I think most of the better grad programs probably do. They have more consistency in personnel and fewer concentrations to evaluate. Most big employers have a lot of turnover in recruiting, though, and smaller employers (smaller meaning number of positions filled each year) don’t have much familiarly with less popular courses of study.
Any particular examples of lesser major requirements?
yeah, my GPA was much higher senior year. I had all 4’s and 3.5’s then. The other years, not so much. Lol. I was not the worlds best student to start. I had no clue how to study freshman year… I was one of those kids who didn’t need to study in high school. Like ever. I think my mom made a joke once that she never ever saw me bring a textbook home. I got A’s and B’s without really trying. Just had to pay attention in class.
Then I got to college and it was like uhhh, what? 
There was so much multiple guess (i realize it’s called multiple choice, but I dubbed it multiple guess). and I’d constantly pick wrong. No clue why. Even if I knew the right answer. You could have asked me the same questions and asked me to explain the concept and I would have done fine. I think they worded things strangely on purpose to trip people up, and they got me every darn time. LOL. In the upper courses, there really wasn’t multiple guess anymore and it was more applied, you know, projects, essays, papers, etc. Much easier… for me anyway.
I remember I had this one accounting class where the professor gave all multiple guess tests and I got a D in it. I swear I knew what I was doing. I took the course again the next semester, with a different professor, as I needed at least a C to take the next class. They changed the text book… I didn’t buy the new book because I felt I already knew accounting… and how much can accounting change from one semester to the next, anyway. All of her exams were like… actually doing accounting tasks as opposed to multiple guess. I got like a 98 in that class, without buying the book. 
Also - it’s quite funny regarding the smart/wise thing - my old boss and I were working on a project together last year and multiple occasions during it he told me that I’m “almost always the smartest person in the room” and he was adamant that this was a praise and a compliment.
I still don’t think I’ve gotten through to him that I’m not, in fact, the smartest person in the room… I might be the most experienced in some rooms… sure… (I’ve worked here 17 years)… and with that you do accumulate knowledge… but everyone has different areas of expertise and brings different knowledge to the table and we all have things we can learn from each other. We all learn new things every day.
And, although he meant it as a compliment, I personally don’t think anyone ever wants to be called the smartest person in the room. I feel like if you get to the case where you are actually the smartest person in the room, you better find a new room to work in… because what is the fun or challenging part in that?
You can’t look at the core curriculum issue in isolation. Yes, Gen Eds continue to evolve at many schools. But that fact has to be considered within the context that a) many top schools no longer give AP credit, b) seniors heading to these schools now take 37 APs and dual enrollments prior to HS graduation, and c) college now costs approximately one trillion dollars.
Pre-meds don’t want to stress over what amounts to their fourth, college level English course, not when their friends at State U are graduating in 2.5 years because the 3s they received on their AP exams all transferred for credit. Pass/Fail Gen Eds split the baby, with neither students nor administration being entirely satisfied.
It’s my understanding Harvard professors have a good amount of freedom in the grading system they choose for their classes. Some choose to curve. Most do not. I’d expect that you are most likely to find curved classes in large underclassmen courses.
My personal experience is with Stanford, which has a similar self-reported cumulative GPA to Harvard. It was my experience that it was quite common for large underclassmen foundation classes to be curved, such as freshman math or freshman science. I recall one freshman chem class where it was common for median exam grades to be in the 30-40% range, with the highest grade in a class of a few hundred students at 80-90%. Curving gave the option to have extremely challenging exams like this that require using general ideas discussed in text/class in original ways at a far high level than done in the text/class, rather than simple regurgitation of the textbook with little thought.
I believe the median grade was a B+ or so in this class. Many students seemed to spend an enormous amount of time studying for exams, which I expect had more to do with the extreme difficulty of exam questions + curve, rather than the grade distribution.
However, this type of class was more the exception that the rule. Curving was quite rare for smaller classes and upperclassmen, major-specific courses that have filtered out a large portion of students who did not get high grades in the field. Even in the large freshman courses, different professors may grade completely differently. For example, in another year, a different professor who taught the chem class mentioned above, and she gave exams where grades of >90% were the norm.
As I touched on elsewhere in the thread different classes have different grade distributions, with some having much larger portion of A’s than others. A few Ivy+ type colleges list the median GPA of the class on the transcript, which gives persons reading the transcript a reference. It also allows students to know which classes have higher and lower median grades. Cornell Reddit posters keep a crowdsourced record of the median course grades reported on transcript. A list showing median grades for 2018 is at Spring 2018 Cornell Median Grades - Google Sheets . Some of the outlier median grades for larger courses were:
AMST 2001: The First American University – Median = A+ (311 students in class)
BIO 1350: Introductory Biology – Median = B- (376 students in class)
1 Like
I’ve seen some examples in the past, but the one currently sticks out in my mind is the requirement for a CS degree at some elite colleges. These colleges have dropped the requirement for (and in other cases, reduced the rigor of) some fundamental CS theory courses. They probably would justify it on the basis that these courses aren’t needed for many of their students, but I believe they did so to make it easier for their students to graduate in CS. After all, these are more difficult courses for students majoring in CS, and there’re lots of other easier required courses that students don’t necessarily need in their careers, but they’re kept.
If this is widespread, surely you can name lots of universities where this occurred.
Cornell is one of them, I recall. Rice is another. There’re many more.
I believe professors at the most elite schools all have a great deal of discretion in grading. However, students these days have access to much more information than the example you gave about Cornell median course grades. Students know which courses are harder and which are easier, which professors are tough graders and which are more lenient. Unless the students are self motivated or the courses are required (and the same professors teaching these courses), they can avoid harder courses and certain professors.
In that case, how about switching to a logarithmic scale? 
I think the availability of grade distribution type information varies quite a bit by college. Harvard has the Q guide, which lists a variety of course information from student surveys. Several years ago, the Q guide used to include a “difficulty” rating, but the average difficulty rating is no longer reported to students. It’s my understanding that one key factor in this decision to remove this information is that some students were using the difficulty rating to find easy A courses, similar to what you describe. They still ask students a question about difficulty, but only the professors can see the students’ reported difficulty ratings. The Q guide does publish other relevant information, such as number of hours per week of work required, which shows a good amount of variation. Some example numbers from past years are below:
CS 161: Operating Systems – Average of 30 hours per week of work
PSY 18: Abnormal Psychology – Average of 2 hours per week of work
Could this be solved, by allowing students to more liberally choose “P/F” for more of those kind of “reach” courses?
My daughter just decided to “keep” a really challenging class, where she barely meets the minimal requirements - but she really likes that (esoteric) topic, and the professor is highly rated. She is not worried about not passing it, but compared to everyone else in class who will be majoring in this particular field, she’ll be at bottom of the totem pole. The alternative would have been yet another “safe”, but uninspiring, class.
So she thought about it for a while, and then decided she didn’t need another “A” that bad. If she does find herself underperforming, she’ll switch that class to a “Pass/Fail” grade, even at the expense of “one less minor” on her transcript.
Maybe offering more P/F options that are not bringing down the GPA in one’s “major” would reduce the “risk avoidance”?
1 Like
Students would certainly be more likely to risk taking courses where they have a good chance of non-A grade, if they had the option to take the courses pass/fail. However, having this option would also likely drive average GPAs even higher since a good portion of the non-A students would choose to instead take the class P/F.
Brown is a good example. Nearly all classes at Brown can be taken satisfactory / no credit, and if the student gets “no credit” , the NC grade is not reported on transcript or included in cumulative GPA calculation. Students can switch from graded to S/NC any time within the first 4 weeks.
This grading system contributes to why, In apples-to-apples comparison of GPA using comparable reporting metrics in comparable years, Brown nearly always has the highest reported average cumulative GPA among Ivy+ type colleges, which was likely near 3.7 prior to COVID. Many of the non-A students in particular classes at Brown choose to take the class S/NC, so the non-A grade does not appear on their transcript.
1 Like
But - isn’t it significantly more important for students to have taken/attempted advanced topics, or in large variety of fields, vs. everyone else’s concern that GPAs look too high?
As long as a certain set of challenging “core” courses required for a major/minor are “graded”, so that one’s level of expertise can be shown, then why no allow the rest of the courses to be “exploratory” and without grade penalty?
If I get a CS Major who absolutely excelled in the core courses that matter to the profession, then let them have their 3.9 GPA - even if the courses in Art History, Music Theory, and Middle-Age English were “only” Pass?
Assuming 3 hours of class time per week in each course not included in the numbers above, that would mean that CS 161 is equivalent of 11 semester hour credits of workload, while PSY 18 is equivalent to 1+2/3 semester hour credits of workload.
However, there could be sampling bias if the student survey was taken around the time a major project in CS 161 was due.
I think that is a great idea. Currently, many kids (not all, of course) avoid challenging classes that interest them for fear of a bad grade. If more P/F options existed (obviously not for coursework in the major) maybe kids would feel free to push themselves without fear of negative effects.
Two examples, though not from a college setting, can illustrate how grade inflation can be a good or a bad thing.
Jack Welch was famous for forcing an employee performance curve at GE where the bottom 10% of performers were fired every year. The toxic culture this created destroyed the company. The “grade inflation” management approach would have recognized their own words that they hire only the best, and that part of what made the employees the best were their skills in innovating and moving an organization forward.
Grade inflation can be bad from a K-12 perspective. Teachers are put under a microscope by parents to work miracles. “Why isn’t my child getting an A?” In many cases, a teacher isn’t in a position to tell the truth about a child not having talent, or not putting in the effort. So the students get the grade they want, and move along.
Also, the obsession with testing, and the absurd expectations for teachers to bring scores up leads to teachers not wanting to work with students that need the most help, teaching to the test, and sometimes even cheating on the tests. There is no room for cultivating creativity, inquiry, etc.
At some point in every process, human judgment is necessary. The idea that anything and everything can be measured and acted upon objectively is a fad that keeps getting stronger for some reason…to the detriment of society.
Really? I think Welch’s earnings manipulation and underpricing of the risk in GE Capital are a better explanation of what destroyed the company. I don’t think Enron using this system was a contributing factor in their collapse.
Amazon had a very similar system for many years (only abandoned once their hiring needs made it impossible to sustain) and they seem to have done OK. Up or out is ubiquitous in top law and consulting firms too.
Of course any personnel management system can be handled either well or badly. But so long as expectations are clear, up or out is not intrinsically bad. Likewise class ranking or high stakes exams are not intrinsically bad: although there may be incentives to game the system, overall it can be seen as fairer to use concrete measures of performance.
2 Likes
“Up or out” is not the same as ranking employees and dismissing the lowest x% (“stack ranking”), which can be applied to job categories where rapid upward movement is not the expectation (where rapid upward movement is the expectation, “up or out” becomes necessary because there is less room at the top).
“Stack ranking” creates incentive not to hire the strongest job applicants when hiring for peer jobs (where it is not “up or out”), since you will be competing against them to keep your job.