<p>
We have no idea what the modal is, I don’t know where you’re getting that from.</p>
<p>
It would be very strange if the mean was higher. The GPA scale tops out at 4 and extends down to 2 (or lower, I don’t know what Harvard’s academic probation policies are). Top 20% is 3.679, which doesn’t indicate any weird clustering.</p>
<p>I know at Hopkins, for our intro chem class, about 1/3 of the class got an A. But in our writing course, almost everyone got an A ( there were only a few people in that class). I think the grade distributions vary greatly depending on the type of class and number of people.</p>
<p>lockn,</p>
<p>I should have said median instead of modal. Too much wine at a business dinner tonight?</p>
<p>It would not be hard to have the average/mean be higher than the median if the distribution were skewed to higher grades. Even though the range is greater on the lower end, that tells us nothing about the distribution. </p>
<p>It is often tempting to imagine that numbers scatter through a broad range just as they do through a narrow range, but real data is not like that. it tends to cluster. </p>
<p>Take human height as an example. Average height for an adult male is about 5-11. Maximum recorded height is 8-11! but that’s 3 feet above the average. range on the other end is theoretically almost 6 feet. so height clusters at the higher end.</p>
<p>
</p>
<p>The honors degrees aren’t awarded by going down the GPA list. 3.417 is the absolute minimum cutoff below which
you don’t get honors no matter the quality of senior thesis or GPA-in-major or how many courses you took.<br>
So 3.417 is probably not the 50th percentile but some lower “quality control” standard such as 40th or 30th.</p>
<p><a href=“newmassdad:”>quote</a> </p>
<p>I would still like to see an actual count, just to see if Harvard is actually keeping to its own limits. I’ve heard anecdotal reports in the past that it does not - that its efforts to reign in honors has been fought by the faculty. siserune, can you give a link?
[/quote]
</p>
<p>Trust, but verify!</p>
<p>[357th</a> Commencement: Harvard confers 6,966 degrees and 104 certificates — The Harvard University Gazette](<a href=“http://www.news.harvard.edu/gazette/2008/06.05/42-degrees.html]357th”>http://www.news.harvard.edu/gazette/2008/06.05/42-degrees.html)</p>
<p>The revealing detail for this discussion is the number of “Cum Laude in General Studies” degrees. According to the rules specified in the student handbook (linked) it seems that, apart from some rare exceptions, everybody above the lowest magna GPA gets a CLGS designation. Thus, magna+summa+CLGS degrees account for virtually all students with GPA above the magna cutoff. In last year’s data this is 358 of 1564 degrees from the college = 22.9 percent. </p>
<p>So we have one data point: a 3.710 GPA (the 2008 magna cutoff) was at the 77th percentile of Harvard class of 2008 GPA. </p>
<p>We also see that students at the 23rd percentile from the top were receiving, and lower ranks than that were being nominated, for an award given to 20 percent of students; magna and summa degrees number 314 of 1564 = 20.077 %. The same happens with the cum-laude-in-major-subject degrees; 50 percent receive that honor or higher, but some nominees and recipients come from below the 50th percentile. The lowest cum laude GPA was 3.414 in 2008 and 3.417 in 2006, according to the online material. This could have been somebody from the 42nd percentile who took a hard courseload. It could also be someone well below that percentile who had mediocre grades but ended up doing world-class research by the senior year. The cum laude cutoff is that of the single lowest-GPA honors recipient and is by definition likely to be an outlier. Extrapolating from the magna figures we would guess the honors degrees are cut off near the 42.5 percentile (2.5 x 23 = 57.5 below the top) but this could be too conservative. The cutoff could easily be anywhere from 35-45 percentile, or even lower.</p>
<p>Bottom line: 3.710 was the 77th percentile, 3.414 is below the 50th percentile and probably somewhere around the 40th.</p>
<p>[FAS</a> Handbook for Students: Chapter 2: Academic Information](<a href=“http://webdocs.registrar.fas.harvard.edu/ugrad_handbook/current/chapter2/honors_requirements.html]FAS”>http://webdocs.registrar.fas.harvard.edu/ugrad_handbook/current/chapter2/honors_requirements.html)</p>
<p>“The idea that it is easy to do well at Harvard is baseless at best and a flat out lie at worst.” is pretty lame. Surely Harvard defenders, of all schools, can do better than that!
Data please.</p>
<p>Data can’t show many aspects of how hard a school is, so to treat it so religiously in this discussion is a mistake. You’re just taking “hard” to mean “hard to get a good GPA,” and not taking into account the actual experiences of students and how hard they work to do well. Which is subjective, but that doesn’t mean it isn’t part of the discussion. You should take a page out of Reed’s book and think outside of grades. Because believe it or not, not all of us shape our lives around grades. Some of us take the hardest classes possible, even though we know that we might get lower grades. And then we sleep 4 hours a night to do all of the work. So i’m sorry that I don’t have data for you, and I’m sorry that Harvard gives out too many good grades to its students. My point still remains that to call Harvard a slacker school is misinformed at best and flat out lying for the sake of belittling the work that is done here at worst.</p>
<p>I did go to more than one of the top 50, and so did many of my friends – I met every Harvard transfer from Fall 1997 to Spring 2002.</p>
<p>I never conducted a poll, but I can tell you with certainty that most people who transferred to Harvard from other excellent schools did not find Harvard to be easier than their previous school. People who came from MIT and Caltech (a sizable percentage of the H transfers) were generally the exceptions.</p>
<p>
[quote=]
Are you personally familiar with the Harvard details I mentioned (and sampling theory, for that matter), or making these comments on general principle? What problems, specifically, are you pointing to as coming from self-selection?
[/quote]
</p>
<p>actually, i am personally familiar with this survey as a harvard senior who just filled it out. it is sent by e-mail, so it is not done in the dining halls. there is definitely selection bias in the types of people who fill out these surveys. conscientious people are much more likely to do so, and they are also more likely to have better grades.</p>
<p>when i worked in a psychology lab, there was a survey i needed to distribute, and the graduate student wanted me to do it in the dining hall instead of via mass e-mailing bc there had been bias using that method in the past.</p>
<p>likewise, as i said before, people who have low GPAs are less likely to report them.</p>
<p>You’ve got to remember too, about rating colleges. Some publishers are in the business of selling college ratings. To succeed, they must convince you one college is far superior to another, and then sell you the details. “Tier1, tier 2, Best in the South” etc. Not to say that all colleges are absolutely equal, but to say a publisher’s ratings are a product for sale, just like the shamwow for example.
Ask yourself who is the best NFL player right now? Some big names come to mind, but as a Colts fan, I don’t want Saturday(the Center) throwing the ball. I don’t want QB Manning trying to make an open field tackle. I don’t want DE Freeney going out for passes. Different levels of expertise and in different areas make it impossible to rate even who is the best Colt, and they only have a roster of 53.</p>
<p>All I know is my kid is a premed junior at Harvard and doesn’t find it all that easy.</p>
<p>
</p>
<p>I agree that the e-mail is much less favorable than the dining halls. Was the survey done after final exams?</p>
<p>
</p>
<p>One would think that, but experience with these surveys, including by e-mail, has been that the selection bias is not what it’s cracked up to be. It isn’t something obvious like conscientiousness that correlates well with the thing being measured. The conscientious people could just as well not answer it because they’re studying for exams, or busy packing, or whatever. The slackers might enjoy wasting time on the survey, or discussing it with their friends. Normally the major sources of bias are predictable and here there is no clear reason to expect a strong skew. </p>
<p>
</p>
<p>They also love the dining halls because of the amazing response rates. e-mail can be hit or miss, and which is the case depends very much on the contents of the survey.</p>
<p>
</p>
<p>One might think so, but experience with similar surveys containing questions more personal than GPA shows that (assuming things are anonymized via surveymonkey or other means) this is not much of an issue. I would expect the main source of bias is that not everyone knows their GPA. The students with higher grades are likelier to know the GPA, or know it to the full 3-place accuracy, and those who misremember probably would overestimate their GPA a bit on average. However, I would not expect any of this to be a large effect. Lying about GPA in an anonymous survey of an undergraduate population totally inured to survey instruments, is probably not a noticeable source of bias. That the survey is an innocuous general one rather than a transparent attempt to compare, say, GPA with race, probably also reduces any inhibitions about answering. </p>
<p>Overall, I would expect a small amount of upward bias, but nothing to fundamentally invalidate the results. That it gets the athletes and final clubs vs general population figures qualitatively correct also increases confidence in the results.</p>
<p>Linear interpolation of the magna/summa calculations (taking the cum laude cutoff of 3.414 to be around the 40th percentile) gives 3.49-3.50 as the 50th percentile. It’s hard to say whether the average should be higher or lower than this; I would bet on lower, very slightly, so something like 3.48 which also, given the margin of error and likely small upward bias, is consistent with the survey. I don’t think the true answer would be nearly as low as Hanna’s guess of 3.3x.</p>
<p>i should’ve mentioned that it came out after finals. </p>
<p>the graduate student i worked for specifically told me that part of the reason he wanted me to do the survey in the dining hall was bc of response bias in e-mail. it is hard to know with self selection; it could have no effect or it could have some unknown effect. this survey certainly wouldnt be approved by peer review (obviously), and i don’t see why we should have less stringent standards. i don’t think they survey tells us much, period.</p>
<p>the cum laude cut-offs are much more compelling. 3.5 still sounds high, though. i think 3.4 sounds about right. as has been amply pointed out, there is a longer left tail in this case. i wish we just had a calculation releaed by the college. some un-official sources say the average GPA is 3.4 (eg, [John</a> Winthrop House: lowGPAjunior.html](<a href=“http://winthrophouse.harvard.edu/web/Academics%20and%20Advising/Pre-Medical/lowGPAjunior.html]John”>http://winthrophouse.harvard.edu/web/Academics%20and%20Advising/Pre-Medical/lowGPAjunior.html)). (just for reference, i am finishing with a 3.88, so i’m not, like, trying to comfort myself or something. 3.88 is around 92nd-94th percentile, based on pbk nominations).</p>
<p>By graduation rates, it would seem that Georgia Tech has almost everywhere beat for difficulty–its 4-year graduation rate is in the 40% range I believe.</p>
<p>^^Although, that statistic by itself could be interpreted in different ways. Maybe it’s easy academically but just an awful place to be in other ways, so 40% of the students transfer. BTW, I have nothing against Georgia Tech. I only spent a few hours there and it was great.</p>
<p>My understanding of what may be behind Ga Tech’s low 4 year graduation rate has more to do with the fact that most of the their majors actively support taking one semester as a co-op work internship, thus automatically delaying graduation. In many cases, this “working” semester resulted in a job offer. If not, it certainly provided valuable relevant work experience very useful on that college graduate’s resume. I think more colleges should do this. Wish my S’s did.</p>
<p>Having attended both Grinnell and Pomona, although many years ago, I can tell you it was a toss up then. Both required steady attention and a fair amount of work, but nothing that would kill the average person who attends. Loved my time at both, although I’d choose Pomona today.</p>
<p>Engineering degrees increasingly are designed to take five years, not four. The graduation rate at Georgia Tech jumps from 33% at four years to 69% at five and 78% at six, according to the National Center for Education Statistics:</p>
<p>[College</a> Navigator - Georgia Institute of Technology-Main Campus](<a href=“College Navigator - Search Results”>College Navigator - Georgia Institute of Technology-Main Campus)</p>
<p>Since not everyone a Georgia Tech is studying engineering, this clearly does not account for all of those who don’t graduate on-time, but it must account for many.</p>
<p>Well-said, cellardweller in post #32.</p>
<p>I don’t know what my kids’ grades are, and actually, I don’t care. (collective gasp) I am actually happy when they tell me “My grades might not be so good this semester because I got involved in this play”, or “…I got asked to teach programming to kids”.</p>
<p>I didn’t want my kids to wind up as grinds, or “ghosts” like they say at MIT, holed up alone in their room tooling all the time. There is so much going on on college campuses, and I’m happy they are involved.</p>
<p>In terms of classes, cellardweller said it exactly right. Even kids who come into college with amazing skills will find they are not good at EVERYthing. It’s good when they try things they don’t know if they will be good at, they take chances and go off the map a bit. </p>
<p>To the parents who want to go to graduation and count the summa cum laudes to try to determine if there’s grade inflation – seriously. Think about that.</p>
<p>AnudduhMom, good for you! Great attitude. I’m sure your kids are better-adjusted and happier for it.</p>
<p>re: Harvard–
I have no personal reason to stick up for the school-- but now, and even back when I was in HS, Harvard had the reputation of being ‘easy’ once you got in. In several cases, the comments were made by people whose family members who had gotten into other ivies, but not Harvard… </p>
<p>I’d tend to agree more with one of my S’s friends from HS who attends, who told him it’s actually quite hard to excel and/or stand out, because there are just so many incredibly bright students. And I’d bet that’s at least partly true for any top school or state honors program–</p>