Dean Says Median Grade at Harvard College Is A-, Most Common Grade Is A | Hacker News Reader

Dean Says Median Grade at Harvard College Is A-, Most Common Grade Is A

thecrimson.com

27 pointsmayneack58 comments

The problem is that grading serves two purposes that are at odds with each other: comparison within a group and comparison between groups.

Harvard's grading system is useless to a student that is trying to figure out how they are ranking among their peers or whether their studying habits are increasing their performance. However, it's great when they're on the job market and their 4.0 GPA is better than the 2.9 GPA of Engineering State that still grades on a sharp curve.

The increase in grade inflation has reflected the trend of college goers valuing extrinsic factors (get a better job) over intrinsic factors (learn more stuff).

A solution to this problem would be to split the two grading systems apart, having one standardized, SAT style test at the end of college that would allow more accurate comparisons between schools while preserving GPA as a within school measure.

In fact, it stuns me that the College Board hasn't positioned the GRE as this kind of service and pioneered some kind of aGPA score that would combine your college GPA with the relative performance of your college against all others.

One of the most interesting arguments against extreme grade inflation I've seen arose in the context of students considering going to grad school in a difficult field. Several professors in the conversation said they weren't comfortable with writing a recommendation or encouraging a particular student to go on to graduate work, but that it was difficult to convince these students that thhe weren't cut out for it after they had given them As in upper level courses.

At my alma mater the grade A+ was used in some departments for this sort of purpose. It wasn't mandatory, and it didn't count towards GPA, sometimes no one got one at all, but it signaled something to the students who did.

Separately I had one professor, in history of all things, who was a notoriously tough grader but was retiring the next semester after I had her. As a one time policy she issued two sets of grades, the one you deserved and the one she decided to give out to be in line with her colleagues. I was much prouder of the B in the first category than the A- in the second.

I went to ETH Zurich, Switzerland for both my BSc and MSc degrees.

The undergrad program that leads up to the BSc degree is ridiculously easy to get into (Swiss nationals even have to be accepted into the program by law AFAIK -- I'm not Swiss) but the first year of that program is so absolutely horrendously hard to survive (one massive set of 10 or so finals at the end of year 1) that you literally come out of your first year with five kilos off the scale and pale like a ghost (I'm Caucasian ... ?).

At any rate, point being at ETH Zurich in particular and other European unis in general (again AFAIK) the goal is to survive and get your degree whichever way you can.

Grade inflation at "top notch" American unis with their laughable grading standards, the constant availability of extra credit to push up your final grade at will, and the weird financial and social hurdles that applicants are presented with utterly confuse me.

Speaking of which, tuition fees at ETH Zurich were 750 CHF per year (830 USD).

Top notch American universities have ridiculously easy grading standards because the filtering happens at the admissions stage. Harvard's undergrad has something like a 5-6% acceptance rate. Admission requires being on the ball since age 14 so you can apply with a perfect high school GPA, getting at least in the 98th percentile on the SATs, and doing a bunch of extracurricular and service projects while maintaining that perfect GPA.

After that, the fact of whether you're smart and hard working is presumed. The grade inflation exists to make college a fun experience where you can try out different sorts of classes without prejudicing your chances at jobs and graduate school later.

At this point what is the logical argument for assigning grades during the four years of study, then? Why not simply rubber stamp a 3.9 GPA on everyone's curriculum vitae and call it a day?

Oh wait ...

I have always found it amusing that the virtue of hard work is bestowed on kids who make it into the most selective American colleges. Another way to look at it is that, it is precisely hard work that these kids(or at least their parents on their behalf)are trying to avoid.

In other words while the rest of us will have to slug it hard until the moment of death (quite literally for many), once you enter a school like Harvard it's all gravy for the rest of your life. Basically put in about five years of hard work to get into Harvard and live a cushy life from then on, sweet deal indeed.

I have meet people who went to Harvard and the kind of accommodation they get is mind blowing to a state-school Joe like me. I dated a girl who was interning in DC a couple years back from Harvard law school, she had the whole summer to turn in what was a five page paper. Mean while I had to write 10+ page papers practically every week in b-school at a public school and if I missed the due date, the penalty can be anything from losing some points to getting no points.

I don't think Harvard grads tend to autopilot the rest of their lives. That said, one of the purposes of going to Harvard is to avoid risk. Harvard students tend to come from the upper middle class (but not from "fuck you" money), and going to Harvard helps minimize the risk of falling out of that class. Hence the popularity of banking, consulting, federal government, etc, among Harvard undergraduates.

You're not on the ball since age 14 - your parents are ages 14 - 17 or 18 and you try and take over after that.

This is the problem with Harvard et al.: you get in b/c your parents have the financial resources to keep you on the ball until its time to sit your Harvard admission exam (or w/e the procedure to get). Financial resources will come in the form of being able to spend quality time with you (and teach you about keeping the ball close) instead of seating you in front of the telly or private tutoring etc.

European unis on the other hand (bar top notch British unis which are another kind of messed up) operate on the principle of offering young adults a bright future if they man / woman up and bash their heads through some severely tough undergrad years.

Social mobility and stuff ...

> In fact, it stuns me that the College Board hasn't positioned the GRE as this kind of service..

The CS department at my alma mater did require the ETS Computer Science Field Exam for graduation. The results are interesting as you can compare your personal score against students from several dozen other universities as well as your department's score.

All the noise about standardized testing aside, I think that this sort of thing should be required for anyone to get a degree. I've met far too many people with Master's degrees who seem easily confused by basic math and science.

The fact that there aren't any measurable standards that go beyond the institutional level really dilutes the value of a degree.

One way to approach this problem is to have accreditation bodies require a more equitable distribution across the student body.

My experience was that it's relatively easy to stack rank Harvard undergraduates, but it often doesn't feel like it serves any educational purpose to do so.

In the class I taught of 12, there were 10 that seemed like they deserved an A or A-. They completed all the assignments and did their absolute best as far as I could tell. Some were definitely better than others (A vs. A-), but giving the hardworking but not genius students Bs just made them stressed out. They would come to office hours to ask how they could improve. The only true answer I could give was "there are 2-3 students in this class that are just incredibly creative and gifted writers, they get the As."

Maybe it's the role of a teacher to stack rank students, but it's no fun. I'd much prefer to try to get the best out of each student. The truly top students are going to have no problem distinguishing themselves later in life anyway, it's just so obvious how good they are.

Ya, that's my difficulty with the article. If all the students learned the material, then what is the problem? OTOH, if they just didn't grasp information, or didn't try, they deserve something other than an A.

There is no easy answer, and it varies by course. In an algorithms course you are going to learn trees, sorting, graphs, and complexity analysis. If you master those, why not an A? Sure, you can make the course insanely hard, and force the students to do essentially original research to get an A, but what is the point there, exactly? You end up with a pressure-cooker, swamped with work school just to create a bell curve. I spent so much of my University time on my own research ideas, helping professors, and so on - course work was almost a back drop. I think I got so much more out of it doing it that way. A competitive fight of churning out massive amounts of work against other straight-A's in high school type-A personalities? Not so valuable (IMO). Time to think, dream, and experiment is what is valuable, and how do you put a grade on that?

edit: I take this back somewhat. In some fields it definitely pays off to have a broader foundation. I took quite a few math classes, but would have profited by by even more. Lots of research papers are tough going because I don't have all the math. So there is value in volume of work, too.

Stack ranking in small groups is stupid and counterproductive because it forces a distribution where none necessarily exists. So for any particular class of 12 I agree with you.

However, Harvard enrolls over 1500 frosh next year. Is there really so little daylight between them that they should all get an A or A-? Graduation is supposed to ensure the minimum fulfillment of standards, GPA is for ranking. Given that you're seeing kids on the far end of the tail, I would expect more variance, not less.

Let's be honest here: grade inflation is not something we do for a reason, but rather something that happens because no one has any particular interest in preventing it.

Full disclosure: I attended a demanding college whose 2007 median GPA was 3.35 (the average was less). Of the 150-ish annual graduates, only 7 have ever achieved a 4.0 since its founding in 1955.

The overall GPA can still be used to rank, even at Harvard. The latin honors are fixed at top 5, 20, and 50 percent of graduating classes.

I don't really care about grade inflation overall because I don't think more competition for grades would improve Harvard as an educational institution.

I do think grading norms/curves may be important to standardize the distribution across majors. It's absolutely the case that some students avoid difficult courses. This is very unfortunate.

I go to Yale where the average GPA is a 3.6. So we suffer from a similar "problem." But honestly, I do not think it is a problem. Kids here put an absurd amount of work into their classes. It's not surprising that the average GPA is a 3.6, because the average assignment actually is A- level work. The obvious argument against this is that the grades should be curved so that the average level of work receives a C... but what do you do when most kids are getting the same top grades on tests, or writing the same high quality papers? It's a hard problem to solve, and as long as other colleges have grade inflation, it would be disadvantageous to your students to not have it.

The problem I have is differing GPA distributions between majors. History majors have a much easier time getting in the 3.6-4.0 range than CS majors, because history teachers will rarely give anything less than a B. So when a CS major applies to jobs that other majors are also applying to (say, finance), he can look bad in comparison.

No one gives a shit about GPA in the real world.

Right, and then there's this. But employers do give a shit about GPA when recruiting college seniors, because it represents one of the only available metrics for quality of work. In my experience, providing employers with other metrics reduces the value they place on GPA.

It may depend on the case - in many cases, the school a student attended trumps the GPA the student achieved at that school. For schools like Harvard and Yale, it is the brand that employers react to most, not the GPA. Having recruited students from those schools while at a strategy consulting firm, typically we looked at school, then major, then other stuff (which may or may not include GPA). Major was a better indicator of talent level than GPA at those schools.

Perhaps schools should adopt the IIHS ratings scale: Good, Acceptable, Marginal, Poor. Doing away with letter grades that have strong historical bias and cultural signaling might be useful.

Example: http://www.iihs.org/iihs/ratings/vehicle/v/toyota/4runner

Nearly all of the most sought-after exits into the "real world" from Harvard care about GPA: banks, consulting firms, Big Corps, graduate schools, etc.

They certainly do if it's your first job out of college. Once you've got work experience, you're right, no one cares about your GPA.

  ...what do you do when most kids are getting the same top grades on tests?

Give harder tests and adjust grades upwards accordingly. If everybody is blowing out your measurement scale, your scale needs to be higher.

> If everybody is blowing out your measurement scale, your scale needs to be higher.

This is only true if the purpose of your measurement scale is to partition scores into a normal (or similar) distribution. This is not always the case - sometimes the purpose of a rank like this is to measure something absolute, rather than relative.

For example, what would happen if every restaurant in NYC cleaned up their kitchens and used such safe practices that they all met the requirements to get an "A" grade on their Dept. of Health inspection? In that case, I would say that they are all safe to eat at, and they all deserve "A" ratings. However according to you, this would mean the scale was broken and we should grade them on a normal curve, giving B's and C's and even F's to perfectly-safe-but-not-quite-as-clean restaurants. This just doesn't make sense - if you meet the requirements for an "A", it's not fair to be put out of business just because someone else spent more time polishing their restaurant floor.

I think school grades should be treated similarly - as an "absolute" measure of whether or not the student learned the curriculum, not a relative measure of how well they performed against their peers. No one should be punished for the "bad luck" of being placed in a class full of wicked-smart students. Now I can't say whether or not the A's given out by Harvard were in fact fair, but the fact that the median grade was an A is not, in and of itself, enough evidence to say that the grading system is broken.

Grade inflation stepped up a notch during the Vietnam War, when professions were reluctant to fail students out of college an into the draft. Now it's done to protect a college's graduation rate, which affects rankings.

Elite Ivy League institutions admit bright, motivated high school graduates and produce bright, motivated college graduates. The institution's value add is mainly networking and branding--kind of like YC.

It's also due to the fact that easier grading brings more positive student evaluations of profs, which are taken into account with salaries and tenureship.

I assure you that at a Research I university like Harvard, student evaluations have at best a pro-forma impact on tenureship, and probably salary as well. Faculty at Harvard are tenured and compensated on their abilities to publish high profile research and attract major grants, showing up to lectures is about all that is expected of them teaching-wise.

Yeah, that sounds reasonable. I don't have firsthand experience with Harvard, this is what I've just heard to be the general case with universities.

Maybe Harvard. But I am employed as a math professor at a Research I state school, in which context you are quite mistaken. Research is most important, but good teaching is also important.

"Universities don't care about teaching" is a fun meme to kick around, and there are anecdotes to support this, but by and large it is not true.

My comment was directed more specifically to Harvard and narrow set of top tier schools, I do not believe that universities don't care about teaching. The list of "very high research activity" (apparently research 1 is no longer used) includes 108 schools [1] of which Harvard is at one extreme end.

That said, my father recently retired after 33 years with his department at a "high research activity", the second tier, and he can't recall a single case when someone was denied tenure where teaching performance was an issue. Research, and increasingly grant acquisition were always the issues.

Personally, I went to a small liberal arts school. Having taught at two top universities, private and public, I'm happy I went where I did. Undergrad teaching was the top priority and the difference is stark.

[1] http://en.wikipedia.org/wiki/List_of_research_universities_i...

To play devil's advocate for a moment, the tenure evaluations I am familiar with at many universities take student feedback into account, as a proxy for "good teaching," but unfortunately teachers students like and teachers that are good at teaching may not be the same. Thus universities seem to care more about keeping students happy than about "teaching" per se.

I always hated this about other schools. While I will be the first to admit that I could probably have tried harder in school, I blame this practice in Ontario for part of why I could never get an interview for medical school. Waterloo engineering was definitely one of the tougher programs, and an 80 average was actually tough to attain. It seems that my friends at other schools or programs had an easier time getting 80s and 90s.

PhD programs are well aware of grade inflation at different universities and take this into account when conducting admissions. No idea if med schools do the same but I'd be surprised if they didn't. Your difficulty getting into med school could also have been because engineering not a typical "pre-med" degree.

Medical schools specifically stated that they did not adjust grades. There was some different weighting mechanisms employed for certain schools, but that was at the school level, not program level.

Furthermore, medical schools here typically encourage students not to pursue "pre-med" degrees. They want students that pursue challenging programs, in order to demonstrate their abilities and to have a backup if medical school doesn't work out.

I don't have any regrets about engineering, it's the best decision I ever made, and I'm sure I'll have an impact in other ways. It's just still a bitter/sore spot for me.

An 80 can be difficult at Waterloo, but no doubt theres over a dozen students in your year with a higher avg. It's good, but not remarkable.

Consistent 90's in any degree program in Ontario is quite the accomplishment, regardless of inflation.

I taught an upper-level undergraduate course in math (combinatorics) at Stanford a few years ago. There were 18 students, and I gave all of them A's or B's.

They deserved it. I was inspired by what they accomplished. I would have felt uncomfortable if I had been asked to make sure the average was around a C. (And all of this is very typical of the experience of people teaching math at Stanford.)

This is why I think concerns over grade inflation are overstated. High grades for outstanding students doing outstanding work are warranted, and these are the kids Harvard by-and-large admits. The apparent grade "inflation" over the past few decades reflects the fact that it's no longer possible to get into Harvard with one AP course: http://www.nytimes.com/2007/04/29/nyregion/nyregionspecial2/...

Your overall point is well-taken, but the Harvard admissions people always stress that there is no definitive profile of an admitted student. If they wanted to fill the class with perfect-SAT valedictorians from NYC and Boston, they probably could, but that is not what happens.

This could be the case, but why couldn't it instead be the case that a 2.9 at Harvard is comparable to a 4.0 at a lesser school, and a 4.0 at Harvard is outstanding and means something truly incredible?

That's all well and good, but the whole point of a scale is that students fall somewhere on it and it reflects work done in comparison to something. If not other students, then just patting ourselves on the back for a job well done.

Would you need a some slacker half-following directions to earn an Average? No, you'd need the Average of the students in the course to earn an Average. Otherwise the distinction is meaningless.

Then again, I've just spent some time re-reading Covey's Seven Habits of Highly Effective people, and I've given up caring too much about grades per-se. They discourage cooperation and encourage competition amongst peers. And in doing that strongly bias their adherents to win-lose scenarios when we should all strive for win-win scenarios.

If you think that's bunk, I encourage you to read Covey's book. Coming at it with an open-mind can truly be life-changing in a positive way.

> If not other students, then just patting ourselves on the back for a job well done.

Or, for the win-win version, how about an objective measure of correct answers/solutions. Grade inflation never bothered me in math and computer science courses, because what were they going to do when the students knew the material? Make 1 point the difference between a A and an C? I remember a midterm where a professor did exactly that (forced a grade-lowering curve on a group that had earned mostly 19 or 20 out of 20). It just didn't really work... A few kids even dropped the course, despite the fact that they basically knew all of the material (and had slipped up on only one or two answers).

Of course, they could just make the material impossibly hard so that only 2-3 students could get the A (after weeks of all-nighters and likely trading off good grades in other courses) but to me that is counterproductive to the whole reason the students are there -- to learn!

Old joke: What do you call a guy who gets straight C-minuses at Harvard?

A Harvard Graduate.

Harvard is effectively graded pass/fail. I'm not convinced that's actually a bad thing.

I never understood the complaints about grade inflation at top schools. The people who care to make distinctions among graduates at Harvard (investment banks, consulting companies, graduate schools), already know how to do so. Everyone else just cares about the "Harvard" on the resume. So who cares what the median grade is? They could hand out degrees "magna cum fruitcake" and nobody would blink.

I went to a liberal arts college that didn't do grade inflation. The average grade was nominally a C. The campus bookstore sold t-shirts with "C+: Better than average!" on them. Oh, and our diplomas are comically large :)

I have to imagine that the long-term solution for this is for employers to find better criteria to judge candidates on than a GPA. There's too much variance in schools and classes to really consider it an accurate judge of ability. It's a measure of how well you did in school, not how well you will be able to do your job.

I am convinced the best way to deal with classes is to make failing a non-factor, i.e. if you fail a class, it doesn't affect anything, and no one really "knows" outside of your school. Then, make all classes graded on a pass-fail basis, perhaps with a professor's ability to note that a few students went above and beyond expectations.

This way, you can actually say "This person knows x and is capable of doing y at z standards. See, they passed the course that required you to do all that." And you don't end up with weird situations like the "gentleman's" C, where you've passed a course but gained next to nothing from it. If you've passed an intro physics course with a C, what does that mean? Can you calculate vector forces? Or are your diagrams sometimes OK looking, but you can't crunch numbers?

So a 4.0 at Harvard is equivalent to a 2.0 at a community college. Niiiiice!

Yes, but the Harvard degree comes with a lifetime of Smug.

Similarly, lack of a Harvard degree comes with a lifetime of hating on people who went to Harvard.

I wasn't aware that observation of demonstrated behavior equated to "hating".

In high school I took number theory at the local community college. We stopped right before the chapter on quadratic reciprocity. Later, I found the Harvard midterm for their undergrad number theory class online. I took a look out of curiosity. The first question was on quadratic reciprocity, and I had no clue what the other 9 (or 14, maybe) were about. They were moving >2x as fast and doing 3x as much homework (the homework was online too).

I suspect the worst student in the Harvard class would have been competitive with the best student in the CC class. If 20% of the kids in the CC class get As, it's not a stretch to believe that 100% of the Harvard kids would have gotten As had they taken the same class.

Obviously this holds for some classes more than others, but the point is that variation in difficulty between classes, professors, and peer-groups can easily dominate variation in difficulty imposed by the grading scale.

You're saying, then, that a 4.0 at a community college is not even a 2.0 at Harvard? But how could it be? It's the opposite! 4 > 2! Those universities are trying to redefine numbers... Or maybe it's in the units! Ohhh, now I get it: Given CC=Commmunity College, then 4.0 GPA CC < 2.0 GPA Harvard. This implies: 2.0 GPA CC < 2.0 GPA Harvard. Now we drop 2.0 GPA from both sides and we get CC < Harvard, or Harvard > CC. But wait, is that true for every community college and for every class and for every course at Harvard and at all those community colleges? So maybe it's not 100%, so maybe 4.0 GPA CC > 2.0 GPA Harvard in some cases. so, the answer, as usual, is "it depends".

What does it depend on? It depends on how things are measured.

Woooooooooooo, wait a minute there Sherlock. Harvard is Expensive man! And hard to get into... How can there possibly be cases when Harvard < CC? Since Harvard is so 1337, there's no way that Harvard < CC, right (in some cases)? But then since it's impossible to measure (accurately and without bias) all of the community colleges out there, it's theoretically possible for CC > Harvard, especially when taking into consideration the tuition paid. So, in some cases, (CC+small_tuition) > (Harvard+expensive_tuition), and thus CC > (Harvard + expensive_tuition - small_tuition), so then one could say that CC - expensive_tuition > Harvard - small_tuition in some cases.

So, when you wrote "it's not a stretch to believe that 100% of the Harvard kids would have gotten As had they taken the same class." You're actually trying to disprove that "CC - expensive_tuition > Harvard - small_tuition in some cases", but that can't be, because there's no way to measure all. Thus your statement "it's not a stretch to believe" has to be incorrect (since can't be verified as correct). Anyway, 100% gotten A? Really? Are you a betting man? What if one student had a bad day that day and got an 89?

Be careful making generalizations, man. You can't know until you measure; and then you can't know until many other people have also measured, and then you can't know until many years have passed and nobody had come forward with substantial allegations of study bias in your field of study.

Find those

Harvard, at least in the last 5-7 years, has been effectively free for the majority of American families.

How did this:

> variation in difficulty between classes, professors, and peer-groups can easily dominate variation in difficulty imposed by the grading scale

Become this:

> 4.0 GPA CC < 2.0 GPA Harvard. This implies: 2.0 GPA CC < 2.0 GPA Harvard. Now we drop 2.0 GPA from both sides and we get CC < Harvard, or Harvard > CC. ... Harvard is so 1337, there's no way that Harvard < CC, right (in some cases)?

And why do you think I'm performing a cost/benefit analysis? I relayed an anecdote about a single class that had to do with course material, not finances or overall GPA comparisons.

> You're actually trying to disprove that "CC - expensive_tuition > Harvard - small_tuition in some cases"

No.

I'll admit that I made a slightly hyperbolic statement when I made a claim about kids in a class I never attended, but you belabor this point to such excess that I have no trouble calling it "nitpicking."

> Are you a betting man? What if one student had a bad day that day and got an 89?

Then my conclusion would not change in the slightest.

> Be careful making generalizations, man.

Right back at you. Slight hyperbole is one thing, straw men are another.

You are assuming that community colleges and Harvard have the same distribution of academic skill in each classroom and that the difficulty of the courses are equivalent.

I would imagine there's a significant selection bias at work here. This is not a random sample of students, and I don't think it's super odd that many of them get As.

Perhaps more get As than should, but I'd look at the grade distribution of a less discriminating institution if I wanted to get huffy about grade inflation. Especially with no meaningful data in the article about historic grade distributions at Harvard.

The problem is that law (and other professional) schools are incentivized to largely disregard the institution of applicants and instead base most of their admissions decisions on GPA and LSAT so they can get high US News rankings. An elite undergrad institution can best serve it's students by making sure high GPA's are achievable to hard working, smart students who want to go to professional schools.

Unless you're planning on applying to grad school, GPA really doesn't really matter for anything after about 6 months out of school.

Preferred admission to kids of alumni at top private school. And easy As that will pave way to other grad schools.

I don't want to believe US is a class based society but ...

Reply on news.ycombinator.com