<p>Xiggi, you took the words right out of my mouth. Great post.</p>
<p>xiggi:</p>
<p>you are correct in the aggregate. But this discussion is generally on top schools which require both strong math/science skills and strong English skills for admissions. Why does HYS award two-thirds A’s in hume/lit and one-third in STEM? Why does Cal Berkeley flunk out Engineers – who had top math grades to get accepted, while “Studies” majors can cruise by with the PC responses?</p>
<p>
</p>
<p>Actually, that’s not ‘attrition’, at least, not the way that we’ve been defined it. Attrition consists of people who start in a certain major and then switch to another due to poor grades (or the fear of such grades). </p>
<p>What your link is talking about is the fact that some students don’t even try out humanities majors in the first place. However, how many students start out in the humanities, then receive failing grades that forces them to flee to refuge in engineering or the sciences? I would say that the answer is pretty close to 0%. {Heck, how many humanities students, even the worst ones, receive failing grades at all?}</p>
<p>
</p>
<p>Actually, I’m afraid that your analysis does not hold, and indeed the data diametrically falsifies your analysis. After all, if it was really true that students self-selected themselves to classes in which they would perform well, then you should see no difference in grading schemes, right? After all, the students who are adroit in humanities would select humanities majors, the students who are adroit in engineering/science courses would select engineering/science courses. Since everybody would have selected the courses/majors that correspond to their strengths at an equivalent rate, grading distributions should be similar across all disciplines, right? So then why does such a large disparity between the grading in humanities/arts and STEM courses exist?</p>
<p>If self-selection is indeed the only answer, then that self-selection must be anomalously asymmetric in nature. Far more students who are humanities-oriented must be “mis-self-selecting” into STEM, relative to the number of STEM-oriented students who are “mis-self-selecting” into humanities. That’s why humanities courses tend to give out few if any failing grades, relative to the STEM courses. But that only raises the question: why would students asymmetrically mis-self-select into STEM? Why wouldn’t any mis-self-selection be evenly spread, or even be asymmetric in the other direction? </p>
<p>The argument for pure self-selection as the only answer becomes even more implausible when you consider the fact that the grading disparities seem to exist across the whole country. Pop quiz: name me some schools where the humanities/arts majors are considered to be the most harshly graded and most demanding majors, such that many students try them, perform poorly and henceforth flee to the ‘easier’ engineering and science majors instead? It’s hard to think of even a single such school. Hence, if self-selection was the only answer, why would that self-selection not only be asymmetric, but seem to exist consistently across all schools? Put another way, why do many humanities-oriented students at every school consistently err in choosing unsuitable STEM majors, but STEM-oriented students at every school never really seem to err in choosing unsuitable humanities majors? </p>
<p>Hence, self-selection is not the explanatory factor. Indeed, if anything, I suspect that self-selection is actually a moderating factor. I can agree that students will probably receive a poor grade if forced to take a course for which they are unsuited. But what is the definition of a ‘poor grade’? In the humanities courses, that probably means a B, or at worse a C. In other words, if we forced all engineering/science students to major in humanities instead, some of them might receive poor grades, but they’d still probably pass. They’d still graduate. {And I suspect that many more such students would actually receive higher grades than they are currently receiving in their STEM courses.} On the other hand, poor grades in STEM courses can easily mean failing grades, particularly in the weeders. In other words, many humanities students who were forced to take STEM courses would probably flunk out entirely. That dramatically demonstrates the difference in grade distributions. </p>
<p>I remember many engineering students who were close to honors GPA thresholds would would load up on with as many humanities courses as possible in order to boost their GPA above that threshold. I can’t think of a single humanities student who loads up on engineering courses in their final year in order to boost their GPA. Heck, I know one engineer - in his final semester - who took a single engineering course (the last he needed for graduation), along with a bunch of ‘elective’ humanities courses, and he had to put in more work in that lone engineering course than in all of those humanities courses combined…and received a lower grade in that engineering course to boot. I wonder if anybody has the opposite anecdote: a humanities student taking a single humanities course in his final semester along with a bunch of ‘elective’ engineering courses, and was forced to work harder in that lone humanities course than in all of those engineering courses combined, while still receiving a lower grade. {Heck, I think it’s difficult just to find a humanities student who would ever take any elective engineering courses at all, let alone finding one who actually receives higher grades for less combined work in those engineering courses than in his humanities courses.}</p>
<p>
</p>
<p>I seem to recall a link that showed that Berkeley’s students in the College of Engineering had higher average SAT Verbal/CR scores, and similar scores in SAT Writing compared to students in L&S. I wish I could find that link. That seems to suggest that engineering students may be just as qualified in the humanities as the L&S students are.</p>
<p>
</p>
<p>And I think that’s the core problem. Law school (and other professional school) adcoms should stop using absolute GPA’s to determine admission. Ideally, what they should do is compare discipline GPA’s against each other, and not against other disciplines (hence electrical engineers who apply to law school would be compared to other electrical engineers). </p>
<p>But professional school adcoms apparently recalcitrantly refuse to take even this simple step. So I think that undergraduate programs should consider responding by providing ‘cleansed’ transcripts for its STEM students that effectively provides them with a GPA boost to increase their competitiveness. This cleansed transcript would be the one that is sent to LSDAS, AMCAS, and other such professional school adcom clearing houses for which the admissions consistently refuse to take grade disparities amongst majors into account. For example, the cleansed transcript might simply delete all STEM grades that are below a certain threshold (i.e. below a ‘B’), or convert them to P (Pass) grades. Or the students could be given the choice of nominating X number of courses whose grades could be deleted from their cleansed transcript. {Note, the ‘real’ transcript’ would not be affected in any way, the only transcript that would be affected is the ‘cleansed’ one that is sent to LSDAS/AMCAS.} </p>
<p>Outrageous? I don’t think that’s any more outrageous than the practice at MIT of having two versions of your transcript, the ‘internal’ transcript that contains all of your grades, and the ‘external’ transcript that is sent to employers and grad-school adcoms that withholds certain pieces of information, notably any failed freshman grades. In other words, employers and grad-schools have no idea that an MIT applicant might have failed certain courses in their first year. That external transcript is effectively a ‘cleansed’ transcript. If it’s not outrageous for MIT to provide cleansed transcripts to boost the prospects of its students, why can’t other schools do the same? </p>
<p>If professional school adcoms persist in misinterpreting and abusing information that is provided to them about their applicants, then I think it is entirely justified for undergrad programs to deny them that information. Sometimes you gotta fight fire with fire.</p>
<p>
I saw that and I’ve been looking for that too since I lost it on my bookmark. </p>
<p>Actually, CoE/Chem students have higher SAT scores even for both critical reading and writing, though the gap wasn’t as wide as the gap for math.</p>
<p>sakky, I know you’ve attended HBS. Can you post the schools with the most number of grads at HBS? I’m sure a lot of lurkers here would be very thrilled to see the numbers. Thanks.</p>
<p>
</p>
<p>If that is true, xiggi, then does it mean that Adcoms are really selecting most students who are “better suited” to hume/lit curriculums? Do you believe that Stanford’s adcoms actually go out of their way to select more hume/lit prospies than STEM prospies? Same with HYP? What about Pomona? Or, are the STEM prospies lying on their apps, by marking pre-science when they are really “better suited” for grade-inflated majors? </p>
<p>Is any of the above logical?</p>
<p>The real issue is the “weed-out” mentality, as alluded to in one of the stories cited in this thread. Everyone says we need more STEM students, and then as soon as they reach college, it’s all about “weeding out” half of those interested in the field in the first place.</p>
<p>We don’t just need “OMG THE BEST!!!1one!eleven!” scientists. We actually need lots of scientists, some of them future Nobel winners, some of them not. That doesn’t mean we accept people who truly can’t hack it. But it does mean we encourage people to persevere in math and science even if they’re not capable of reaching the top of an incredibly harsh grade curve.</p>
<p>
</p>
<p>And if that is indeed true, then that would indicate that the elite schools are clearly doing a terrible job in admitting and yielding students who are actually going to perform well in their STEM majors. HYPS have developed some of the best science/math departments in the world, and the latter 2 (especially S) have built some of the best engineering departments in the world. Why would these schools persist in admitting ‘relatively subpar’ students into those departments, relative to the ostensibly excellent humanities students? </p>
<p>Or take those engineering programs who run admissions schemes separate from that of the humanities/lit students. You would think that, say, the Cornell University College of Engineering would admit only those students who are actually suited for engineering and screen out those students whose talents lie in humanities. So why do they persist, year after year, in admitting students who do not perform well, relative to Cornell’s humanities students? </p>
<p>Anybody (such as xiggi) who truly believes this must concede that a clear problem therefore exists in the admissions process. Schools should be doing a better job of admitting a student body that is properly matched with the majors that they offer. Obviously you would never have perfect matching. But the mismatching would be symmetric. For example, in some years, the student body would be slightly mismatched such that the engineering students earn worse average grades than the humanities students, and in other years, the student body would be mismatched in the other direction, such that humanities student would be earning worse average grades than the engineers. But you wouldn’t have consistently asymmetric mismatching.</p>
<p>
</p>
<p>Weeding out - if it needs to happen at all - should happen only at the lowest tier schools with the least stringent admissions standards. At those schools, there may well be some students who are truly incompetent at STEM and perhaps should not graduate with a STEM degree. </p>
<p>But why does STEM weeding have to occur at the top schools? Presumably, top schools are only admitting highly qualified students (and if they are not, then that’s a problem with admissions, as alluded to in my post above). Hence, I suspect that most, if not all, students at the top schools are probably qualified to earn a STEM degree and work in the STEM industry in some capacity. Granted, they shouldn’t all get A’s, but I don’t see the need to weed them out of the industry entirely either. Somebody who flunks out of engineering at Berkeley probably could have successfully graduated and become a engineer had he just gone to an easier school. Granted, he probably wouldn’t be a star engineer, but he’d still be a perfectly serviceable one. So why weed him out of engineering entirely? That’s one more engineer that the country could have used. I hardly think he’s any less capable than the guy who went to some 4th tier engineering school and graduated with a 2.1 GPA, yet that guy gets to be an engineer.</p>
<p>
</p>
<p>The numbers are on page 1 of this thread…</p>
<p><a href=“http://talk.collegeconfidential.com/12626981-post7.html[/url]”>http://talk.collegeconfidential.com/12626981-post7.html</a></p>
<p>
</p>
<p>BB, I can’t really comment on the logic of what you wrote, at least not as a sequitur to what I wrote. My comment was that there is no evidence that the students who get lower GPA or flunk out in STEM-oriented curricula would do better in hum/lit classes. My second point was that students who opt for hum/lit classes might do well (GPA wise) because they are well-suited for such programs and respond to the expectations placed upon them. </p>
<p>As far as admissions, I am only familiar with schools that do not accept students in specific majors and value a liberal arts education.</p>
<p>“My comment was that there is no evidence that the students who get lower GPA or flunk out in STEM-oriented curricula would do better in hum/lit classes.”</p>
<p>Indeed, many I’ve encountered over the years would do rather poorly. Many people gravitate to engineering -type majors not only because they’re good at math and science, but also because they’re horrible at writing papers, or writing effectively, relative to the standards of arts & sciences programs of repute comparable to their engineering college. Or they just don’t like reading and writing period.</p>
<p>On the other hand, others would do just fine.</p>
<p>
</p>
<p>Not sure why this is addressed to me, but in the case of Cornell, I would agree that there there is a problem in their admission process. Many of students the school wants seem to prefer to attend a different school. :)</p>
<p>On a more serious basis, perhaps we should remember that adcoms accept 16 and 17 years old who THINK they’d like to be engineers or math majors without really knowing what will be expected from them in a college setting.</p>
<p>1) Agreed, and I don’t think anyone is making that claim.</p>
<p>2) Which is back to my question: are students being admitted to STEM majors who are not as “suited” for success as those being admitted to hume/lit majors? If so, why do you think this might be?</p>
<p>For example, Stanford has one of the top STEM programs in the world and can pick and choose who it wants – losing only a few hundred (?) primarily to HYPM. Can S not find STEM prospies “well suited” for success, such that 2/3rds of them would earn an “A” (similar to hume/lit)? Is Stanford really admitting STEM Frosh that aren’t as “suited” for their major as hume/lit prospies are for their major? Why would S do such a thing? </p>
<p>Dartmouth is a liberal-arts-focused college and its numbers are similar.</p>
<p>
</p>
<p>How is this any different than those that think they want to be a Russian Lit major? Or English major, or Econ major? (Could it be that the Intro courses in the non-STEM fields also have 2/3rds A’s?)</p>
<p>
</p>
<p>
</p>
<p>And we can surely all agree that plenty of humanities students would perform poorly - indeed probably fail - if they had majored in STEM. </p>
<p>Yet the fact of the remains that the empirical evidence demonstrates an undeniable asymmetry: STEM students earn lower average grades than do humanities/arts students, and this seems to be true across schools. If intra-school grade inflation did not exist and self-selection was the lone issue, then that begs the question of why so many more humanities-oriented students mistakenly choose STEM majors (which would supposedly then account for the low STEM grades), but relatively few STEM-oriented students mistakenly choose humanities majors (and hence would earn low humanities grades). </p>
<p>Put more starkly: where are all of the poorly performing humanities students? Why does the mismatching always seem to be asymmetric? </p>
<p>
</p>
<p>Which school would that be, pray tell? And would that school also exhibit intra-school grade inflation?</p>
<p>I don’t understand what you are saying about Cornell. I am not aware of a problem in the engineering college’s admission process. </p>
<p>The most recent data posted shows 83% of freshmen entering the engineering college graduated from it, 6% graduated from one of the other “endowed colleges” there, 3% graduated fom one of the “contract colleges” there. </p>
<p>A feature of Cornell is its broad strength in a wide variety of subject areas, outside of engineering as well as inside of it. This feature of the university disproportionally attracts matriculants who have interests outside of just engineering, who intend from the outset to take advantage of the university’s strengths in areas outside of engineering. It is inevitable , given such profile of matriculants, that some decide they prefer another area outside of engineering. Nevertheless that retention profile is actually very strong for engineering colleges. I can’t see that they’re doing anything wrong at all.</p>
<p>Engineering programs/colleges in general should have relatively high attrition, and not just because it’s hard. Few people really understand what engineering is when they are seventeen and applying to colleges. When they get to college they find out. They also get exposed to other subjects. Interests naturally evolve as people’s minds get opened. There is nothing wrong with that at all. Nevertheless, despite this natural process, and a body of matriculants who might be more susceptible to it than is generally the case, Cornell’s engineering college has a reasonably high retention and graduation rate.</p>
<p>I’m not sure Cornell is really such a good example of whatever it is you were trying to show. But I guess I missed the point.</p>
<p>
</p>
<p>Xiggi has contended that students properly self-select into humanities courses, hence accounting for the high grades in those courses. {That doesn’t explain why students are able to properly self-select into humanities courses but apparently improperly* self-select into STEM courses, but let’s put that issue aside for a moment.} In other words, self-selection -> high humanities grades, according to Xiggi.</p>
<p>I would submit that the far more plausible argument is the reverse: high humanities grades -> self-selection. As evidence of that, I would point to the experience of Cornell. Cornell implemented a policy in 1997 where course grades of prior years were publicly posted. As soon as they posted the average grades assigned in particular classes, students (unsurprisingly) began to gravitate towards the higher-grading classes, now that they had reliable information about what those classes were. If Xiggi’s hypothesis is correct, then you would expect that Cornell’s grade distribution would fall, or at least remain the same, as students in the post-policy era are now choosing classes only because they had provided high grades in the past, regardless of how well those courses matched their interest level or talents. </p>
<p>Instead, the opposite occurred: grade inflation increased. Students self-selected into classes that happened to assign high grades but may have had little to do with their abilities, which boosted grade inflation even further.</p>
<p>In other words, self-selection is not a cause of grade inflation. It is actually an effect. Students tend to prefer (and hence self-select into) courses who assign high grades. If self-selection was a cause of grade inflation, then you would expect Cornell’s overall grade distribution to decline, or at least remain flat, post-policy. Keep in mind that the policy revealed no new information about the content of the courses (hence, students would not have improved their matching based on course content). The policy only revealed information about the grading of the courses. </p>
<p>*In 1997 Cornell University began posting median grades for every course online, so that students could put their own grades in a larger perspective. (After all, an A in “Physics for Poets” is presumably less impressive than an A in “Physics for Physicists.”) The university’s theory—which reveals an astonishing naivet</p>
<p>
</p>
<p>I would submit that few people know what many arts/humanities majors are when they are 17 and applying to colleges. It is often times only when they are in college that they are ever exposed to, say, Russian Literature or Film Studies (no, watching Thor on Imax with your date on Saturday night is no more equivalent to ‘film studies’ than is me playing Angry Birds on my Ipad equivalent to EECS)</p>
<p>So why isn’t there comparably high levels of attrition and low grades within the humanities? Why aren’t there hordes of humanities students being weeded out, just as there are hordes of engineering students being weeded out? </p>
<p>The core issue at hand is why are certain majors graded harder than others. It is the contention of Xiggi that perhaps engineering majors are graded harder because of shoddy engineering admissions policies which admits unqualified people. But if that is really true, then the next logical question is why can’t/won’t schools do a better job of admitting more qualified engineering students who earn high grades, in the same way that they admit supposedly “highly qualified” humanities students who earn high grades? </p>
<p>Maybe Cornell CAS knows something that the College of Engineering doesn’t know? If that’s the case, then maybe the CoE adcoms need to call a meeting with the CAS adcom. </p>
<p>The far more plausible story is that humanities classes are simply less demanding - they assign higher grades for less work and less talent - than do the engineering classes.</p>