Columbia math lecturer Peter Woit writes:
There has been significant grade inflation over the years, so having a transcript with a string of As isn’t worth what it once was. This is not good for the unusually talented, who now need to find other ways to distinguish themselves.
That’s a good point! I’ve typically thought of grade inflation in isolation (as in my post asking why weren’t the instructors all giving all A’s already?) with the problem being that inflated grades provide less information to future employers.
Woit’s point is related but goes further. Now that A’s are given out like candy corn in the world’s worst Halloween party, they don’t provide much signal, first because, as Woit says, non-unusually-talented students can also get strings of A’s on their transcripts, and also because if you’re competing on grades, the occasional slip can be so costly. Either way, ambitious students have to distinguish themselves in other ways—for example, by publishing articles in journals and conferences. This propagation of “publish or perish” down to the high school level just exacerbates the explosion of publications—apparently, zillions of medical students are kinda required to publish research too, and if publication is a requirement, then the quality is not gonna matter so much, and these papers just get stirred in with whatever remaining legitimate literature is being produced.
So, yeah, if we were to give out more B’s and C’s, maybe the world would be a better place.
I’m not planning to first, though. As I wrote a few years ago, the real mystery to me is not, Why is there grade inflation?, but rather, Why is there any room left to inflate: why weren’t the instructors all giving all A’s already?
At that time, I recommended statistician Val Jonhson’s plan to “make post-hoc adjustments to assigned grades to account for differences in faculty grading policies”—basically, fit a multilevel item-response model to estimate students’ latent abilities based on their grades. As I wrote at the time:
The beauty of Val’s approach is that it does three things:
1. By statically correcting for grading practices, Val’s method produces adjusted grades that are more informative measures of student ability.
2. Since students know their grades will be adjusted, they can choose and evaluate their classes based on what they expect to learn and how they expect to perform; they don’t have to worry about the extraneous factor of how easy the grading is.
3. Since instructors know the grades will be adjusted, they can assign grades for accuracy and not have to worry about the average grade. (They can still give all A’s but this will no longer be a benefit to the individual students after the course is over.)
I still like Val’s idea, but at this point there may be too much grade inflation at some schools for it to work. At some point there is so little signal left that you can’t recover the information you want.
OK, at this point you might say, sure, grades are B.S., whatever. But that puts in the worse position of implicitly requiring students to have other qualifications. At best, this sends students to interesting research projects and internships, but many times it just pushes them into trying to hop on projects to get credentials. Rather than writing some crappy Neurips paper and then learning the tricks to get it accepted, I think they’d be better off taking interesting courses in college, working hard, doing well on exams, and writing good term papers.
If almost all grades are A’s, then classes essentially become pass/fail (with a B tantamount to an F). Ideally, this would lead to employers taking a ore holistic, in-depth approach to evaluating candidates. However, it seems like they have taken the binary thinking (have they published? Yes/no, doesn’t matter how good the paper actually is). So, this seems more like a problem with society/culture/markets, a problem that a fancy statistical model can’t fix.
Adede:
I don’t think that a list of classes that a student has passed will be very useful in allowing employers to “take a more holistic, in-depth approach to evaluating candidates.” The problem is that passing provides so little information. It’s too easy to pass a class. And then there’s reliance on letters of recommendation–that’s really horrible.
> list of classes that a student has passed
I think that *could* provide some information, if the student had considerable choice in which courses to take. It would at least give a sense of what the student was interested in, and maybe whether they had the training to develop unique perspectives in some domain. For example, a student who only had technical courses on their transcript might be someone who is really good at solving problems that are laid before them, but maybe not so good at communication or creativity or critical thinking. On the other hand, maybe that same student is reading a bunch on their own, plays in several bands, and volunteers in science outreach on Saturdays.
So I guess I just refuted my own example! The ambiguity I described in the previous paragraph could, in principle, be resolved by the student’s personal statement or letters. But as Andrew says, letters of rec are usually pretty surface-level. And, in my experience, most personal statements read like they were produced by a chatbot even before we had ChatGPT. So it seems like very few of these formal channels carry much useful information.
This also depends on the university system. If someone earned their degree in a highly technical/mathematical subject, but obviously picked the easiest and softest possible courses then that is a red flag. In the American system you are of course forced to take more courses on soft topics (nothing fundamentally wrong with that).
I have interviewed some people, who looked impressive based on their CV but transcripts of studies suggested something was off, and every single time they had trouble answering technical questions.
I meant something more like employers saying “If the transcript is just a list of A’s, then we should evaluate applicants in depth instead of just reducing them to a transcript.” But instead they just move on to reducing applicants to a list of papers (not bothering to read the papers), or other bullet lists of exrracirriculars. And it’s too bad so many application processes have this superficial, assembly-line approach that is kind of dehumanizing and also puts everyone on a Goodhart’s Law treadmill.
What does “holistic in-depth approach” mean? Do you think there is wide-ranging consensus on paper quality? How do you know that employers don’t read applicants’ papers? have you ever been on a hiring committee?
Your griping about dehumanization is amusing
I don’t think that a list of classes that a student has passed will be very useful in allowing employers to “take a more holistic, in-depth approach to evaluating candidates.” The problem is that passing provides so little information. It’s too easy to pass a class. And then there’s reliance on letters of recommendation–that’s really horrible
I think this a fairly useless way to conceptualize the problem. I don’t think grade inflation is a problem so much as the very practice of grading. It largely reinforces a passive approach to the most important aspect of education – empowering learners to have executive control over their own learning. It says to the student they don’t have ultimate responsibility for assessing their own learning.
As most of these comments underline, there’s nothing intrinsically valuable about grades because they’re necessarily dependent on a cohort which is significantly arbitrary. And even behind that, why is it useful to judge a student relative to other students.
Meaningful assessment should be criterion-refeemced not norm-referenced.
Regarding grade inflation, this really did happen when I was a college freshman in the early 1950s: At a large gathering of freshman students, a dean, with great pride, said to us, “Look around you and note how many of you won’t be here next year.” He was extolling the high standards of the institution and his measuring stick (shtick?) was the number/percentage of students who were “weeded out” because of low grades.
Then came the G.I. bill and the increased temptation to look upon it as a cash cow and a source of growth and prestige. Another way, a sort of new way, to indicate growth and prestige is to extol the average GPA and this resulted in an ever-upward climb such that the measuring stick was turned upside down with the average grade slamming into its upper limit. Needless to say, no dean in his right mind would today take (public) pride in telling freshmen that the money spent this year was cash down the drain.
Now, it’s the “college rankings”, rather than the GI bill, that’s driving colleges to lie on their college statistics (like Columbia) or inflate grades so their graduates compare well and get good jobs. With the ongoing trend towards fewer students going on to college (and colleges shutting down in response), schools need to attract students, train them well enough, and then promote their success. When applying for jobs or graduate school, anything short of grade inflation is a penalty.
That’s an interesting perspective. Although I think there is something to be said for the counterpoint, that a high rate of flunk outs may be indicative of poor pedagogy.
Am the founder of a tech. company. We try and hire as many people as we can. This is a difficult process (which is why we’re always trying to hire). In any case, we often hire new college graduates. Yes, we’ve observed and quantified that there’s little difference in grades between most students. We’ve also observed that there’s little difference between almost all college graduates in how they pass the quantification tests (which are required to get hired) between almost all students. That is, if, say, you graduate from Columbia university and are an excellent student, there’s almost no difference in your predicted ability to pass the quantification test from an above average student who graduated with also a CS degree from, say, Ohio State university.
Am mentioning this for two reasons: 1) the students who graduate from Harvard, MIT, Stanford, Caltech, the university of Illinois (Urbana) and the university of Wisconsin (Madison) are predictably much better than every other university. That is, they’re about one standard deviation more likely to pass the quantification tests than any other student from any other college. So, if you assume that every university and every department has grade inflation, there appears to be a quantifiable way to discern, at least for my company, a way to distinguish between how much students learn. (It’s either this or the students who are admitted to these universities and apply at my company are more likely to learn what’s needed to pass the quantification test. That is, these students would pass regardless of which university they attended.)
2) Because it appears there’s little difference between students (and very few from the “elite” colleges apply to my company), we’ve done away with the requirement to have a university degree and replaced it with a series of quantification tests and a three-month evaluation, to see the perspective employees ability to code and fit in with the respective team. Interestingly, the only people who apply for this are those who are about to graduate from high school. As far as I know, no one else has applied.
This is fascinating — thanks! About college graduates not applying for the tests + evaluation period path: Are they aware it exists? (Especially students with non-CS degrees?)
A more interesting question, but related to the topic of awareness of alternative paths: do you think your tests would be much different than those required by other somewhat similar tech companies? If not, could coordinating other companies make it easier to implement these, and also easier to advertise that this path exists?
Raghu:
There’s a standardized testing protocol called Leetcode that, as far as I’m aware of, almost all tech. companies use. We also use it, with some slight modifications.
Regarding if non-CS degree holders/non-college graduates are aware of how we hire, everything we do when we contact or interact with potential employees or customers, we always a/b test each email, social media post, interaction with a journalist, etc., when we advertise for job openings. That is, we’re always looking to reach as many people as we can. Also, we use an algorithm called community detection, to track how consumers use our products and also who responds to our job posts. The purpose of doing so is to initially interact with people or media outlets that will have the most impact (which is defined within the context that we’re interested in). This is just a clustering algorithm. Then, the short answer, it’s VERY difficult to get Americans to be interested in quantification tests. Therefore, it’s very difficult to hire qualified people, even with the high salaries we offer.
Thanks! I have a few questions that are unlikely to be of general interest. If you don’t mind my sending them to you, can you send me an email? [raghuveer.parthasarathy … gmail] If you’d rather not, that’s certainly fine!
Re the forced publication issue:
For the last ~10 years I’ve helped medical residents with statistics in their journal club articles. Occasionally one will ask for help with a research abstract or paper. I won’t say this happens every time, but often they start with asking how to calculate a p-value. That turns out not to be so easy if there’s no null hypothesis, no control group, etc. One was proud of being listed as a coauthor of a publication on some tissue imaging technique but IIRC didn’t know whether the technique was optical or magnetic, etc.
The academic publication obsession has been a problem for many years. So much wasted effort! It’s disappointing to see it being spread to undergrads, med students, etc. They’d be better off doing almost anything else, e.g. taking a nap.
I’ve also been helping residents with research for the past 15 years, in my medical field. And I too would say that the vast majority of students (and medical faculty, both senior and junior) want to know how to calculate a p-value with their data.
But thinking back to my knowledge base from a couple decades ago, I too didn’t grasp how to analyze results correctly either. What little that was taught in undergraduate school about evaluating experiments did focus on p-values. Medical school didn’t help at all.
So even 25 years ago, resident time could be better spent actually learning about medicine (and how to comprehensively design and evaluate research) than by trying to publish something. That doesn’t seem to have changed. But with grades providing less information, applicants are correctly adapting and focusing effort where it can help with future jobs, fellowships, etc.
When I was an undergraduate at Columbia, we had a Physics professor named Joaquin Luttinger. The year before I took his course in Lagrangian Mechanics, the average grade was below a C—not an exaggeration—and the students who selected the course were already a smart group. Luttinger was an unusually tough grader, both by the standards of the school and even within the Physics department. I suspect his grading style was, in part, a departmental strategy to winnow down further the small group of students pursuing the major.
The issue we faced—though this wasn’t true in the Math department at the time—was that Physics courses were graded on an unusually low curve. Meanwhile, in Political Science, for example, the average grade was already hovering between an A- and a B+. As a result, when your GPA was calculated, you were penalized compared to students in other departments, even though you were taking harder courses with arguably some of the brightest students in the school. Employing some mechanism to weight GPA by both the class composition and course difficulty seems fairer and more useful than what exists today.
Regarding the comment above on Sam about hiring, At the tech companies I ran, I always found that the top 2% of most schools were better than the middle of, say, Harvard and Columbia. The only exceptions were Cal-Tech and MIT, where the cutoff was further down.
I’ll note that the frustration for me from a student’s perspective (master’s student planning to apply to PhD’s) is the incentive is to take somewhat easier classes than I could handle or would most enjoy because it’s better to get an A in the second hardest version of X that’s available than risk an A-/B+ in the hardest. This despite the fact that taking a harder class and not acing it would almost certainly teach more (sometimes a lot more) than taking the easier version and getting an A.
Anon:
Relatedly, I often get emails from students saying they want to work on a research project with me. I always recommend they first take a couple of my classes.
Grade inflation causes numerous problems. Here are more: Since it is easier to achieve high grades, students who earn good grades today learned less compared to students who earned the same grades a decade ago. The content covered in courses has often been simplified, reduced, or streamlined, to achieve the good grades.
Furthermore, since more students pass high school, more go to university. Many of them may not really belong there. Professors are under pressure to make sure a high percentage passes, so standards are lowered. In Germany since 2011, around 55% of young people enroll at a university, up from 35% in 2006. (Although other factors matter here as well. For example in 2011 mandatory military service was abolished)
This may lead to a suboptimal allocation of workers. Jobs where no university degree is required, lack skill. And many who do study may have difficulties finding a job.
At least in the US, there are not more students going to university. There’s a definite decline in enrollment leading to some schools closing. It’s estimated that by 2041, there will be 14% fewer. This is partly due to demographics and partly political.
I’d argue that some (hopefully many) students aren’t learning just the bare minimum to reach their “grade”. But perhaps that’s just my hope. That could hurt finding or keeping a job.
Finally, the content covered doesn’t need to be watered down to lead to good grades (again maybe only in the US). Grades are chosen by the teacher, so grade inflation can happen by adjusting the curve. Or if grades are chosen by a standardized test, the test covers everything deemed important (chosen by a national body) so the content cannot be simplified, reduced, or streamlined without really hurting grades. Instead, if grades are test-derived, then the incentive is for classes to teach to the test and skip anything not tested.
In the last few years the trend has reversed, but over the last few decades *vastly* more students in the US are going to university. It is over this scale that this topic is being discussed.
I have an interesting perspective on this issue, since I taught for many years at Evergreen State College, which doesn’t have grades but instead a rather demanding narrative evaluation process. You’d think the requirement they write a page or so of detailed reportage on student learning would force professors to be at least sort of honest, but no. OK, many (myself included, I hope) were, but quite a few wrote puff pieces for nearly all their students. I think this is less effective in written narrative, because, if you’re going to, say, praise a student’s paper, to do it with impact means saying something specific about what made it noteworthy. But still there was the strewing of garlands of vacuous praise by lots of faculty, effective or not. My perception was that it was part of an unwritten compact: students wrote positive evaluations of faculty, and vice versa. And there was less friction that way too.
The difference was that a conscientious, positive written evaluation could make a big difference for a student all on its own. The rate of acceptance into grad programs was high, partly for this reason and also because the evaluation system forced faculty who took their jobs seriously to pay close attention to students and their work. The bottom line is that eval inflation was a thing, but not as damaging as grade inflation.
Educators need to be specifically taught how to evaluate students. Just like writing effective letters of recommendation, writing an effective evaluation takes time and work. Without putting in the work, you get form-like recommendation letters or unhelpful, terse written evaluations.
Out of curiosity, what is the timing for the narrative evaluations at Evergreen State College? Are you able to provide feedback scattered throughout the course so that students can adapt and improve? Any recommendations on how to do this well (or encourage fellow educators to make the effort)?
There’s a midterm evaluation and then a final. The midterm doesn’t go on the transcript; it’s just a progress report. As for recommendations, I can just say what I did: I kept records of student performance on each piece of work, organized so that, at the end of the term, I could peruse what each student did separately. (And written comments should be as thorough as possible on exams and papers, for ongoing feedback as well as facilitating the final evaluation.) I also kept a spreadsheet on which I could briefly comment on students’ in-class work, etc. But no matter how much I prepared ahead of time, writing the evals was a huge job. The whole point is to get away from blanket assessment and identify particular areas of achievement or the need for more work. Oh, and it’s crucial to get a baseline at the beginning of the term, since the evaluation should focus on progress along multiple fronts.
Peter Dorman wrote: “I could peruse what each student did separately.”
The word, “peruse”, should be banned because it is one of those nasty words which is a contranym.
transitive verb
1
a
: to examine or consider with attention and in detail : study
b
: to look over or through in a casual or cursory manner
The web site presents a lengthy list which, quite properly, also includes the word “sanction”
https://en.wiktionary.org/wiki/Appendix:English_contranyms
but also includes other words which don’t quite fit because, unlike “sanction” or “peruse”, the pronunciation or spelling or accenting is different for the opposite meaning..
Paul:
In the immortal words of Run-DMC, “not bad meaning bad, but bad meaning good.”
Paul, every time you demand someone change their verbiage (I think you were trying to censor “unknown unknowns” last time) it makes me want to do them four times as much.
Medical residency training is being hit from multiple angles, grade inflation being only one of them. This past year, the residency program I help with had 700+ applications for 5 positions. In years past, some programs would use standardized-score cutoffs or GPA cutoffs to whittle down the number of applications read.
Grade inflation is an issue, but the vast variation in grading paradigms has always been a bigger problem. Schools don’t have identical curricula, courses, or rotations. On that 4 week rotation with orthopedics, did you have simple patients (with no other medical problems distracting you from excelling in ortho) and your proctor was on their 1 month-a-year inpatient shift so needed all the help they could get? If so, your grade is an “A”. Unless they were the “old-school” doc who worked 60 hours a week “in his day” and so your grade is a “C”. Everyone involved in medical education for decades knows this and so more programs are moving to be pass-fail. Some standardized tests are also now pass-fail.
So if you have 700 applications (each 20-40 pages long) to review each fall to choose a few dozen to interview, how do you do it? Some programs give weight to honors societies (but aren’t available at each school), care about pedigree (which benefits well-financed applicants who can afford top undergraduate schools and top medical schools), use outside companies (AI?) to “screen” applications for you, etc. At my program, we tap every able body to read and evaluate applications. Then multilevel modeling (including a reviewer bias term) is used to put applications on the same scale. The more application reviews with a smaller team, ranking, interviews, discussions, more ranking, …
EB:
Also think how inefficient this is, given that each of these 700+ applicants is applying to many training programs, each of which is independently and redundantly going through this reviewing process . . .
Yes! For the past few years we’ve received applications from a large percentage of applicants to this field. The national organizing body has tried to discourage applicants from applying to every program by instituting a fee per application (none of which goes to us). But if it’s your future on the line (there are more applications nationally than spots in some medical fields) then it’s only reasonable to apply to almost everywhere…it’s cheaper than finding giving up on a career in your dream medical field.
But programs do look for different things in applicants. We try to take a holistic approach and find a good set of students who would support each other and complement the department. Some programs definitely prefer to recruit from “top” schools. Some programs can train international students (some can’t because it’s expensive). Some programs want to train local students…
At least there is a common application that massively reduces the redundancy. There is one clearing house for applications and they come through as one PDF per applicant, so we don’t have to wrangle letters, faxes, emails, and other electronic submissions.
If you don’t make an holistic in-depth evaluation of all applicants, you are dehumanizing them according to a comment on this post. What’s your take on that?
It continues to amaze me that approximately everyone agrees that grade inflation is bad, but no one does anything about it. Of course, doing something would require coordination — any individual professor who single-handedly introduces standards that would be commonplace 20 years ago would just penalize his or her students; any single university would penalize its own students. And individual students think that artificially high grades benefit them, while fairly few realize that this contributes to the overall well-justified cynicism about the value of college degrees.
By the way: as always, I think these discussions are heavily weighted towards “elite” schools where a tiny fraction of US college students study. Grade inflation is also rampant at standard public universities (like mine), and an A means less than it did even 20 years ago, but paradoxically there are lots of students who fail. The difference between A/B and C/D/F is basically “does the student show up and complete assignments with some basic level of consciousness?” Surprisingly many fall in the latter camp. I teach a lot of them, and I feel sorry for them — they’ve been conned into the notion that everyone needs to go to college, and they suffer for it, at great expense.
My grades in this class
Total As: 22.6 percent
Total Bs: 18.9 percent
Total Cs: 29.2 percent
Total Ds: 11.3 percent
Total Fs: 7.9 percent
I’m not typical, and this probably hurts me.
How do you know you’re not typical (at your university, presumably)? Do you have an access to a probability distribution of grades for your department (or the equivalent, your class)? Why do you think it hurts you and doesn’t benefit you? Is this just intuition?
I should clarify the “this probably hurts me.” It doesn’t hurt me in any direct sense. I like my courses, most students seem to like them, and the ones who like them seem to like them quite a lot — I get nice notes, sometimes even years later. However, conversations with students make it clear that there are other “general education” courses that are easier in terms of assessment and workload. In a sense, we (the physics department) are competing with other departments for student course enrollment, since this determines our budget, our ability to hire, etc. For many students (not all), the ease of getting an “A” is a significant determinant of their choice of courses. So the impact on me is indirect, via my department.
This is a bit old (2017), but local trends in grades are here: https://uomatters.com/2017/02/uo-grade-inflation-accelerates-after-reform-effort-fails.html
Raghu:
Wait, you’re citing UOMatters? That scandal sheet?? I won’t believe it until I see a report directly from the provost.
Regarding your question of why nobody does something against grade inflation:
At least in Germany, the discussion around grade inflation is complicated by concerns about social inequality. Students from low socio-economic backgrounds often face greater challenges in school. To avoid the injustice of letting them fail—since failure would further reinforce social inequalities—they are given more and more chances to pass, even with poor academic performance. (This example focuses on grade inflation at the lower end of the scale, not a scenario where everyone receives top grades.) However, the root problem is that these students often grow up in families where education is not a priority. They lack role models for good study habits, and their parents are often uneducated. These systemic issues cannot be resolved simply by passing everyone.
In Germany, there are essentially three types of secondary school diplomas: an easier certificate after 9 years (Hauptschulabschluss), a standard one after 10 years (Realschulabschluss), and the university-qualifying diploma after 12 years (Abitur). After the fourth grade (or sixth grade in some federal states), students are typically assigned to one of these school tracks. In some states, parents have a say in the decision, but not everywhere. Historically, this system created relatively homogeneous classes based on students’ intellectual abilities.
Today, however, many advocate for a more inclusive and equitable approach, where all students learn together in the same environment. In these integrated classrooms, students of varying academic abilities—and even those with disabilities—attend the same classes. Interestingly, research shows that in such environments, *all* students perform worse in standardized tests. Turns out is difficult to design a course when abilities vary greatly (surprise).
If everyone gets As, then Val’s approach doesn’t work. The benefit is that it gives teachers the freedom to grade without inflation.
In my department the norm is that, if grades in an undergraduate course are too high (average > about 75 on a 100-point scale), or variance too low, the HoD has a chat to the instructor about their methods for grading. This generally means that issues with grade inflation are only a problem once per instructor/course combination, if that.
This is at a small public university in Europe, by the way, and with students who are mainly trying to get into very competitive professional graduate programmes. (Typically only the best handful of students in any cohort will manage to do so, so this approach tends to help them stand out from the crowd.)
Probably the most difficult aspect of this is that grade inflation is a massive issue in the high school system here, so students enrol expecting to get 90 + on everything and then get a rude awakening when they’re scoring in the 60s or 70s instead. Fortunately though, I don’t have to teach first-year courses, so don’t have to deal with their anger. (And sometimes their parents’, believe it or not—colleagues have reported angry calls from parents about why their kids are being given such low grades!)
> fit a multilevel item-response model to estimate students’ latent abilities based on their grades
My alma mater does this, or at least they did when I was applying. The coefficients on the high school level variables were secret and much speculated upon.
Back when I was involved in recruiting college graduates for consulting research positions (roughly from 1985-2015) we never really looked at grades anyway. You got an interview if you went to one of the proper 20 schools, took the right courses, and passed a fairly modest interview process. Rarely, the HR department would relax the process slightly to include a few more schools. A bad grade in a course was actually slightly helpful because it provided an opportunity for the applicant to discuss the experience and explain.
I’m surprised no one has mentioned the possibility of refining the grade scale by adding plus/minus. If “average” is a B+ or A-, a transcript that consists of A’s and A+’s will stand out. More than one or two institutions already take that approach–it could easily become more widely adopted.
?? Every institution of higher education already uses +/-. (i.e. grades are A+, A, A-, B+, B, …).
(Technically, I have not checked every US institution. I’ve seen a lot of transcripts, though.)
Almost no places in the US uses A+. Naturally, Harvard does.
I think the problem is that grades do not have a clearly defined purpose. A common definition of the purpose of a grade might be ‘it is meant to show the student’s level of understanding’, but that is very ambiguous (level of understanding relative to what?). Anything goes with such a definition. Another purpose of grades that I have heard quite often is that ‘grades are meant to motivate the student’. The comments above also mention a third goal of selecting the best (e.g. due to limited resources for admitting students or hiring new staff). There is some overlap between these purposes, but I do not see a way of creating a unified purpose for grades. Without a clear purpose, grades will be used and abused in all sorts of situations. I see grade inflation as a symptom of the ambiguity of the purpose of grades. Once the purpose is clearly defined, it is possible to create grading metrics that fulfil that purpose.
Related topic – what about award inflation? An institution advertises that 10% of faculty hold special chairs in recruitment materials.
Lots of great arguments for standardized testing.
What? Standardized testing? A CRIME AGAINST HUMANITY!!!! CALL THE UNITED NATIONS!!!!! IT’S IMPOSSIBLE TO MEASURE A STUDENT’S ABILITY!!!! NO MORE GRADES OR TESTS!!!!
Standardized testing is dehumanizing!
The issue isn’t standardized testing per se – it’s norm-referenced standardized testing.
It’s always remarkable to me how many sophisticated thinkers more or less just accept the value of norm-referenced testing. Although it does make sense that a crew like this, who almost certainly have all benefited from a norm-referenced testing paradigm, would assume assume it’s value as a starting point.
Sigh. When I was teaching CMPSC at PenN State in the early 1970s, grades were usually inverse curve, in course usually rated by students as most work/credit (we were missing a course, so jammed some more material into this one).
Anyway, most students got A’s or B’s, as they had to do a lot of work. If they really struggled, a C. If they cheated, and F.
I don’t know how most recruiters evaluated grades then, as my only experience was being recruited by Bell Labs, which:
a) Assigned technical management teams (not separate recruiters) semi-permanently to each school where they recruited.
b) They’d come visit several times/year, get to know faculty, take them out to dinner, ask them which students were especially good, keep an eye on them.
c) Then some day they’d show up in grad student’s office, ask them if interested and invite them to visit for interviews.
I don’t think they looked for letters of recommendation,as they already had familiarity.
===
Of course, that sort of recruiting was a bit of a luxury, we always said “monopoly money is nice.”
The good old days, back when employers actually made an effort to do recruitment, rather than expect everything to come to them on a silver platter. Unthinkable these days, like so many other things.
Another annoying thing is that tests like the SAT math are set up to distinguish between students at the lower percentiles but compression at the top scores make it difficult to separate top performers accurately. What is the point, the lower quantiles are not going to pursue or be successful in quantitative fields. That also results in misallocation of talent and resources and rewards minutia and anal perfection rather than true creativity and risk.
Under the system I had in a third world country As started at 75 percent, an A+ at 90 percent accuracy and were rare as hens teeth,but engineering slots were limited and valuable. There was no interest in outside activities or sorting anyone at the middle of the distribution. Yet it was less stressful, easier on the kids.