For someone who attacks the “generalization” that Stanford’s grading is easier than Berkeley’s, it is quite a psychological leap to accuse Berkeley students of trying to make themselves feel better, and to say that Stanford students work harder than Berkeley students, enough to entirely account for the higher average GPA (of course ignoring for the moment that work ethic has nothing to do with how hard a school’s grading is - if anything, it only has to do with how “just” it is).</p>
<p>In fact, I personally believe that Berkeley could use some grade inflation.
The simple matter is that Stanford has easier grading than Berkeley. I never stated this was a bad thing.</p>
<p>They’d likely still successfully pass and graduate. After all, even in the worst case scenario, while at Cal they could simply switch to one of the creampuff majors (e.g. the ‘Studies’ majors) as the pathway to graduation. There is no creampuff pathway to graduation from MIT or Caltech - everybody has to pass the institute requirements which comprise a slew of rigorous technical coursework.</p>
<p>What I did say is that no matter what major you choose at MIT, you still have to pass the rigorous General Institute Requirements, or you don’t graduate.</p>
<p>^ and Stanford’s financial aid is no-loan, covers 100% of students’ need, and grants on average 80% of the COA (about $42k), none of which can be said about Berkeley. Nearly 80% of the students are on some kind of aid. (Fun fact: for ~94% of the households in the US, Stanford is cheaper to attend than Berkeley.)</p>
<p>flutterfly_28,</p>
<p>
</p>
<p>It’s hard for me to take your ideas seriously when you assert such garbage. The minority students at Stanford are perfectly capable and hard-working. If Stanford “instituted the policies” that Cal has, URMs would do just as well as the rest of the student body. It’s also worth noting that most of these URMs (at least the ones from California) got into Cal, but chose Stanford.</p>
<p>
</p>
<p>You know that isn’t true. Cal gets around the ban on AA by emphasizing socioeconomic factors (‘holistic admissions’), which it touts just as much as (if not more so) Stanford touts its ethnic diversity (“x% is first-generation,” “y% receive the Pell Grant,” etc.). If its student body were more uniform, you wouldn’t see such a large spread in its SAT ranges. Not that SAT scores are truly indicative of the relative strengths of student bodies (e.g. HYPS’s average SAT is 100-130 points higher, but such a difference doesn’t accurately demonstrate how much harder HYPS are to get into). </p>
<p>The fact that you’re contesting this point - and even asserting that Cal’s student body is more uniform - is surprising. Most Cal students concede this point, as far as I’ve seen.</p>
<p>terenc,</p>
<p>
</p>
<p>Notice that I emphasized, multiple times, that it’s “on average” (more uniformity), and also said that some Berkeley students are harder-working than some Stanford students, and vice versa. Inferring the motivation to assert that “Stanford students have it easy” isn’t a psychological leap at all - it’s a pervasive, repetitive sentiment that I’ve encountered among Cal students innumerable times. Other common motifs in the Berkeley lore: Stanford is a country club, Stanford students are all rich legacies, Stanford students waste money on a $50k education, Stanford students are coddled and will fail once they get to the real world, etc.</p>
<p>
</p>
<p>You’re looking only at the average GPAs (which, by the way, are not recently available for Stanford, so any comparison you try to make is by definition unsupported). Grades don’t exist in a vacuum, and it is essential to account for the rigor of the standards themselves in assessing the “laxness” of grading. Thus, the nature of the student body is important to consider if you want to make a claim that an A at X university is a B at Y university. </p>
<p>My point is that you cannot draw from GPA statistics the conclusion that Stanford has “lax grading,” since grading is inherently tied to the rigor of the standards and in turn to the abilities of the student body. Stanford could actually have harder grading than Berkeley, but the student response to this standard is such that they tend to do well. “A B at Berkeley is an A at Stanford” is just as plausible as “a B at Stanford is an A at Berkeley,” depending on the rigor of the standards.</p>
<p>In short, you and I simply have different conceptions of what ‘grading’ means. Unsurprisingly, this difference is the crux of every debate on grade inflation/deflation (throughout academia, not just on CC), and is unlikely ever to be resolved.</p>
<p>flutterfly, I remember your posts from before railing non-stop on Stanford’s AA and its supposed bias against Indians/Asians. It’s been a few years - time to let go of this absurd notion, esp. when the reality is that ethnicity is only a tiny consideration. Stanford has shifted heavily toward considering socioeconomic factors, just like Cal. For example, see this FAQ:</p>
<p>
</p>
<p>This is a stark contrast from its previous statement on it, which outright said “yes.” The dean of admissions, Rick Shaw, has repeatedly emphasized Stanford’s shift in philosophy toward background/socioeconomic factors.</p>
<p>Even if AA were a significant part of its admissions decisions, it’s insulting to every URM at Stanford to assume that these students are less capable. It’s simply not the case. It’s also a stretch to assume that elite private universities like Stanford aren’t able to attract a critical mass of equally capable minority students (with nearly 40,000 applicants from across the country/world, this isn’t a problem).</p>
<p>I agree with almost every point you’ve made on this thread. Berkeley classes are not harder than Stanford classes. Stanford is much harder to get into than Berkeley, and Stanford students OVERALL are much stronger and more uniform than Berkeley students. Berkeley admits a lot of students that are seriously unqualified. I definitely also agree that a lot of Berkeley students say that Stanford is easier to make themselves feel better about their poor GPAs at Cal. </p>
<p>But I simply cannot agree with you about the affirmative action issue. I don’t know when Stanford changed its policy on AA, but I am certain AA was practiced with regards to admissions from my high school the year I graduated (2010). </p>
<p>For background, 11 people from my high school were admitted to Stanford that year: 5 URMs, 3 legacy’s, and 3 students that I would call “academic superstars.” 2 of the URMs were NOT admitted to Berkeley, while the other 9 were. I know all 11 students personally, and went to school with them for years. Meanwhile, around 75 students from my high school got into Berkeley in 2010, and around 40 of those are currently attending, including myself.</p>
<p>I can say for sure that the URMs who got into Stanford were clearly less qualified than MANY of the students who were rejected. The URMs were not TERRIBLE students or anything, and they were definitely better than a decent number of the Berkeley admits, but they were not in the league of quite a few of the students who were rejected from Stanford. Several of them struggled heavily in classes that I took with them (but they got decent grades because my high school heavily inflates GPAs). Their extracurriculars were good, but not outstanding. I didn’t read their college admissions essays, but I DID proofread some of their essays in AP Comp Lit, and they were ordinary at best. And they were all mid-high income students (very financially well-off). They got in because they were URMs with decent stats, decent extracurriculars, and decent essays. I don’t see any evidence against that claim. </p>
<p>And before you suggest that I am just being bitter, that is really not the case. I am thrilled with my experiences so far at Berkeley. I have a 4.0 after two years and I’m doing interesting research with plenty of funding. I couldn’t really ask for more. And I am quite happy with affirmative action overall, too. I just think affirmative action does exist at Stanford, or at least it did 2 years ago.</p>
<p>singh2010, thanks for the courteous insight. To be clear, I’m not suggesting that AA is absent from Stanford’s admissions, which would be an absurd assertion. But as I said before, flutterfly_28, like so many others, is overestimating the degree to which ethnicity matters. Anecdotes such as yours unfortunately can’t really tell much; you can’t be certain that ethnicity was important, or even relevant, for those URMs who were admitted. It’s very easy to be wrong about a student’s family’s assets, even if it seems like they’re middle class. Unless you know all about their background - their incomes, their parents’ education, all their personal circumstances - it’s impossible to tell. It doesn’t make sense to extrapolate from their AP essays to their college essays, which are likely far better and could have had very significant content that you were unaware of. You also didn’t see their recommendations, which are important in Stanford’s admissions. In short, you have truly only a glimpse of these applicants and are attempting to judge their admission as “decent applicant, but URM, so admitted” - which is impossible, considering how much hard work/time/effort are part of an admissions officer’s job. It’s just not that simple.</p>
<p>As I said before, Stanford’s emphasis on background is able to explain most URM cases, wherein their race becomes largely irrelevant compared to their background. Berkeley and UCLA are perfect examples of how an emphasis on socioeconomic factors (‘holistic admissions’) can maintain ethnic diversity. I’m not sure when Berkeley started doing this, but since UCLA began following Berkeley’s example in 2007, the (domestic) enrollment of black students has increased 45%, Hispanic 18%, and Native American/Hawaiian/Alaskan 33%; for comparison, the (domestic) enrollment has increased only 3% during this time. 23% of UCLA’s student body today is URM, compared to Stanford’s 25%. Berkeley’s hovers between 15-20% (currently 17% of domestic students). Does this mean that UCLA is practicing AA more? Of course not.</p>
<p>The point here is that it is entirely possible to maintain high enrollment of URMs without paying any attention to ethnicity, and in Stanford’s case, very little. I know, this runs counter to the CC wisdom, but that’s because much of CC refuses to pay attention to the shifting reality of elite admissions. Another reason is that many students prefer to believe that they didn’t get in because they weren’t URM, or a legacy, or a recruited athlete, or a resident of a rural state. It’s easier and more comfortable to think this than to face the reality that a group of other students simply beat them out in admissions.</p>
<p>I’m not saying you’re part of this group (your post doesn’t indicate bitterness), but that this is a common sentiment on CC, continually perpetuated by such bitter students. Every time you see someone ranting and raving about this topic, a glance at their past posts usually reveals rejection from elite private schools. It’s tiresome, esp. since the extremity of their ideas is simply off-base and out of touch with reality.</p>
<p>Well, actually, phantasmagoric, on the particular point regarding AA, I’m afraid I have to agree with singh2010 and (surprisingly) flutterfly_28. </p>
<p>Now, to be clear, I agree with you that nobody here can ascertain the intimate and intricate details of any one particular admission decision. But I don’t think that’s relevant. What is relevant is what occurs at the aggregate, statistical level, that is to say, within the wide population of applicants, does Stanford in fact place some non-zero admissions weighting upon ethnicity? Regarding that question, I think that you would agree (and in fact, seemed to have already agreed) that Stanford indeed does. Indeed, on this point, there seems to be no dispute. </p>
<p>The point of contention then seems to be how much does Stanford weight ethnicity within its admissions practices, and put more precisely, how much does ethnicity matter entirely separately and residually from the ‘holistic-background’-oriented variables that you cite (or to be truly precise, how much admissions variance associated with ethnicity is orthogonal and therefore non-covariant with those other cofactors). I notice that you have accused flutterfly_28 and others of “overestimating the degree to which ethnicity matters”, or that Stanford pays “very little” attention to ethnicity. </p>
<p>Fair enough. If you truly really believe that Stanford indeed pays “very little” attention to ethnicity per se, then perhaps you should then declare that Stanford ought to pay zero attention to it. After all, by definition, there must be ‘very little’ difference between placing “very little” weight upon particular variable, and placing no weight upon it at all. Such a move ought to then similarly have “very little” impact upon future admissions decisions. </p>
<p>The benefits that Stanford would reap would be substantial. By entirely abolishing any influence that ethnicity alone has upon the Stanford admissions process, and perhaps adopting the UCLA-style holistic admissions process that you cite, Stanford would align itself both with the political sentiments within the state of California (as evidenced by the solid 10-point victory of Prop 209, with no credible attempts to repeal on the horizon) and probably eventually with the nation as a whole, as race-based AA seems to be losing support nationwide over time. Stanford would also immediately quell the ‘tiresome bitterness’ and the ‘ranting and raving’ that you cite is common amongst Stanford rejectees who perhaps wrongly blame ethnicity for their rejection. {In contrast, while those rejected from Berkeley or UCLA are surely also likewise bitter, at least they’re no longer blaming race for their rejection.} Perhaps most importantly of all, it would help future Stanford URM’s who would no longer be insulted and questioned as “asterisk-Stanford-students” over any controversy regarding how exactly they were admitted in the first place. Again, that stands in stark contrast to post-Prop-209 URM’s at UC who never have to feel defensive about their admissions process. </p>
<p>In short, why exactly should Stanford continue to suffer the slings and arrows of political outrage for maintaining the controversial admissions policy of ethnicity weighting, if doing so, as per your words, has “very little” impact upon admissions anyway? Like I said, to maintain its commitment to opportunity for the underprivileged, Stanford could simply declare that it is shifting to the far less politically controversial UCLA-style holistic admissions process that abolishes race as a standalone variable. So why does Stanford persist? Seems to me that the only logical conclusion is that the admissions impact of race is indeed far larger than the “very little” impact that you assert. </p>
<p>{To be clear, none of this post should be construed to state that race-based AA is right or wrong. That’s ultimately a normative, philosophical question that is beyond the scope of this thread. Whether I personally support race-based AA, or whether others should do so is likewise beyond the scope of this thread. I’m simply discussing why Stanford, or any other university, should continue to do so, particularly if its impact is truly “very small”.}</p>
<p>sakky, I don’t want to turn this into a full-on debate about the merits of socioeconomic vs. race-based AA (which is tiresome and not germane to the discussion), but I’ll explain my thoughts on why the two aren’t incongruous. Some of this has been corroborated, in a roundabout way, by admissions officers at Stanford and elsewhere.</p>
<p>It isn’t the logical conclusion that Stanford, or any other school that practices race-based AA, should simply ignore race altogether if they’re already practicing mostly socioeconomic AA. You assert that Stanford stands to substantially benefit simply by eliminating it, and I agree with those benefits. But in the end, tracking race is still important enough to maintain in the face of those benefits. It allows them to know, year over year, how many of each ethnicity is applying; what the socioeconomic makeup of those applicants is; how many of them are admitted (regardless of AA); and how many of them attend. This allows them to shift their recruitment and outreach (if, say, one year there are substantially more well-to-do black students applying, with fewer low-income black students applying, as opposed to previous years; this would indicate weaknesses in their outreach), but also allows them to monitor how important race-based AA is in any one year and, in turn, how important socioeconomic AA is in any one year.</p>
<p>This is why I doubt that Stanford and its peers will be ignoring race completely any time soon. Even though I’d say in most recents years, for most applicants, it’s a small - or even irrelevant, in some cases - part, it’s still essential to keep track of. Outreach and yield temper the degree to which race and socioeconomic factors matter in any one cycle. But note that that doesn’t mean that Stanford will necessarily increase its race-based AA to reach a target enrollment. For example, some years the black enrollment in the freshman class drops to 8%; other years, it’s gone as high as 12%. For Hispanic students, it’s gone from a high of 17-18% to a low of 13%.</p>
<p>In short, the importance of both kinds of AA varies from year to year, and from applicant to applicant, depending on outreach, yield, and the non-racial characteristics of the applicants. Despite this, Stanford and its peers are able to do significant recruitment so that race is generally a small or even irrelevant factor. </p>
<p>Just look at the change in elite admissions over the past 20 years: applications to HYPSM-type schools exploded during this time, beyond the ‘echo boom’ numbers, largely because of more national outreach. These schools are far less regional now than they were then, which has completely changed how race is viewed in admissions. This also explains why Berkeley and UCLA have struggled: they aren’t more national, even when you include the recently higher OOS % (who are typically well-to-do, i.e. their only form of AA is moot). While their enrollment of Hispanic students is high, their black student enrollment is low (California has a much smaller black population). Of course, the downside of having a more national student body is the far higher complexity of outreach to and yield from numerous populations that vary along racial, socioeconomic, and geographic lines. So while HYPSM-type schools would have an easier time - relative to Berkeley and UCLA - enrolling a racially diverse class with only socioeconomic AA (given that they aren’t constrained by the demographics of their home states), it still wouldn’t make sense to eliminate it completely. You might consider it “insurance in case we mess up.”</p>
<p>To be honest, I don’t think that this is the crux of the ‘ranting and raving’ about AA for posters on CC (and perhaps for certain posters on this thread). I think it mostly has to do with the treatment of Asian applicants, which is unsurprisingly also explained by socioeconomic AA - but this is yet another way in which students will ignore reality and reduce admissions to a simplistic, zero-sum game in order to pad their egos. I think this anger though does turn into a hatred toward AA in general.</p>
I hope that you see the irony, as I do, in your conclusion that Berkeley students “on average” tend to declare that “Stanford is a country club, Stanford students are all rich legacies, Stanford students waste money on a $50k education, Stanford students are coddled and will fail once they get to the real world, etc.”</p>
<p>
Of course, this statement ignores and does not even attempt to refute any of the details I mentioned here and other details (A+ = 4.3 at Stanford, Berkeley EECS grading guidelines, etc., for example):
</p>
<p>
“Grades don’t exist in a vaccum.” OK. Agreed.
What exactly is meant by “rigor of the standards” is left unclear. If you are referring to the rigor of the grading policy, that renders the entire paragraph tautological. I assume it refers to the strength of the student body, as the next sentence suggests. I do agree that Stanford has a somewhat more intelligent student body (on average) than Berkeley’s student body.</p>
<p>Your argument seems to rest on the implication, however, that Stanford’s stronger overall student body fully accounts for the difference in average GPAs.
Firstly, there is little basis for the idea that Stanford students are harder working than Berkeley students. Secondly, there is little reason to believe that Stanford’s slightly more intelligent student body accounts for the quite substantial difference in average GPAs (roughly 0.3).</p>
<p>Thirdly, the actual grading policies of Berkeley and Stanford have substantial differences. Certain departments at Berkeley will set a specific target for the average (refer to the EECS grading guidelines I referenced), which is relatively inflexible. Only X% of a given class will get a certain grade, even if it was the most brilliant class ever. At Stanford it does not appear that such an inflexible rule even exists, or if it does, is set with such at such a low grade.
In the sense that no matter how hard a class works, only a certain % will get an A, Berkeley’s grading is tougher than Stanford’s, since Stanford’s grading is more flexible with where the grade distribution is centered.</p>
<p>I already refuted the difference in grading scale, but you apparently didn’t understand it. Yes, an A+ is a 4.3 on Stanford’s scale, while it’s a 4 on the Berkeley scale; but an A- at Stanford is a 3.7, while it’s a 4 on Berkeley’s scale. See how the +s and -s balance each other out? As I suggested before, this argument works out in your favor only if there are more +s than -s. In fact, it’s just as likely in my favor as it is in yours: who’s to say there aren’t more -s than +s? If so, this difference in grading strengthens my argument. If not, then it strengthens your argument. Since we don’t know which way it goes, it’s a moot point. Unless you can pull a rabbit out of your hat and support this with data, it’s probably a good idea to drop this point.</p>
<p>I cannot comment on Berkeley’s EECS grading guidelines because the data for such at Stanford is not available. Yet you’re trying to draw conclusions about it anyway.</p>
<p>
</p>
<p>Yes, “rigor of the standards” is unclear. That’s precisely why I said these debates will never be resolved. Inevitably, the debate on grade inflation reaches an impasse on whether the grade distribution or grade-independent rigor matters more. For example, compare Sonoma State’s intro programming course to Berkeley’s 61A or Stanford’s 106A. Do you think they’re equivalent in difficulty? I don’t think they are. The rigor of the course is set to the abilities of the students.</p>
<p>
</p>
<p>Actually, what I’m saying is that you can’t conclude that Stanford has “lax grading policies” based simply on this or that GPA statistic (which isn’t available, anyway), without accounting for the abilities of the student body. It’s entirely possible that, even after controlling for the abilities of the student body, there’s still some grade inflation. Notice that my second post in this thread said “exactly” in response to texaspg’s statement: “This statement has merit if you can look at grade distributions in a class and show that Stanford awards As disproportionately compared to Berkeley after eliminating the bias for caliber of the students.” [Addendum: of course you’d also need to account for differences in the concentration of STEM majors, the socioeconomic makeup of the student bodies, etc.]</p>
<p>I’m just saying that your simplistic view of the issue is leading you to come to conclusions not supported by any evidence available. </p>
<p>
</p>
<p>I haven’t asserted anything otherwise, but rather I’m quelling this idea about “a B at Berkeley is an A at Stanford,” which indicates that Stanford somehow goes easier on you in grading. It doesn’t. It holds students to the same standards - it’s just that more students are able to meet those standards.</p>
<p>Think of it like this: Stanford has about 1/4 the # of undergrads. So take the top 25% of the students of the student body at Berkeley (using metrics not dependent on their GPA at Berkeley - say you got the admissions office to give you a list of the top students, in the theoretical possibility that only those 25% were to have been admitted and to matriculate). Then, of those that are taking CS 61A, look at how they do relative to the rest of the class. Do you think that these students would be disproportionately higher-performing? Of course they would. Remember, in this case all students are being held to the same standards, so no group has a more “lax grading policy.” </p>
<p>Now pretend that there’s a hypothetical CS61A class containing only these top students, and they’re being held to the same standards as the CS61A class that runs the gamut of all students, which of course results in higher grades on average [this is the functional equivalent of the previous example]. You’re essentially saying that the resultant grade distribution in the former case is evidence of “lax grading.” But that completely ignores the fact that, in the latter case, you simply have a larger number of students who are not on level with those in the former [again, according to metrics outside of their Berkeley GPA], which gives the appearance of more difficult grading (the top grades are more rare when the denominator is a vast array of students who perform at a lower level). Yet both these groups of students are being held to the same standards. How can one of them be “lax”?</p>
<p>Yes. the grade distributions are different. But that doesn’t mean that anything is “lax.” Again, as I said before, this is the difference you and I have here: in trying to decide whether grading is “lax” or not, I am not looking merely at the grade distributions, while you are. I am considering the grading in relation to the rigor of the standards the students are being held to. I have no way to quantify this (and indeed no one does, hence why this issue still stands today). I am making only the assertion that the quality of the student body is in some way related to the standards that professors hold them to. I’ll add that though I have no statistics on this point, this is a frequently asserted argument, e.g. a Stanford professor explained in a debate about grade inflation that the same class he’s taught over the past 20 years has become harder and harder as the students became markedly more able, and he has found it difficult to make the class harder than it already is. </p>
<p>Another piece of evidence that supports my point: there is an extremely high (negative) correlation between selectivity and average GPA. What do you have to say about this? Why is this the case?</p>
<p>
</p>
<p>And that, we’ll have to agree to disagree on. [To reiterate; I’ve qualified the above statement many times and in many ways, so no one can misconstrue my assertion on this point.]</p>
<p>Hmmm out of curiosity, where did you get the info about A- being a 4.0 in Berkeley? I’m in L&S (perhaps there’s a diff between L&S and engineering?) and got a 3.7 for an A- last semester.</p>
An A- is a 3.7 at Berkeley. The GPA comparison stands.</p>
<p>
My conclusions rest on the fact that the guidelines rigidly set an incredibly low grade average (B-) accross the board for upper division classes; I surmise (and I acknowledge that this is merely an inference, albeit an inference that has some basis in fact) that such a rule does not exist at Stanford because there is no such published guideline (as there is for Berkeley).</p>
<p>
</p>
<p>You are saying (your assertion):
Student body abilities vary, thus you cannot draw any conclusion as to whether Stanford has lax grading.</p>
<p>Having established (on my end) the difference in grading distributions, in order for your assertion to negate my argument that “Stanford has more lax grading than Berkeley,” the following statement that goes unsaid but is implied in your argument must be true:
Stanford’s higher student body ability accounts completely for the difference in grading distribution.
That is, Stanford’s stronger student body accounts for the difference.</p>
<p>I understand that your argument is one of “insufficient evidence.” However, for your argument that we have to eliminate the bias for caliber to be a valid counterargument, it is necessary that the bias in caliber fully account for the difference in distributions.</p>
<p>
The fatal flaw in this analogy is that if you took the top 25% of students at Berkeley and had these top students in CS 61A, the average grade would be a B-, as imposed by EECS grading guidelines.</p>
<p>
Eh… I previously shared an anecdote in which a Berkeley professor who had taught the equivalent class at Stanford had admitted that students getting a B in that class would have gotten an A at Stanford.</p>
<p>^ cross-post - the above was a reply to dacheesefreakk.</p>
<p>terenc,</p>
<p>
</p>
<p>You haven’t actually established this…</p>
<p>
</p>
<p>No. I never said “completely” - again, I’m pointing out the flaws in your reasoning and demonstrating a very plausible - and likely - reality. Whether this point accounts for all differences remains to be seen. </p>
<p>
</p>
<p>For the last time, yes, we know about the grading distribution. The EECS grading guidelines apply to the general CS 61A class, which runs the gamut of students. That’s why I proposed a hypothetical class composed of the top 25% of students. In reality, this portion of students is integrated with the general class, wherein these grading guidelines apply. These students do far better. That’s the point. Within this hypothetical group, the grading guidelines don’t apply. That’s why I posed two different possibilities: one in which you looked at a normal offering of CS61A (that contains all students, but you only considered the top 25%) and one in which you looked at two different classes (one that has the top 25%, and one that has all students, including the top 25%). My hope in this analogy was that you’d see why the grade distribution is different because of the difference in student ability - despite the fact that both are held to the exact same standard. In short, you’re (incorrectly) trying to apply this real fact (the grading guidelines) to a hypothetical situation, and in the process missing the point - yet again.</p>
<p>
</p>
<p>Fair enough. For the record, though, unlike you I wasn’t basing my entire argument on an anecdote, but simply using it as an example (hence the use of “e.g.”) of a side point under my larger argument.</p>
<p>But that’s not what you were talking about earlier (or at least, it’s not what I thought you were talking about). It’s one thing to track/record racial information for future data analysis to enable post-hoc assessment of the impact of various admissions policies upon the student body. It’s quite another thing to actually utilize racial information, even if the impact is “very little” as per your assertions, within the admissions process on an a-priori basis. For example, if Stanford admitted students used solely the socioeconomic factors that you cited and it turns out after the fact that those students just all happened to be URM’s, that’s far less controversial than if Stanford specifically embedded URM-status within the admissions process itself. </p>
<p>As far how a school could record/track such information without actually using it within its admissions process, that’s actually quite elementary. Universities routinely quarantine and segregate their student databases for privacy purposes. For example, right now, many universities provide student health-care insurance, with some universities even having on-campus clinics (e.g. MIT Medical Center), and therefore likely have numerous student medical records. But those records are shielded as a matter of law from the prying eyes of school administrators. Surely Stanford, given its vast resources, could easily design an admissions process that tracks race but shields that information from the adcom. </p>
<p>
</p>
<p>And now we’re getting to the brass tacks, for now you’re arguing that Stanford (and the other elites) should continue to use race, even if as you assert ‘only slightly’, as an insurance policy. Translated to ‘stat-speak’, I take your statement to mean that it is important that Stanford continue to capture the ‘orthogonal residual term’ that is absorbed by the URM-status term that its current socioeconomic factors fail to capture. </p>
<p>But to that, I would then ask, why can’t you fully capture that residual? Why do you need such insurance? Again, the schools that we’re talking about - HYPSM - have immense resources that most other schools cannot ever dream of having. And with the possible exception of MIT, the most difficult step towards graduating from any of those schools is simply being admitted in the first place. Given the crucial role that that admissions office plays, you could think that schools with such immense resources could design a new admissions process that truly does capture all of the relevant socioeconomic factors with no reference to URM-status, or renders that residual URM-status to such a tiny value that it truly can be safely dropped from the process. That actually seems to be a fairly trivial statistical problem to solve. </p>
<p>
</p>
<p>Maybe so, but whether we agree with it or not, the reality is that nobody enjoys the perception that they’ve been judged by their race, even if only mildly. Whether we’re talking about the criminal justice system, the job hiring process, or, yes, the college admissions process, the role of race is a highly emotionally sensitive topic. This is particularly so in the United States given its painful history of race relations. </p>
<p>And it is the nature of highly emotionally sensitive topics that otherwise highly intelligent and rational people often times will not behave entirely logically. As an analogy, I might be able to take the most peaceful, rational and logical person in the world and aggravate him enough to want to punch my lights out simply by loudly and provocatively insulting members of his family in front of him. Nor would it absolve me of responsibility by simply declaring that his desire to fight me is simply ‘irrational anger/hatred’, and condescending claims that he’s simply ‘ignoring reality’ would surely only inflame the situation. The bottom line is that if I don’t want to start fights, then I should refrain from using fighting words.</p>
<p>Similarly, if Stanford persists and insists on using the deeply politically controversial and emotionally-fraught measure of race within its admissions process, well, then they will inevitably suffer the consequences of applicants suspecting that their rejections were because of race. Such a reaction might well be emotional, but that’s the nature of the beast. If you don’t want emotional reactions, then stop pressing people’s emotional buttons.</p>
[To reiterate; I’ve qualified the above statement many times and in many ways, so no one can misconstrue my assertion on this point.]</p>