Harvard Professor Concerned about Grade Inflation

<p>

</p>

<p>Your example does not quite work because the median grade is an A-, which means the percentage of As must exceed 25.<br>
[Report:</a> Grade Inflation Persists | News | The Harvard Crimson](<a href=“http://www.thecrimson.com/article/2007/5/9/report-grade-inflation-persists-over-half/]Report:”>Report: Grade Inflation Persists | News | The Harvard Crimson)</p>

<p>

</p>

<p>At Yale, their study of grading suggested something similar. The departments whose majors entered with the highest academic indexes gave out the lowest grades. So, there was a negative correlation between perceived quality of majors (students) and grades.<br>

</p>

<p>

</p>

<p>I.e. A, C, and F grades (no B or D grades)?</p>

<p>It is my surmise that the “least generous grading departments” are likely to be STEM departments where grading is based to a large extent on testing in which there are right and wrong answers, while the more generous departments are those in which grades are based on more subjective evaluation of papers. I think this is somewhat inherent in how these subjects are taught and evaluated. If 12 students in an English seminar turn in excellent papers–something that could easily happen at someplace like Harvard or Yale–it is not obvious to me why, or even how, one would decide that some of them must be given lower grades. In math, on the other hand, you can make the test harder and harder until you get more stratification. But even in math, in a small class you might have mostly, or even all, students who are able to fully master the material.</p>

<p>Note that STEM majors have sequenced prerequisites, so there is likely pressure among the faculty to give grades which realistically indicate how well the student knows the material for the purpose of using it in subsequent courses. I.e. if a lot of students got A grades in calculus 1 and then failed calculus 2, that calculus 1 instructor might get some backlash over that.</p>

<p>

</p>

<p>That is certainly true, but it is probably not the entire explanation. There are large differences across departments in the amount of grade inflation. At my own institution, there used to be much less of a difference across departments than there is now. Grades in the sciences and certain social sciences have stayed relatively constant while those in the humanities have risen (e,g., math grades stayed low while English grades rose). The explanation offered by my humanist friends is related to what you wrote, though. Specifically, they find it harder to defend a low grade than do faculty members in the sciences and certain social sciences. Yet, they had no such problem 25 years ago, so competing for students may be a (large?) part of the explanation. </p>

<p>There are some good references here, particularly the Achen- Courant paper, which studies grading at Michigan. [CONVERSABLE</a> ECONOMIST: Grade Inflation and Choice of Major](<a href=“http://conversableeconomist.blogspot.com/2011/11/grade-inflation-and-choice-of-major.html]CONVERSABLE”>CONVERSABLE ECONOMIST: Grade Inflation and Choice of Major)</p>

<p>

</p>

<p>To me, the whole concept of instituting a bell curve makes no sense. When I think about my employees at work, my goal is not that x% of them become A workers, B workers, C workers, D workers, F workers. My goal is that my instruction / coaching / management style makes them all capable of doing AND doing A work. I don’t see a thing wrong with giving everyone in a class A’s if they indeed mastered the material at a higher level, and I think it’s kind of dorky to “want” to institute a bell curve on top of a system just because some people can’t think outside of numbers.</p>

<p>Some colleges apply a fairly-strict academic sorting mechanism before they will let you out. Others apply it before they will let you in. Pretty much by definition highly-selective colleges fall in the latter category.</p>

<p>

</p>

<p>Ask a humanities professor at Harvard whether she can distinguish the work of a typical junior from that of a typical second-year graduate student. The answer will be in the affirmative. It may be that the junior who gets an A is doing work that is only “excellent” relative to the current demands (just like the math student who gets an easy test perfect). Having said that, I am not a fan of mandated curves.</p>

<p>Is Harvard populated almost exclusively by very bright people? Yes. But I would guess that even Harvard admissions wouldn’t claim that they are admitting the 2,500 smartest kids who apply, or the 2,500 highest achievers. If Harvard were accepting students purely on the basis of demonstrated academic achievement (which, by the way, I don’t think they should do), it might be reasonable to assume that most of the students should be getting As. But if you’re going to take into account things like athletic ability, other extracurricular achievements, geography, race, legacy, financial background etc, then don’t try to justify your inflated grades by claiming “we only take the best and brightest.” What you are actually doing is taking some of the brightest, and also a cross-section of far more ordinarily bright people who add to your class in a variety of other ways - not to mention people who are extremely bright but not terribly well-prepared for college, or people who are great in their concentration but not renaissance men who should be getting As in the obligatory distribution requirement course.</p>

<p>I’ve deliberately never said where I go to school, so you’ll have to take my word for it that it is good enough that I think I can speak with some knowledge about grading elite school students. Based on those experiences, I totally agree that it would be just silly to expect anything remotely close to a full A-F grade distribution. THAT would be massive deflation; there are certain baseline standards of quality, and the average paper here is not a “C” paper. It is almost as silly, however, to expect that the the vast majority of the class should be getting As, because there are meaningful differences between student papers, and I don’t just mean degrees of excellence. My best student this semester is brilliant, far above the level of even other very strong students. Some of the other students, however, are also producing A papers, and they will get As. But there are also a lot of students not producing A papers. Some of these might be A papers at other schools; I don’t think it means it has to be an A paper in this one. Some of them - perhaps the bottom 25 % - I don’t think would be As at a lot of non-elite flagships either. </p>

<p>Jonri, I appreciate your point about GPA cut-offs and law school admissions. I do wonder, however, how much of that, at least in the case of admissions, is reality and how much is perception. Have Princeton students had a much harder time getting into professional schools since grade deflation began? I also think that while some students would be hurt by deflation (or, as I like to call it, “honest assessment”), most would be unaffected and others would be helped by it - the really outstanding student whose “A” means more in a school with real standards, or the STEM student who looks worse on paper than a student with a different major.</p>

<p>

</p>

<p>I don’t know if factually they have, but the perception is that the grade deflation has hurt Princetonians trying for law schools. (P’ton students have not done as well as H and Y students in admission to top law schools in the past–or as well as Williams College students for that matter.)</p>

<p>See [Princeton</a> announces committee to review controversial grade deflation policy - The Daily Princetonian](<a href=“http://dailyprincetonian.com/news/2013/10/princeton-announces-committee-to-review-controversial-grade-deflation-policy/]Princeton”>http://dailyprincetonian.com/news/2013/10/princeton-announces-committee-to-review-controversial-grade-deflation-policy/)</p>

<p>[New</a> research on grade deflation?s impact concerns Eisgruber ?83 - The Daily Princetonian](<a href=“http://dailyprincetonian.com/news/2013/10/new-research-on-grade-deflations-impact-concerns-eisgruber-83/]New”>http://dailyprincetonian.com/news/2013/10/new-research-on-grade-deflations-impact-concerns-eisgruber-83/)</p>

<p>I think the greatest impact is probably in employment. If Candidate X has a 3.2 from H and candidate Y has a 3.0 from Princeton, I think most employers will think the H resume is more impressive. When it’s the difference between being above or below a cut-off like 3.0 or 3.5, I think folks just focus on the number.</p>

<p>Perhaps more importantly, the cynic in me believes that P’ton thinks this will hurt its “yield” among students who are admitted to HYPS. Personally, I wouldn’t be at all unhappy to weed out students who are that into grade-grubbing, but yield uber alles.</p>

<p>Excellent point post #24.<br>
The top students can master and apply the material to problems.</p>

<p>Way back when I was interviewing college grads for entry engineering jobs it was difficult to compare GPAs. We wanted the grads that earned top grades with less effort rather than those working 80 to 100 hours to earn similar results. We often steered our lunchtime casual chats towards part-time jobs, working out at the gym, or volunteer activities.</p>

<p>Lastly, I don’t really care if colleges have grade inflation. I do think they should make that information easily available for prospective students, grad schools, and employers. Because admission at some colleges will not make you look better or worse than your graduating peers.</p>

<p>

</p>

<p>Isn’t this “cured” by introducing the A+ and making it rare? A few years ago, Harvard Law reported that 12.5% of its entering class had GPAs above 4.0. Stanford has them and so graduate schools know the difference between the good student and the brilliant student at Stanford. Perspnally, I think H and Y should do the same, rather than try to deflate.</p>

<p>Yale kind of does this. It’s possible to graduate from Yale “with highest distinction” in your major. Most departments limit this to about one every 5 years. Unfortunately, you don’t get that designation until graduation, so it’s not much help if you are applying to grad school as a senior. If you take some time off, it matters to graduate programs. (I doubt most employers realize that it means that much.)</p>

<p>And, I doubt that most employers who see a 4.0 from Stanford on a resume realize that you can get a 4.0 with a B+ or 2 along the way–as long as you balance it out with a A+ or two.</p>

<p>

</p>

<p>Stanford does give 4.3 for A+ grades in calculating GPA over courses taken there:
[How</a> the General University GPA is Determined | Student Affairs](<a href=“http://studentaffairs.stanford.edu/registrar/students/gpa-how]How”>http://studentaffairs.stanford.edu/registrar/students/gpa-how)</p>

<p>However, not all universities which have A+ grades count A+ grades as more than A grades in calculating GPA over courses taken there:
[GPA</a> Calculator](<a href=“http://academicservices.berkeley.edu/advising/gpa/]GPA”>http://academicservices.berkeley.edu/advising/gpa/)</p>

<p>Professional school admissions may calculate your GPA differently from how your undergraduate school does.</p>

<p>Law schools give 4.33 for A+ grades in calculating admissions GPA:
[Academic</a> Record](<a href=“Academic Record | The Law School Admission Council”>Academic Record | The Law School Admission Council)</p>

<p>But medical schools only give 4.0 for A+ grades in calculating admissions GPA:
<a href=“https://www.aamc.org/students/download/181676/data/[/url]”>https://www.aamc.org/students/download/181676/data/&lt;/a&gt;&lt;/p&gt;

<p>ucb–my main point was simply that if you want a way to “signal” truly exceptional work, giving out A+ grades and making them rare seems to me a better way to do it than saying we have a policy of asking each department to limit As to 25% [or # of your choice] of grades awarded.</p>

<p>Whether or not an UG institution calculates an A+ as a 4.0 or 4.33, receiving an A+ indicates truly exceptional work. Grad schools know this. Law school admissions recognize this. I suspect med school admissions know this too–but choose not to include it in formal GPA calculations, since many schools don’t have A+ grades.</p>

<p>Yale’s “with highest distinction” signals the same thing, but in one field. (Although I actually know of someone who double majored and got a highest in both–and nobody knew whether anyone else had ever done it, and if so, how many people and when, since the statistics are kept by department ).</p>

<p>You need transparency in the grading system. Obviously receiving an A+ at a school that awards one every five years is not the equivalent of a school that awards multiple A+ per semester.</p>

<p>

</p>

<p>It’s definitely ridiculous to want some students to not learn the material to enforce a curve. But assigning grades in a curve doesn’t mean that you have to make it hard for students to learn, just that you have to raise the standards. The reason why I or someone else would want to have curves like this is because without them it can make our degrees seem worthless. If it’s easy to graduate with a high GPA then how can I show that I learned something or that I work hard or whatever?</p>

<p>Oh, I don’t disagree on the need for transparency in the least–I like the Dartmouth system or I should say, what I understand to be the Dartmouth system. My understanding is that Dartmouth transcripts give a letter grade and then the distribution of grades in EACH course–not overall. So, your transcript might say you got an A- in introductory microeconomics and then say there were 150 students in the class that semester and 15 got As, 45 got A-, etc.</p>

<p>But there’s no transparency now at most schools. Heck, trying to find out the grade distribution in each graduating class is darn near impossible.</p>

<p>The use of “curve” grading is presumably for the convenience of instructors, since it allows writing different tests and assignments each semester without having to closely calibrate the difficulty to match previous semesters or a fixed grading scale. Of course, it assumes that the students in the class will have a distribution of abilities matching the assumed “curve”, which may be reasonable for a large class, but less reasonable for a small class.</p>

<p>

</p>

<p>But does it really do this? </p>

<p>In the humanities, often your grade is very dependent on how well you write…even then when writing isn’t the “point” of the course. You may have read a novel and truly understood it, but when you have to explain the themes in an essay, you get a lower grade than your buddy Mike because Mike is a better writer. The person looking at your grades isn’t going to know that you worked as hard as Mike did in the class or even that that you actually understood the themes just as well as Mike did. Nothing tells him that Mike is simply a better writer than you are. And, if it’s freshman year, he may only be a better writer because he went to Hotchkiss or Andover and you went to a high school where 98% of your tests were multiple choice and only half the class went on to college.</p>

<p>My kid had a history prof who actually gave all the students in a class the option of taking the exam orally or in writing. He did this because he knew that the math & science types–especially internationals–were often disadvantaged because they didn’t write that well. But, if he gave them the option of explaining e.g.,the causes of World War I orally, some of them did a much better job. </p>

<p>I thought that was an interesting approach.</p>

<p>

You need to look at the data and decide for yourself.
[url=<a href=“http://www.gradeinflation.com/Harvard.html]Harvard[/url”>Harvard]Harvard[/url</a>]
1889 2.46
1967 3.00
2005 3.45
You are not suggesting today Harvard students are a lot smarter than their predecessors, are you?</p>