedience to authority or the “status quo” regardless of whether it is right or wrong (Milgram Experiment, Stanford jail experiment, etc.). I was just saying that unless the cause of justice is fairly clear, it is hard to come up with a firm opinion on it.</p>
<p>I personally believe a big reason the Civil Rights movement succeeded is because people also felt double standards and discrimination to be a “morally” reprehensible thing to do. But affirmative action doesn’t seem to have the same strength of conviction in that same regard.</p>
<p>I mean if Gandhi or MLK jr. tried nonviolence tactics to convince radical religious terrorists to stop harming people of other faiths, I honestly don’t think they will succeed because the terrorists don’t feel what they are doing is “morally” reprehensible. In fact they feel they are rewarded for it. A big reason I think they succeeded is that their enemies or oppressors had some ounce of “moral” conviction (In their cases it would be Caucasian Americans or the British). It wasn’t an accident that both of them should some form of “God” in their speeches. I feel it was intentional.</p>
<p>I could be totally wrong but that’s my guess.</p>
<p>No, you are basing it on a 10-year-old mention of an increase of 5% over another 10 years without even knowing the baseline in the first place, and are assuming that it went up or stayed the same over the past 10 years. These are guesses. Educated ones, but still guesses.</p>
<p>
</p>
<p>No, you can’t. So you know what an average is - that says nothing about any constituent that went into the average. Still a guess, and not even an educated one this time.</p>
<p>
</p>
<p>Yes, you’re right for once; I looked at gradeinflation.com but didn’t see the note that you needed to add 0.12 to all of UCI’s listed GPAs if you wanted to compare them to other schools. Most schools don’t have this caveat.</p>
<p>Either way, my point stands: UCI’s GPA is 0.17 lower than Berkeley’s. I guess UCI must have harder grading, right? So do CSU-Fresno (2.85), SJSU (2.85), CSU-Fullerton (2.68), and a ton of others. These are just a few examples of less selective schools with lower GPAs that were extremely easy to find on gradeinflation.com, yet you somehow missed them. How do you explain their lower GPAs?</p>
<p>
</p>
<p>That’s not what I’m talking about. Academic intelligence is only one piece of the puzzle, and is also poorly defined, hence why I used several characteristics that tend to correlate with selectivity (natural ability, talent, work ethic).</p>
<p>As I’m sure you know, after a certain point GPA and SAT become useless at differentiating students. (In fact, you were previously arguing against the use of these statistics because of their lack of predictive ability.) That does not mean that all students after that point are equal. Among these students there are meaningful differences that will bear out in the classroom and that must be probed without the use of numbers. You don’t believe that these factors have an influence on student performance in college (i.e. their GPA); I think they certainly do. It’s just that no one has been able to study this yet because any stats that admissions offices do have aren’t released. So what I said - that you reject selectivity as a (full/large) factor in GPA - stands.</p>
<p>
</p>
<p>You’ve drawn a large argument on a point I haven’t made. I have intentionally not used admit rates in any one comparison because that doesn’t allow you to draw many conclusions. One problem is the issue of self-selectivity; Stanford has a far more self-selecting pool, largely because it is much harder to apply to (you have to pay $90, write more essays, and send recommendations, transcripts and midyear reports). Berkeley’s isn’t as self-selective; it doesn’t even have its own application (Stanford at least has a supplement to the Common App), and all it takes is checking another box to apply, it’s cheap, no more essays, no recs, etc. hence how it is that Berkeley can get 10,000 more apps in a single year.</p>
<p>This also should answer your question on why Berkeley’s admission of 5x as many students is relevant. Hint: proportions cannot tell you everything (as is with most stats in life, you also have to look at raw numbers, like GDP vs. GDP per capita). </p>
<p>
</p>
<p>Such as… prestige. So what? Prestige would draw in more students - esp. those who are at the top of the admit class. Yield is important here since the majority of the 12,000+ students who are admitted choose not to attend Berkeley. If you looked at the top 2,400 admits (let’s say the admissions office gave you a list again), what would their yield be? Indeed, the further you go down this hypothetical ranked list of 12,000 admits, you’re likely to find that the yield steadily increases (students probably didn’t get in anywhere better at these points). This is the case for every school of course, even Stanford and Harvard, but the point here is that Berkeley likely fares far worse in attracting its top admits. IIRC even the yield for R&C scholarship recipients is poor, about the same as the general yield, so you can bet that if these students weren’t getting the money and perks of the scholarship, their yield would probably be much lower.</p>
<p>Sure, there are confounding variables, but do they erase the relationship between yield and the quality of the student body? Most likely not.</p>
<p>
</p>
<p>Trying not to palm-desk here. Academic statistics like SAT and GPA are the only way we have, so by definition they’re the ‘purest and most untainted.’ This again comes down to availability of data: the absence of evidence showing the true gap in student body quality (on average) is not evidence of absence. SAT shows some gap, but not nearly as much as it really is. This is because, again, numbers stop being useful after a certain point in selectivity. That’s why I’ve focused on the latter, in general, rather than the SAT, which has extremely limited utility here. </p>
<p>While I’ve said many times that data on selectivity isn’t available to us, I’m not throwing the baby out with the bathwater - selectivity still matters, far more than SAT, which becomes relatively useless long before you get to the level of 2100-2200 SAT student bodies, as at Stanford and Berkeley.</p>
<p>
</p>
<p>Read more carefully and you’ll notice that I said “We both have statistics that suggest our respective point is right.” You look to a 2001 Daily article suggesting a 5% increasing over the past decade, a 1992 official stat, etc. I look to stats on the difference in selectivity, the lower average GPAs for less selective schools, etc. Neither of these proves our respective point - they only suggest. More importantly, you have presented no statistics that would refute the relationship between selectivity and the average GPA. It seems your tactic is to assert, over and over again, that there isn’t much difference in the student bodies at Stanford and Berkeley - perhaps in the hopes that, if you say it enough, it’ll be true. Oh wait, you did bring up the fact that both schools draw on the top 10%, so I guess your job is done, right? :rolleyes: </p>
<p>I think the motivation to stick to your guns on this point is demonstrated quite clearly with this particular repetition of it:</p>
<p>
</p>
<p>In other words, your only argument is one of incredulity. You just can’t fathom that the difference in student bodies, on average, could be of that magnitude (to be clear, you reduce it to ‘intelligence,’ which I have repeatedly suggested is too narrow). And I can understand why you would want to reject that conclusion at all costs. To be clear, this discussion has nothing to do with any claim to quality or anything indicative of value; it isn’t inherently better to have grade inflation or grade deflation. And it doesn’t mean that Berkeley’s larger/less selective student body makes the university any ‘less,’ since Berkeley’s mission is, after all, to be accessible to a wider range of students. It provides an excellent education to a broader cross-section of society, which is commendable in itself.</p>
The data point is 10 years old. The increase of 5% is calculated off the 1990 data point, and that data point is an OFFICIAL data point that was released by Stanford - this is the known baseline.</p>
<p>
School Average = [(department 1 average) * (num of students in dept 1) + (dept 2 average) * (students in dept 2)+…+…) / total students.</p>
<p>Yet you claim there is no relationship (that is what is communicated when you say that it “says nothing”) between average and department 1 average (CS department, in this case).
The logical reason is that there is a relationship between school average and department 1 average (in this case CS).
I think what you meant is that there is not necessarily any guarantee of a correlation between school average and CS dept average. </p>
<p>However, this actually weakens your point. In a simple example with 2 departments, if Stanford’s average GPA is a 3.5, and the CS dept has a GPA of a 3.0, then Department 2 would have to have a GPA of a 4.0 - this example only serves to illustrate that if the CS dept had an average “out-of-whack” with the school average (lower than expected), then another department would have to have an even higher average GPA than expected.</p>
<p>I also don’t know why this got side-tracked into CS in particular. The issue is whether Stanford overall has easier grading.</p>
<p>
</p>
<p>In this case, the difference in student quality DOES account for the GPA gap.</p>
<p>There is a HUGE student gap between someone who goes to Berkeley and someone who goes to CSU-Fullerton, or even UCI. In order for Stanford’s higher student ability to account for the GPA gap with Berkeley, the difference between Stanford and Berkeley academic ability would have to be just under twice the difference between Berkeley and UCI student academic ability (this is a rough estimate based off the fact that the Berkeley-Stanford GPA gap is larger than the Berkeley-UCI gap). Big difference.</p>
<p>Furthermore, I don’t see how this furthers your point. Let us accept, for the sake of argument, your assertion that this means UCI had tougher grading than Berkeley. OK. So what? This has little to do with Stanford vs. Berkeley.</p>
<p>
</p>
<p>The aspects of natural ability and talent that contribute to academic performance would be accounted for by SAT scores / HS GPA. Extracurricular activities and personal stories - the differentiating factor with a school like Stanford, have little correlation with what kinds of grades people get - it may or may not correlate with whether someone will be more successful in future careers (due to soft-skills), which is what Stanford looks for, obviously.</p>
<p>What GPA and SAT test score DO tell us is how able someone is in an academic setting - how good someone is at taking tests and doing homework and doing schoolwork - things that correlate to what GPA you get. </p>
<p>You claim that selectivity and other factors also influence academic ability. This is true. However, they influence academic ability only because they correlate with an “intermediary” variable, which is work ethic and natural ability. These factors are accounted for by SAT scores.
The fact is that the test score differences between Stanford and Berkeley cannot account for the difference in college GPA.</p>
<p>
My technique was reductio ad absurdum. You did not say that or even claim that. However, I extended the logic of your arguments to make that statement.
</p>
<p>All these are possibly valid points. All you have given me are reasons why Stanford might be more selective than outwardly visible. </p>
<p>How exactly does this refute my point that the selectivity difference between Berkeley and Stanford is not relevant? </p>
<p>
Greater prestige increases yield, which is why Harvard has a higher yield than Stanford - this is the confounding variable. </p>
<p>Even if there is possibly a relationship between yield and the quality of the student body, the problem is, you are trying to draw a relationship between yield and the grades that student will get.
Yes, a higher yield means getting more of the top students. However, any effect that the yield has on the caliber of students would be fully accounted for by SAT/GPA differences, because yield only affects the grades a student gets insofar as it affects work ethic / natural ability, which are accounted for by SAT/GPA.</p>
<p>
I don’t get this. Selectivity data is available: admit rates and yield rates. It just so happens that when I used admit rates to show, how, reductio ad absurdum, admit rates are irrelevant, you dismissed that statistic.</p>
<p>Absence of evidence is not evidence of absence. That is true. The absence of evidence that a teapot is not orbiting the Sun between the Earth and Mars is not evidence disproving the existence of such a teapot. Yet the absence of evidence means that my assertion is to be preferred over other assertions, if were are to maintain any respect for empirical standards of proof. See: Russell’s teapot.</p>
<p>Furthermore, as far as standards of evidence go, SAT scores are to be preferred, since we cannot rigorously quantitatively analyze the other factors you mention. Standards of evidence are important - assertions go to the person who can actually bring evidence to the table.</p>
<p>
Is there any reason for this?
As I have argued, it does not matter, and SATs matter more, because the issue here is not whether Stanford students are better leaders than Berkeley students, or whether they’re more active in their community, or whether they’re more entrepreneurial. The only issue is whether they are academically more able.</p>
<p>
Certainly, you have brought up numbers. These numbers show that Stanford is more selective than Berkeley (admit rate and yield rate). OK. Great. That has not shown us in any way why or how Stanford’s higher academic ability accounts for the GPA gap. You have not brought up any percentages that you can point to and say Stanford students are (an estimated) X% more intelligent.</p>
<p>
That burden of proof falls on you. You have not shown any statistical evidence implying this relationship.</p>
<p>
No. You are twisting my argument. I have taken every step to ensure that wild “belief” and mere emotion / “feeling” plays no role in my arguments.</p>
Very well. I will use the term “academic ability,” which would account for natural talent and work ethic - this is what I have always meant, anyway.</p>
<p>
Is this playing the armchair psychologist? This isn’t quite ad hominem, but it is a rather thinly veiled attempt to inject into this debate my alleged “motivations” and my personal thought processes (my ability to fathom, apparently).</p>
<p>No, that’s what you understood, and to be pedantic, you have nit-picked the words. You know what it means: knowing the average doesn’t mean you know anything about any one of the parts.</p>
<p>
</p>
<p>No, it doesn’t weaken my point - that is exactly my point. I never claimed that some departments don’t have higher-than-average GPAs, because some certainly do. In order for some to be lower (e.g. STEM departments), others have to be higher. But whether English or psychology has a higher-than-average GPA is irrelevant.</p>
<p>
</p>
<p>Because the OP specifically mentioned CS…</p>
<p>
</p>
<p>In other words, you’re arbitrarily choosing which pairs of schools that this is the case for. You’re willing to say it about Berkeley and UCI, because that props up Berkeley. But when the same is applied to Berkeley and Stanford, of course not!</p>
<p>I think that in terms of selectivity, Stanford:Berkeley :: Berkeley:mid-tier UCs (whose average GPAs are all about the same). You are free to disagree. (And no, don’t bother bringing up SAT scores, since we’ve already gone over how useless they are.)</p>
<p>
</p>
<p>It’s to get you to admit that yes, average GPA has a lot to do with selectivity. You backed away from this before.</p>
<p>
</p>
<p>I disagree. Statistics capture only some of that natural ability and talent. There’s a reason that high-scoring people are routinely rejected.</p>
<p>
</p>
<p>Well isn’t that a reversal. You were previously arguing against SAT to be able to account for success in college (success as a result of work ethic and natural ability); from post #56 of yours, which you followed up with a quote from gradeinflation.com:</p>
<p>
</p>
<p>Please, stick to an argument - even when that argument isn’t to your advantage anymore.</p>
<p>
</p>
<p>Yes, again that’s the point. Admit rates tell you a little; yield rates tell you a little; things like SAT or GPA tell you a little; and so on. It all tells you a little about the quality of the student body. Which tells you quite a lot about the average GPA.</p>
<p>
</p>
<p>I highly disagree with this point.</p>
<p>
</p>
<p>If you had brought up admit rates before, I would have dismissed them then. You made that argument as though it somehow disproved something I was arguing, but I have never based my argument on admit rates. Ever.</p>
<p>
</p>
<p>It’s a little different from religious arguments involving teapots. When Russell made that analogy, he was specifically referencing the burden of proof for unfalsifiable claims. By any scientific standard, my claims are falsifiable. But of course, one can’t falsify them without the said data from admissions officers. In this case, “absence of evidence is not evidence of absence” should be more like “lack of availability of data does not prove that such a relationship [that Stanford is more selective than any available statistics suggest] does not exist.” But the former is more pithy. :p</p>
<p>
</p>
<p>We could if we had the data. I know for a fact that Stanford’s admissions office has collected data on them. So in other words, you’re agreeing with me that we simply don’t have enough data to make any strong conclusions: we have the SAT, but as you first established, it’s not particularly useful for this discussion. So any conclusions we do make are personal opinion, biased and all. Remember that “agree to disagree” bit? :)</p>
<p>
</p>
<p>I’m not saying that. I’m saying that non-objective factors account for a difference in the student body that a) plays out in the classroom (i.e. academically, as you emphasize), and b) the SAT and other statistics do not account for. It’s not about being a leader or entrepreneur or whathaveyou, but rather what qualities go into such, like tenacity and hard work and commitment and initiative, qualities that will make one more successful in classes - qualities that are not accurately captured by SAT and the like.</p>
<p>
</p>
<p>I’ve brought up gradeinflation.com several times. Either out of laziness or an unwillingness to see the truth, you have refused to look at average GPAs there in relation to selectivity. Yet you have acknowledged that CSU-Fresno is far less selective and that’s why its average GPA is lower. You just don’t want to make that conclusion about Berkeley.</p>
<p>Interestingly, the point about selectivity and GPA is never contested. It’s a well-known fact. There isn’t data on the intelligence of the Stanford and Berkeley student bodies, on their qualities of work ethic and so on - and I have said that many, many times. So when we consider “does Stanford have easier grading?”, it isn’t enough to just look at average GPAs, as you would like to. You simply have to account for selectivity. But we don’t have the data to make strong conclusions. Yet you’re willing to make strong conclusions anyway (as indicated by your first post in this thread).</p>
<p>Which leads me to the point I’ve been making since the beginning: stop making “strong” conclusions with such a simplistic analysis. The data isn’t there for us to decide, so it comes down to personal opinion as to why the two differ (although we each have some statistics that suggest our respective opinion is more reflective of reality). You prefer to think that it’s because Stanford goes easier on students. You are free to believe that. I’m free to disagree. (Seeing a pattern here?)</p>
<p>
</p>
<p>You’re the one who said “the point is that is absurd to think that Stanford students are so intelligent that their intelligence outstrips Berkeley students’ intelligence to the tune of 0.3 points in GPA on average.” Not “absurd to conclude based on the available data.” But “absurd to think.” You can’t blame me for wondering whether the idea is unfathomable to you because you’re incredulous or because you simply hate the conclusions.</p>
<p>The fact that you’ve been arguing tooth-and-nail on something that is simply not “winnable” without more data suggests to me that you really want to cling to this notion that Stanford goes easier on its students, no matter how absurdly you have to apply tenuous facts and statistics. Notice that throughout this thread, I mostly didn’t affirm that the difference in student bodies accounted for all the difference in GPA, just that we didn’t have enough data to say whether it does or not. In other words, I was being cautious about making conclusions, and trying to get you to admit that you are coming to a conclusion without sufficient data. We can both state our beliefs on the discrepancy in average GPA, but I am not pretending I have the hard data that proves that the quality of Stanford’s student body is far enough ahead of Berkeley’s to account for the GPA difference. You are pretending, or at least refusing to admit that your conclusion is mostly opinion and unsupported by available data.</p>
<p>So after pages of arguing with you, I can’t help but think that it doesn’t matter whether we have the data or not: I’m willing to bet you would reject the conclusion about the difference in student bodies no matter what.</p>
I mentioned this in the previous post:
“School Average = [(department 1 average) * (num of students in dept 1) + (dept 2 average) * (students in dept 2)+…+…) / total students.”
While this is not a guarantee that CS GPA avgs would track the school avg, but it is more likely than not that a positive relationship exists, given the mathematical formula.</p>
<p>
</p>
<p>What I said previously:
I didn’t extend that example far enough, so allow me to do so now. The reason that argument weakened your point is because if Stanford’s average GPA is a 3.5 and the CS dept has a 3.0, a hypothetical 2nd department would have a GPA of a 4.0. Now if Berkeley had a GPA of a 3.2, and CS had a GPA of a 3.0 (same as Stanford CS), then a hypothetical dept would have a GPA of a 3.4 - 0.6 points lower than Stanford’s hypothetical department.</p>
<p>
True, but the OP was mainly talking about the two colleges in general.</p>
<p>
I have always agreed that student quality does account for a difference in GPA. The question is whether the differences in student quality between any two given schools are sufficient to cover the GPA gap.
Evaluating differences in academic ability requires evaluating the two specific colleges in question.
This is not a situation where “if it is true for UCI-Berkeley, it MUST be true for Berkeley-Stanford.” That would be a logical fallacy (hasty generalization).</p>
<p>
And why is that? Have you provided a reason they are useless?
You might have gotten the impression that the studies I presented showed how useless the SATs are. That was not actually part of my argument.
I was arguing that:
SATs are the most direct way of measuring academic ability (since HS GPA data is not available),
Studies have shown the SAT has only slight correlative effect on college GPA.
THEREFORE, academic ability has only slight correlative effect on college GPA, in the aggregate.
IMPLYING that grading policies have a greater effect on college GPA, in the aggregate.</p>
<p>The problem with this: “Stanford:Berkeley :: Berkeley:mid-tier UCs” is that the SAT gap is ENORMUS between UCI and UC Berkeley.
UCI has roughly an average SAT of 1760 (calculated from 25th adn 75th percentiles, 2011-12). Berkeley has an avg SAT of 2050. Stanford’s is at 2185. UCI to UCB’s gap is 290. UCB to Stanford’s gap is 135. The UCB-UCI gap is 215% that of the UCB-Stanford gap.</p>
<p>
Just to clarify, you mean college, GPA, right?
Furthermore, even if you are right that such a correlation exists, how can you prove causation? (i.e. selectivity CAUSES higher college GPA).</p>
<p>
Notice that I said “academic performance,” not “whether you will be president of the US, a la George Bush,” “whether you will go on to become a CEO,” “whether you will found a non-profit to solve hunger,” etc. The reason that top schools routinely reject top scorers (a reason you did not provide) is because these schools are looking for the latter type of people.</p>
<p>What HS GPA and SAT score tells a college is whether that student would do well in classes - whether they would be able to academically perform.</p>
<p>Colleges don’t admit people with strong ECs/essays because they think those EC’s/essays are better indicators of academic ability than GPA/SAT, they admit them because ECs are better indicators of “entrepreneurial drive,” “service to community,” and other soft-skill factors.</p>
<p>Harvard/any top tier school doesn’t admit a football player with a straight B record with low academic class rigor because they are confident that he has the soft skills that will ensure he performs well academically. No. They are admitted because they have soft skills like leadership and social skills which often translate into career success, though not academic success.</p>
<p>
See my clarification above. SATs DO account for GPA, but the correlation is weak, meaning that college grading policies play a bigger role (as I argue).</p>
<p>
And can you provide a reason as to why yield is a better or equal indicator of average student academic ability as SAT scores?
I can point to several confounding variables: weather of campus, “prestige”, US News rank, etc. Not to mention the fact that Harvard has a significantly higher yield than schools like Yale, Stanford, and Princeton, despite the fact that academic ability differences between these schools are nil.</p>
<p>
Ok. Now provide a reason.
What does yield indicate? Whether students like a college better; how students “perceive” the reputation and prestige of a college. Where does a direct relationship to academic ability come into the picture?</p>
<p>
Ok.</p>
<p>
Yes, your arguments are falsifiable in the general context of the world. But in the context of this debate, your arguments are unfalsifiable.</p>
<p>
I have highlighted the two operative words - two hypothetical words.
Allow me to show you an example: I could disprove your points if the dean of Stanford and the dean of Berkley jointly wrote an article about how Stanford has easier grading than Berkeley.</p>
<p>SATs remain the best data available.</p>
<p>
You have yet to give me a reason WHY those non-objective factors would create an academic difference andwhy those factors are not accounted for by SAT scores / HS GPA.
Why would tenacity and hardwork not be accounted for by academic statistics, for example?</p>
<p>
False dichotomy. That statement sure sounds great on paper, with its rhetorical flourish, but it’s a highly fallacious statement to make.
I have looked at the site. I don’t see a relationship. There. See how pointless you “evidence” was?</p>
<p>
The only thing I acknowledge is that the academic ability is much lower. I never mentioned selectivity. For your second statement, refer to what I said about the hasty generalization fallacy.</p>
<p>
I don’t get why “selectivity” keeps being interchanged with “intelligence”/“academic ability.” What we’re talking about here is whether or not the difference in academic ability account for the difference in GPA, specifically for Stanford and Berkeley. Selectivity is only relevant to the extent that it affects “intelligence”/“academic ability” (and to what extent it has an effect is one of the subjects of our arguments).</p>
<p>
One pattern I do see is how you keep wanting to “agree to disagree,” which in my opinion is a total cop-out and a pernicious, thought-terminating cliche.</p>
<p>**
Furthermore, regarding the making of “strong conclusions,” allow me then, to qualify my main argument: It is more likely than not that Stanford has easier grading than Berkeley, given available evidence.
**</p>
<p>
Ah. Forgive me for the poor choice of words. In that context I used the word “absurd” in the same context as “reductio ad absurdum” - meaning “unreasonable.” Furthermore, at the very least I backed up that claim with statistics (for example, that Stanford’s avg SAT is only 135 pts higher), and logical reasoning (that both schools are top tier and compete for top students) - whether or not those reasons were necessarily accurate.
The statement you made, however, tried to frame my arguments as one of mere “incredulity,” which I felt ignored the nuances and varied reasons that I provided supporting my arguments.</p>
<p>
When I have to repeat myself, it means this statement really does not say anything new (particularly relevant words highlighted):
“Is this playing the armchair psychologist? This isn’t quite ad hominem, but it is a rather thinly veiled attempt to inject into this debate my alleged “motivations” and my personal thought processes.”
Also, I don’t think it’s good form to use a word right after criticizing me for using that word, although it doesn’t really matter, since I interpret that word as “without reason.”</p>
<p>
</p>
<p>Sadly, a hypothetical situation that is especially misleading because it
1)attempts to deduce my personality (my propensity for self-delusion, apparently)
2)is a self-contradiction, given your entire emphasis on not making strong conclusions without all the data (and this is a strong conclusion that is being made with data about my personal self, nor data on how I have acted in situations where I faced insurmountable, contradictory statistics).</p>
<p>Maybe I go through pages of arguing because I actually believe my points to be valid, actually enjoy arguing, actually enjoy logical dissection, and/or actually enjoy practicing logical fallacy detection.
Or maybe not. </p>
<p>
As I have said before,
Your position is unfalsifiable in the context of this argument/thread, and should be rejected at least until new data arises.
If I qualify my statement as “It is more likely than not that Stanford has easier grading than Berkeley, given available evidence,” (and I do qualify that statement as such), then your objections no longer apply.</p>
<p>When will you understand that no, it is NOT more likely that a positive relationship exists between the overall average and the average for any one of the constituents? Lord this is stats 101. Please, educate yourself.</p>
<p>
</p>
<p>And again, I never said this wasn’t the case. Of course it is, so it doesn’t weaken my argument. To be clear, even departmental average GPAs need to account for student caliber.</p>
<p>
</p>
<p>That’s your interpretation. I saw “especially in the CS department.”</p>
<p>
</p>
<p>We’ve talked about their relatively weak predictive ability. And I’ve mentioned that after a certain point, differences in students cannot be seen in SATs. For example, last year, over 50,000 students scored above a 2100; at least 15,000 scored above a 2200. Not particularly useful if you’re trying to whittle your applicant pool of 37,000 into 2,400.</p>
<p>
</p>
<p>I know that, which is why I brought it up before you could. Yes, the gap between UCI and Berkeley in SAT is larger than that between Berkeley and Stanford, because in the latter case, the utility of SAT has largely run out. That’s why it isn’t useful in trying to find analogous pairs of schools in selectivity here. The degree to which Berkeley is more selective than UCI is about the degree to which Stanford is more selective than Berkeley, if not more so. You don’t see it in SAT, but that doesn’t mean the selectivity difference isn’t there.</p>
<p>By your logic, UChicago and Harvard are about equally difficult to get into because they have the same SAT score; you’d be hard-pressed to make that argument. Harvard could easily have a 2300+ average SAT score, but it chooses not to, because other factors matter more, both for academic and extracurricular reasons.</p>
<p>
</p>
<p>Admissions officers know that GPA/SAT account for only part of the success - academic or otherwise - in college. The rest of the application accounts for the rest of the academic success. It also accounts for out-of-class success.</p>
<p>
</p>
<p>It doesn’t matter whether those reasons went into a student’s decision or not. The only thing that matters is that the yield is high: in other words, it’s pulling in the students it admits - i.e. the strongest students - and not having to resort to its waitlist. It also means that it gets to be one of the first to pick among top students; the only school that Stanford ever loses to in cross-admit battles is Harvard, where it has nearly drawn 50-50. The point is that the yield is indicative of Stanford’s ability to bring in the “best of the best.” </p>
<p>
</p>
<p>Again, all that matters is that the top students *are attending<a href=“which%20partly%20supports%20my%20contention%20that%20the%20quality%20of%20the%20student%20body%20determines%20the%20average%20GPA”>/i</a>. Their reasons for attending are irrelevant.</p>
<p>
</p>
<p>Yup, just head-desk’d. My claims are certainly falsifiable even in the context of this debate. Think of it like this: someone claims that at the bottom of the Mariana Trench, there would be single-cell organisms. Until recent decades, we didn’t have the ability to go all the way down to the trench. Would any scientist have said, “your claim is not falsifiable”? No, they would say, “it remains to be seen,” but the claim in and of itself was falsifiable. In other words, you’re misusing the word “falsifiable,” and trying to apply a quote from Bertrand Russell and in the process making your argument seem not only pretentious, but completely and utterly wrong.</p>
<p>
</p>
<p>And it’s still not enough to parse out the differences in student bodies. So in short, you’re admitting that the best data we have is simply insufficient for our purposes.</p>
<p>
</p>
<p>HS GPA can be manipulated easily, with some high schools having far easier classes. SAT is largely dependent on your ability to take the SAT; indeed, SAT has been shown to correlate almost perfectly with income. Likewise, if you had a lazy student who tests well and gets high grades simply because the classes are easy, he probably wouldn’t do well at Stanford. In short, colleges simply have to look at far more than SAT or GPA to puzzle out how well a student would do in college. That you’re contesting this shows how very little you know.</p>
<p>
</p>
<p>You’ve already admitted the relationship exists with the several schools I already pointed out, so now you’re changing your viewpoint - yet again, when it’s convenient to you and allows you to reject evidence that doesn’t work to your advantage.</p>
<p>
</p>
<p>To be clear, you’re the one who keeps bringing up intelligence. I have repeatedly focused on points like work ethic and natural ability. I posit that they are highly correlated with selectivity. Since we cannot measure work ethic and natural ability, we can try to gauge selectivity to determine just how different these student bodies are.</p>
<p>
</p>
<p>That’s because I accept the realities of available data and have been a few steps ahead of you on every point, since I’ve debated this topic numerous times, heard every argument, seen all the data, and none of it has changed. I was attempting to be nice and let you say “all right, we really can’t conclude much, agree to disagree,” so you wouldn’t continue to make absurd arguments to save face. Apparently, you really want to do that.</p>
<p>I now can see that you only just graduated high school and haven’t even attended Berkeley yet. Once you do, I think your viewpoint on this will shift to be more in line with the other posters on this board who agree that there are large portions of Berkeley’s student body who are rather unqualified (and tend to bring the average GPA down). Even die-hard Cal fans on CC, like RML, agree with that. I won’t bring up why I think you’re hellbent on proving that Stanford is “easier,” but I will state my confidence that you’re likely to agree with my points in this thread after you’ve spent a few semesters at Berkeley.</p>
<p>Hmm…well I always thought a B was a B and an A was an A, I would have never guessed it was more complicated than that. lol.</p>
<p>Anyway, this thread got 4000+ viewed, I wonder what that says…</p>
<p>PS: I agree with Phantasmagoric’s most recent post regarding actually spending some time here at Cal. Actually come to Berkeley, and after your freshman year is over, you should come back to this thread and share any new insight you’ve gained or contest that you were right all along. I’m actually kinda curious to see what happens.</p>
<p>It is indeed sadly far more complicated than that. </p>
<p>To wit, I’ve always been far less interested in the inter-school comparisons of grade deflation, and rather am far more intrigued and vexed by the intra-school grade deflation amongst different majors. Exactly why do certain majors - notably the technical majors such engineering, physics, or math - inevitably tend to be graded harder on average than, say, the arts/humanities majors at the same school? This pattern seems to hold at every school: Stanford, Berkeley, or anywhere else. Regardless of the school, you never hear of students saying that they tried to major in humanities but found the courses to simply be too difficult such that they would likely flunk out, and therefore in order to successfully graduate, they had ‘no choice’ but to switch majors to engineering.</p>
<p>Therefore as an addendum to the title of this thread, I would ask: “Is a B in EECS equivalent to an A in American Studies?”</p>
<p>I believe it lies in the nature of the subjectivity of the grade given. If the grade is based on performance that are less quantitatively measurable, like the quality of an essay vs computing a double integral properly or correctly proofing the Cauchy-Swartz (mispelled) inequality, then I think you will notice that grades tend to be higher in those courses.</p>
<p>I’ve never proven this but perhaps human beings, as a whole, on average, have an easier time improving their abilities in composing or analyzing things of a more qualitative natural than those of a more precise quantitative nature.</p>
<p>There is no spell check when I make an algebraic error on a STEM exam, thus resulting in using that wrong answer in future parts, and if those parts require analysis, a wrong analysis will then be made due to the wrong answer, and all those errors propagate and result in lower grade on that paper or exam.</p>
<p>Before I turn in an essay however, there is time to spell check and time to revise and edit the paper. And in the case of in class essays, I’ve noticed most professors to focus on overall content rather than grammar (as long as there isn’t too many grammar mistakes).</p>
<p>In STEM, we do the same kind of open ended analysis like they do in the humanities, but there is more rules that are based on quantitative factors, and computing those quantitative factors incorrectly will result in incorrect analysis. Perhaps this is less of an issue in the humanities, allowing a larger proportion of humans to do better since our human minds were designed to allow open ended thinking, not to be human calculators.</p>
<p>In the real world however, people can about one’s ingenuity in understanding the concepts of STEM, thus being able to innovate and design properly, since the computation part is done generally by a machine.</p>
<p>Ex. Babak’s EE20N exams focus more on concept but my Math 53 and 54 classes focused more on computation (and I heard that changes in upper div math). So I guess the general feeling is make sure they can do the stupid computation first before we start teaching them the concepts and philosophical aspects of STEM (which imo is the most important part). Like many people can solve differential equations, but do they truly understand what it is that is actually happening, why they are doing what they are doing? When you manipulate a summation or tranform a function into another function, ask yourself “Why am I doing this and what purpose does it serve?” Unfortunately a timed exam doesn’t give much room for this type of thinking, but research and/or the working world does, so I guess it all works out.</p>
<p>While I can agree that subjectivity may indeed be correlated with easier grading, it doesn’t have to be that way. In fact, that’s not the way it is at Berkeley (or Stanford) right now…within the PhD programs. Humanities PhD programs at Berkeley are at least as difficult to complete as are the sci/engineering/math PhD programs and probably more so: tending to take far longer to complete, with a median time to completion of 7+ years (with some students taking well over a decade) compared to ~5 years for the tech majors, and exhibiting notably higher non-completion rates, with far less than half of PhD students ever completing the PhD in many of the humanities. </p>
<p>That is despite the fact that the work necessary to complete a humanities PhD is necessarily subjective and open-ended. You can’t simply slap together a bunch of ill-formed notions regarding the deconstruction of Shakespeare and then demand that Berkeley grant you a PhD in English. You must meet the exacting standards established by your dissertation committee. Like I said, many such students effectively fail in the sense that they never produce a satisfactory dissertation and hence never complete the PhD. </p>
<p>While I’m obviously not demanding that humanities undergrads produce work of the same quality as the PhD students, surely Berkeley (and other schools) could demand higher quality than they do now in order to graduate. They could institute a policy that states that humanities undergrads who fail to meet a particular standard of quality, however subjective, will effectively ‘fail’ in the sense that the department will refuse to grant them a bachelor’s degree by the same manner that the PhD students are forced to comply or fail. Right now, how many undergrads actually fail the humanities? </p>
<p>
</p>
<p>Let’s not neglect the crucial difference between scoring and grading. Scoring in quant courses may indeed be highly mechanical as you had described, for you either completed the steps of the computation/proof, of you did not. But the translation of that score to a letter grade is highly subjective, being based on a curve, which has nothing to do with how many computational steps you correctly completed but rather depends on the ‘meta-characteristics’ of the class itself: namely, the quality of the other students in the class as well as the letter-grade boundaries that the the prof subjectively wishes to assign to certain sections of the score distribution.</p>
<p>Allow me to give you two examples. I know somebody who once scored a 30% on an engineering exam…and celebrated afterwards. Why? Because the mean score was a 25%, and the prof’s translation of scores to grades effectively means that he received an A. Let’s face it - he understood practically none of the material on the exam, getting over 2/3 of the questions wrong. But that didn’t matter. All that mattered is that he knew more than most other students, who understood even less. In the land of the blind, the one-eyed man is king. </p>
<p>For the second example, I recall one guy who scored in the 80’s% on an exam…and was driven nearly to tears and depression. Why? Because the mean was a 95% with a tight distribution and harsh curve such that his score translated to at best a D, and possibly an F. It didn’t matter that he demonstrated that he could successfully compute the vast majority of the questions on the exam. All that mattered is that other students were able to compute more. In the land of the eagle-eyed, the one-eyed man is a pauper. </p>
<p>The upshot is that students in the quant majors cannot only concern themselves with knowing how to perform correct calculations. They also must be intensely worried about their standing relative to other students, for that determines their grades. It doesn’t actually matter how much you know on an absolute standpoint, rather what matters is how much you know relative to how much the other students know. Those students who happen by sheer luck to be members of an unusually talented group of students will earn far worse grades than vice versa. </p>
<p>The most infuriating aspect of grading is that many post-graduation opportunities care little about differing grading standards. Law school and med-school adcoms are notorious for demanding high grades with practically willful disregard for how you obtain them. The same is true for many of the most prestigious international scholarships. The Marshall and Churchill Scholarships specifically state that you must have a 3.7 to even apply. If you have a 3.6 because you took unusually difficult coursework, they don’t care. All they’ll see is that you lack a 3.7. Meanwhile the guy who schemed his way through school by cherry-picking the easiest possible coursework and accumulated a 3.9 is laughing all the way to a top law/med-school.</p>
<p>Yeah I think that system is completely ridiculous, but I guess there is no better alternative since apparently it is somehow wrong to give everyone decent grades if they score according to what the professor believes is an A,B,C,etc as opposed to using a curve system. Maybe the professor doesn’t even know how much knowledge on his exam corresponds to what letter and thus using the curve system is easier? I’m sure there must be a reason.</p>
<p>And that is precisely the crux of the matter. Why is it ‘apparently somehow wrong’ for profs in tech majors to give everyone decent grades when the humanities profs do so routinely? </p>
<p>To be clear, the issue is not with the curve itself. Indeed, often times the curve can actually prove to be helpful. Let’s face it: without the curve, the guy who scored the ‘celebratory’ 30% on his engineering exam would have surely flunked (along with everybody else in the class). The real issue is the culture of grading that predominates in a particular major. Why exactly do certain majors always feel that the worse-performing students must be weeded out through harsh grading when other majors feel no such compunction? If you want to weed out all (supposedly) ‘incompetent’ students, fine, then let’s weed out all incompetent students across all majors. Why single out certain majors?</p>
One can note that instead of actually providing any reasons refuting my arguments, your only response has been along the lines of “it’s obvious that I’m right - it’s Stat 101!” In the hierarchy of disagreement, this would be a “contradiction” - a statement of the opposing case without providing any reasons.</p>
<p>
I didn’t say you said this. I said this. The point is that even if CS GPAs are equal at Stanford and Berkeley, if Stanford overall school GPAs are higher, other departments at Stanford will have even higher GPAs.</p>
<p>
And I did too. The title of the threat is “A ‘B’ in Berkeley is an ‘A’ at Stanford.” OP continued to say “I heard this saying on Cal Day, especially in the CS department. Does it hold any validity?”</p>
<p>The main point is clearly that “A ‘B’ in Berkeley is an ‘A’ at Stanford.” A secondary point is that this is especially true for CS (but the main point, again, is that this is true for the overall schools).</p>
<p>
You missed my point. My point is not whether they are useful in whittling an applicant pool into 2,400. I was questioning your implied assertion that SATs are not the best indicator of student academic ability.</p>
<p>
I’ve tried to number specific parts of arguments that I make.
Not true. As I pointed out before, the SAT gap between Berkeley and UCI is twice the gap between Berkeley and Stanford. As I’ve repeated many times, selectivity is irrelevant (see below)</p>
<p>
By my logic? I never said that. This entire time I’ve been arguing the selectivity is irrelevant to determining academic ability. What my logic actually implies (and not “your” version of “my” logic):
UChicago and Harvard have similar avg SAT scores
Your average UChicago student is as academically capable as your avg Harvard student - meaning they are as good at taking tests, doing homework, writing papers for class, doing problem sets, all the things required to get good grades.
This does not imply UChicago is as hard to get into as Harvard. Admissions into Harvard vs. UChicago often depends on having a good personal story (essay), leadership, community service, etc.</p>
<p>
It’s very interesting to note that you use the two kinds of success in the same phrase, “academic”, and “otherwise.” Furthermore, “out-of-class” success is utterly irrelevant to our discussion of GPA.
You’ve also just made a fallacious appeal to authority (“Admissions officers know…”) - this is hardly a solid logical reason backing up your refutation of my point.</p>
<p>
You have just made two enormous and unsubstantiated logical leaps:
That several confounding variables are irrelevant, even though you’re trying to draw a correlation between yield and student academic ability (so much for the Stats 101 reference)
That pulling in the students you admit = students <em>must</em> be smarter.</p>
<p>And EVEN IF your point about how yield = stronger students is true, again, we go back to the arguments made previously about “academic success” and “out-of-class success”: yield does not distinguish between those two kinds of success, one of which is relevant to our discussion about GPA, and the other of which is irrelevant. This inability to separate out what exactly yield is indicating about the student body renders the use of that statistic useless.</p>
<p>
Your example is not valid and is overly simplistic.</p>
<p>If someone says “there are single-cell organisms at the bottom of the Mariana Trench,” despite the fact that the data that is available indicates there ARE NOT single-celled organisms, scientists would reject that assertion and tell the person asserting it that they do not believe him/her unless s/he brings up some evidence. If that person then replies, “You can’t reject my argument because scientist X has evidence locked up that might prove me right,” and of course no one can talk to scientist X, then other scientists would say “Your assertion that scientist X has evidence possibly proving you right is irrelevant and unfalsifiable.”</p>
<p>
An irrelevant rhetorical flourish. On the hierarchy of disagreement, this would fall on the third-lowest level: “Responding to Tone” - criticizes the tone of the writing (although I acknowledge that you do try to address the substance of the argument earlier in the paragraph).</p>
<p>I’ve added the numbers.
That HS GPA can vary is precisely why I have avoided using it and have mostly mentioned SAT scores.
SAT correlated with income. Ok. You do realize that IQ correlates very highly with income, right? Heck, college admissions into Ivy League Schools/Stanford/Berkeley/even simply going to college correlates with income. It’s quite obvious why - high income parents are likely to be educated and thus raise their kids with a focus on their education - a significant portion of IQ is determined in early childhood.
All this contributes to the argument is the assertion that high SATs are not a guarantee of success. Which I don’t think anyone would disagree with. But that assertion is not very useful for your argument.</p>
<p>
Ad hominem - Attacking an opponent’s motives or character rather than the policy or position they maintain.</p>
<p>
I never admitted that such a relationship exists for all schools. Please quote me if necessary.
Furthermore, saying that “the relationship exists with the several schools I already pointed out” is quite useless because the sample size has been constrained by your choices. I mean, I could draw a correlation between being white and being a Democrat if my sample size consisted of Harry Reid, John Kerry, and Edward Kennedy.</p>
<p>
Your argument conveniently ignores how I mentioned “academic ability,” which includes work ethic and natural ability. Whenever I say “academic ability,” I mean academic intelligence + work ethic + natural ability (it’s just a shorter way of typing out those three phrases). </p>
<p>As for the relevance of selectivity, see my previous arguments. A summary (only a summary) of the primary two arguments I have made are that
SAT scores are a “purer” statistical test of academic ability since you can’t distinguish between the effect that selectivity has on “academic success” and “out-of-class success”/
Selectivity is influenced by several confounding variables.</p>
<p>
There are so many logical errors here that it is hard to lose count:
Merely saying that you’ve “been a few steps ahead of” me, and how you’ve “heard every argument” is logically irrelevant and mere fluff.
Attempting to frame your position as one of pity for my arguments is logically irrelevant, again.
And…
Your arguments again fall into ad hominem and irrelevant discussions of my motivations and behavior.
I could repeat my quote about the armchair psychologist, but I think repeating that for the third time is simply unnecessary.</p>
<p>
Again, these arguments are utterly devoid of logical substance. I skimmed them over and found nothing logically pertinent to our preceding arguments.
It is quite ironic that, despite the fact that we are on an online forum and thus we both have ZERO personal or anecdotal credibility, since we can “say” we “are” anything, you continue to make assertions backed with statements like, “Even die-hard Cal fans on CC, like RML,”. This is why I refuse to resort to these sorts of wild insinuations and rely on logical reasoning that is based off published statistics and common assumptions.</p>
<p>As for your last sentence, I could repeat my quote about the armchair psychologist, but I think repeating that for the fourth time is simply unnecessary.</p>
<p>I will also note, that several of my points went unaddressed in your refutations.
My assertion that academic ability has a weak correlative ability with a college’s average GPA. (An argument with four points to it)
The argument that: “You have yet to give me a reason WHY those non-objective factors would create an academic difference and why those factors are not accounted for by SAT scores / HS GPA.
Why would tenacity and hardwork not be accounted for by academic statistics, for example?”
My assertion that:
If I qualify my main argument as “It is more likely than not that Stanford has easier grading than Berkeley, given available evidence,” your objections are no longer valid.</p>
<p>My arguments do not rely on insinuations of someone else’s beliefs in the present or the future, do not rely on criticizing personal motivations in an online forum, do not rely on irrelevant appeals to authority, and do not rely on several logical fallacies.</p>
<p>**
I do not care about any inferences of your character, your motivations, your beliefs, nor your education. All I know is that I know nothing about you; all I believe is that your arguments are wrong - that’s it. All I believe is that it is readily apparent whose arguments lie on firmer ground.**</p>
<p>Wow, you can go on and on. Based on all the bold text, I’m guessing I touched a nerve or two, so it’s probably best to end this discussion here. Come back when you have a little more perspective on the issue. More data would be great too. :)</p>
Based on all the bold text, I’m guessing I touched a nerve or two, so it’s probably best to end this discussion here. Come back when you have a little more perspective on the issue. More data would be great too. :)</p>![https://en.wikipedia.org/wiki/File:Graham’s_Hierarchy_of_Disagreement.svg[/url]](https://en.wikipedia.org/wiki/File:Graham%27s_Hierarchy_of_Disagreement.svg%5B/url%5D){kind=link}
