It's insane to me that no one's written the big expose on college GPA adjustment formulas yet. Every remotely competitive college has GPA weighting formulas that the use to adjust GPAs in an effort to correct for grade inflation and the profoundly different meaning of the same grades at different schools. They're proprietary and secret and probably play a profound role in who gets into college, which is a constant source of clicks. As far as I know they often involve curving for a given high school based on previous applicants from the same school, but nobody really knows. And there's as yet no big NYT or New Yorker or whatever investigation. They're secret formulas that decide who gets into college, and nobody cares! Baffling.
Freddie deBoer in the SB comments section?!?! What new devilry is this?
My own theory: reporters don't want to spill the beans because they think the current formulas will help their kids get into Harvard, the alternatives are just too awful to contemplate: https://www.youtube.com/watch?v=osJyfLATy1s (That or the universities are doing a "catch and kill" type of thing where the promise spots to the editors overseeing these pieces and they agree to spike them.)
I suppose my question is why would this be controversial? Grading likely varies quite widely from high school, and to be fair to applicants (and to try to identify the actual best prepared students), you’d want this.
In all likelihood, it’s probably just a regression model with the dependent variable being college gpa (of past students who have matriculated), dependent variables being high school gpa and a high school specific intercept. If they’re smart they are maybe using a random effect for high school (for regularization purposes). If they’re really smart, they’re accounting for truncation but I doubt it.
A regression model implies that you know a statistically significant number of “Y’s” for all of your “X’s”. They only know the grades of their applicants and maybe results for a few accepted students.
"statistical significance is a property of the result of an experiment."
That's what we want it to mean. But the technical definition is a property of an *analysis*, which is not the same thing as an experimental result (you can do multiple analyses on the same dataset; this is why statistical significance is a useless indicator).
Just made a similar comment. It's not particularly controversial (as opposed to pretending all high schools are equally rigorous) so the practice isn't really a juicy target for expository journalism.
Can you share more about what you know here? I work in the industry and have long assumed some kind of school-weighting algorithm, but I've never read anything or had any conversations that put a fine point on the subject.
High school GPA is just a terribly noisy indicator; there are schools where the valedictorian has a 3.8 and schools where the entire top 20 is above 4.2. Schools try to adjust for that sort of thing in various ways, and some college websites have explicit language about making those kinds of adjustments, but it's always vague. As I said above, a lot of people talk about schools tracking previous applicants from the same high schools and using that to set a curve. (Rank is valued so much because it's the only other way they have to adjust for a given school's grading tendencies.) This sort of thing has gone on for a long while, as I understand it, but it's taken on more salience as a) more and more kids are applying to elite colleges (even in an overall enrollment turndown) while schools refuse to increase enrollment, b) the SAT is being phased out for political reasons, and c) grade inflation is absolutely rampant and it just doesn't mean anything to have a raw 4.0 GPA anymore.
It's actually fairly obvious that there's a lot more elite college chasing than there was 20 years ago (where your high school's valedictorian MIGHT apply to one of the Ivies, at least if you went to a normal public high school) and it's hard to explain why.
Oh I don't think it's hard at all -- you can draw a direct line from rising income/wealth to elite-school applications. As jaw-dropping as tuition numbers are these days, there are more and more people every year who can afford them.
And since enrollment numbers basically never go up, actually getting into one of these places is perhaps the ultimate signifier of parent prestige. Drive around in a tony suburb for a day and count the number of top-25 university bumper stickers you see.
It's a common knowledge effect. Anyone can go on reddit or one of the college advice forums and learn how to tailor their application, read stories about how somebody like them thought they would never get in but they did, etc. Students used to be limited to advice from their parents and guidance counselors, but they're not anymore. Everybody knows that everybody else is applying everywhere, so they might as well do that too.
How are they tracking previous applicants? Grade inflation at most of the Ivies has also climbing rapidly, so are they doing a curve on their own grading, or are they using other metrics. If other metrics, would be fascinating to know which ones.
In general, yes, these things are all happening, but I've never seen anything to suggest they're as heavily quantified as your original comment implied. My understanding is that a student's high school could move their application from "no" to "maybe," but it's not doing anything more emphatic than that. And basically all of this is up to the discretion of individual admissions readers -- admissions departments aren't ranking high schools on some sort of master list.
I know for a fact that a least a could schools are running GPAs through literal formulas but again they all are so secretive about it I can't say for sure how pervasive any of it is
I had long assumed admissions offices do this. I guess it's...an interesting topic? But it doesn't strike me the least bit "baffling" that their methodologies haven't been a major target for journalists. If anything, the existence of this curving—which indicates US post secondary institutions are well aware that not all high schools are equally rigorous—suggests colleges haven't quite abandoned meritocratic norms to the extent argued by the right.
If I recall from my college APs days the UC system will explicitly recalculate your GPA. It has a certain list of things it deems “honors” and will award a 5.0 for an A in those classes only.
I have to say that this is not my experience at all. I have done >100 engineer evals and school + GPA is a pretty consistent predictor of interview performance. As you mention below you are using GPA as a screening metric, which is precisely the point -> below certain levels it's not even worth an initial phone screen.
We interview them for 8+ hours, 1 hour per interviewer and confirm that they have some fundamentals, are capable of solving problems, and don’t pretend to know stuff they don’t.
But you have to narrow them down somehow, if you have 50 candidates presumably you don’t spend 400 person-hours interviewing them all? What criteria do you use to choose whom to interview?
We phone screen down to a relatively small number of candidates. I’d guess we hire between 1/3 and 1/8 of candidates we bring (fly often) in for in person interviews. The variation is listed because sometimes you need to staff up, and other times you’re just looking for the superstars that every company wants to hire even when they haven’t got a rec.
As someone who's been a college professor for 26 years, I would say that my students are spending less time on their studies than 20 years ago. Even though much of the reading is posted online, few students actually complete it. Instead, they have become heavily dependent on notes, power points, and slides.
As an instructor, this makes it easier to separate the "A" students from the rest of the students. In their in-class essay exams, the "A" students have done the readings and can tell me (accurately) what Scholar X said vs. Scholar Y vs. Scholar Z. Everyone else is just spouting off generalities and summarizing stuff in the notes. They get low Bs, Cs, Ds, and even Fs.
It is what it is. I write a lot of letters for students who want to go to law school, and I always laud the students who have put the work in. For the rest, I have what I call my "mediocre but well-intentioned" letter where I damn these students with faint praise.
Damn it this explains me a bit. I was always the one who did the readings but only when the readings actually came up in class. Online I also give minimum viable effort in my research.
"As an instructor, this makes it easier to separate the "A" students from the rest of the students. In their in-class essay exams, the "A" students have done the readings and can tell me (accurately) what Scholar X said vs. Scholar Y vs. Scholar Z. Everyone else is just spouting off generalities and summarizing stuff in the notes. They get low Bs, Cs, Ds, and even Fs."
This sounds exactly like it was when I went to school way back in the stone age. Are you saying that there are fewer people getting As? Or what exactly has changed?
Interested on your thoughts about the dearth of real textbooks as it relates good work. I find online reading difficult and not as conducive to actually pulling in the content. The cost of the physical books was already way to high decades ago when I went.
This is a good time to bring up that the average full-time college student probably attends a non-flagship state school and majors in accounting or some shit.
One of the annoying things about the Discourse about college is that the vast majority of college students are attending a Cal State-type school, not an Ivy.
interesting fact, the number of accountants is actually going down. We've lost about 300,000 accountants over the last couple of years. Personally I think I've done pretty decent in the profession though
"STEM" is just so broad that I feel like it probably depends much more on what specific major you're in than what school you're at.
Computer science, specifically, seems like the kind of dumbed-down thing that would appeal to people who got told to major in STEM but aren't actually interested in STEM.
I agree. Individual majors like engineering, math or philosophy on the other side of things heavily select for the better students. In the past one could have made the assumption the guy above does.
But the monetary sweet spot for colleges is in easy, not actually so technical stem courses. The parents think STEM (copyright trademark) is a pathway to riches, the kids certainly don’t want to get their asses kicked in engineering and these are the archetypal adjunct taught courses so everybody wins.
Yeah, I wish people should be more honest about this. One of our school board members is a PhD biologist. He suggested recruiting science teachers at Masters and PhD programs. His colleagues said, "why would they go into teaching." He said - "are you kidding? A stable job in one place with excellent health insurance and retirement is much better than chasing postdoc after postdoc." We do live in a place where teachers get a decent wage and aren't subject to some of the nutso school boards we have been seeing lately.
Maybe things have changed in the last 20 years (god I'm old) but few people considered CS an easy major at my school. You could definitely reduce the difficult and only have a handful of really challenging classes, but the major had a significant number of the "hardest classes there are" as options and most people took 1-2 of them at least.
I think "CS" can mean a couple different things, and which particular one a given school prioritizes can dramatically impact how hard it is.
If CS is "coding" or "web development" then it's probably a comparatively easy major but still has decent career prospects (you just need college to get you enough skills to pass a coding screen and talk your way through an entry level interview then a decent dev can probably cruise through a basic career trajectory).
If CS is math and theory that's a good bit harder, but also probably tends to select for students who are good at that kind of thing, so I could see this path being harder without seeing a significantly higher workload in practice.
If CS is engineering-adjacent and heavily project focused, then that's where I'd expect to see heavy workloads: even if it's not quite as rigorous as some of the tougher STEM majors, but it does tend to have lots of work that consumes significant time even for strong students.
My program (grad 2011) was the latter: we didn't take quite as much math as the Electrical Engineers, but were encouraged to do a math minor that mostly brought parity there. We didn't have to do the Fundamentals of Engineering exam or anything comparable, but we did do a year long software engineering practicum capstone that was definitely equivalent to their senior design project. If you were a strong student who only needed to do graded homework it was probably a bit lighter than an EE load, but if you needed to do work out of class to master the material, we workes about as many hours, IMO.
20 years ago was about the lows in interest in CS as a major - the bubble had burst and then a few years for people who had already picked it as a major to graduate out. It was certainly hard then. Plausible that’s changed as interest has boomed and more programs have been added, although I don’t think it was easy in the 90s either.
I majored in computer science at a likely mid school. We didn't study so much as just demonstrated that our code worked. Didn't even have a formal thesis or final tests.
I was an English Columbia 2001. I never stopped studying, ever. I’d study in the library for eight hours, have dinner, and then feel guilty if I didn’t get back to it. I read hundreds of pages of books each week. And this notion that liberal arts majors are some sort of cake walk…I did not go to the college where that was so. College was so much work, I graduated in a shattered heap of broken pieces.
People complaining about liberal arts majors at serious colleges being less demanding were not a liberal arts major at a serious college. Majors like philosophy, history, english, etc. still attract many of the best students. I studied philosophy and chemistry, and O-Chem is significantly less callenging than Kant and Aristotle.
Yeah, the reading assignments for English back then were pretty intense. Ironically I was much more easily able to cram for science classes, whereas falling behind humanities reading was deadly. Though I was good at turning out humanities papers the night before, though could often be a nearly all nighter. I distinctly remember having to play with fonts for an English take home mid-term essay test because I was bigly over the word count to try and make it look normal length.
I was an English major at a liberal arts college who graduated in 2022 and definitely did not have that experience; I'm sure different schools can vary a lot but think this one has definitely changed. Professors would sometimes ask us to read a book in a week but most students just didn't, and post-covid it got a lot worse. Even in my majors-only junior seminar you could tell a lot of people were skipping much of the (considerable but reasonable amount of) reading. I think in a weird way this is probably part of why humanities majors are declining--if a field of study isn't hard, people won't take it seriously. But if you just made an English major way harder overnight you'd lose a lot of students...kind of a vicious cycle.
Likewise. In 2004 I was an English major at a third tier state school that once claimed it had "one of the top six public university business schools in Alabama" and that still meant reading like 2 novels a week on average.
I was an English major because I like reading novels, so it wasn't like the hardest work in the world, but that shit still takes time! Writing papers is time consuming too if you have even the slightest care about them being good. At my school I'm certain I could have spent a lot less time worrying about them and still gotten As. But again, this place was a far cry from Columbia.
This is why I found law school to be much easier than undergrad. The reading in law school was so much lighter than in humanities undergrad classes because in law school you are expected to discuss every case that you read in class which really limits what can be assigned each week. I came to realize that the people who thought law school was a lot of work were not doing the reading in undergrad so relatively it was more difficult because with the Socratic method of law school you had to actually do the reading. But it was honestly never more than a couple of hours a day (at UC Berkeley law in the 1990s).
Ha, yeah, I was a political science major. I find the "real world" to be a fair bit easier than college. The very stressful thing about college was that if you were sick for a few days the clock kept rolling, and it was very hard to catch up
I hear you. When I went home at the end of each semester, my temperature would immediately shoot up and I had to spend the first couple of days in bed, for four years.
I definitely have still not worked as hard as I did during a few of my college quarters except for maybe 1-3 weeks at a time during the craziest crunch times at my startups. I don’t think I’d be physically able to.
Yeah, I'm assuming this is somehow factoring in the weekend into the numbers and screwing up all of the data? Otherwise, 1.2 hours of class? I distinctly recall many days having 4 1 hour classes.
Even if you use 7 as the denominator, the only explanation is that there is rampant class cutting. I was a square who went to class, so I can't say if the average person actually only shows up to half their classes.
As a professor, I think my small classes (under 30 students) have about 10% average absences, while my large lecture classes (300 students) probably average over 50% absences.
Heck, I was a slacker that skipped more than a few classes. Still sounds low? Googling, I see a "credit hour" is supposed to be about 3 hours per week. Assuming a low target of 12 credit hours, you would still expect that to come in around 5? Which, yes, is closer to the 2.8 in this poll, but wow.
It seems like there must be something funky with the data. How are students defined? Is it only full time students? Is the data annualized or is this only when class is in session. (There really aren't that many full weeks of class during the calendar year.)
It says full time students. But I’m guessing this is from time use studies at random points of the year, so that it’s 365*1.2=438 hours a year in class. If we assume 30 weeks of actual class time (two 15 week semesters or three 10 week quarters) that works out to 14.6 hours a week in class, which actually sounds maybe a little high! So I don’t know.
I confess I would be surprised if anyone working their way through school is close to that low of a number? I think I would be fine with that as an end goal, but I was getting 35-40 hours a week just for work back in college. And I did not think of myself as coming close to struggling with work and school. Folks I knew that were struggling to work through school were working more than one job.
I suspect this is how time use surveys are usually reported, but yeah it’s definitely misleading (as it also is when you average across workers and retirees, or other heterogeneous populations).
My first observation is that the slow boring readership almost certainly studied quite a bit more than the average student. Two might not seem like a ton but I think you could have gotten by with two even in the old days if you were consistent about it.
I think I’m changing my mind. Even including the wide array of institutions that are out there 1.2 hours of classes per day (yes including the weekend but still) seems crazy.
Do we know if they sampled only while school was in session, or if they are averaging in the whole year? 1.2 hours per day in class is 438 hours per year, which is 14.6 hours a week for 30 weeks and 0 hours for 22 weeks.
I found a study with more details and similar numbers. They didn’t count summer or “holidays” but they did count weekends. Not sure whether winter break or spring break are “holidays”.
Where the heck are they getting their data? I wish students were getting over 8 hours of sleep a night, but sleep deprivation is a big health issue on campuses. And even if you no homework whatsoever, you will likely be in class ca. 10 hours a week if you are a full time student. Based on those two stats alone, I can't take the rest of this chart seriously. I have two kids in college now - one at a very demanding school and one at a middle of the road one, and these numbers make no sense to me.
Turns out they don’t do June to September, and they also throw out “holidays” but there’s still probably a good number of non-school-days (at least weekends, also probably finals week, spring break, and maybe the week or two between terms).
Just realized that this is from 2016, so predates (1) time wasted on more attention-grabbing social media, (2) remote learning induced absenteeism issues, (3) [insert your favorite weird social fad over the last eight years]. As someone who was in college at this time I buy this chart as the experience for the large sample of college students who went there without a real plan and partied for 4 years.
Oh wait this whole thing is overweighting weekends anyway? "To ensure good measures of time use on weekdays and weekend days, the sample is also split evenly between weekdays and weekend days. Ten percent of the sample is allocated to report about each weekday, and 25 percent of the sample is allocated to report about each weekend day." Source: https://www.bls.gov/opub/hom/atus/design.htm
Probably they are trying to get good size samples for each day, and then afterwards they reweight them to make the numbers add up right. That’s what exit pollsters do to get cross tabs on minority voting - overweight minorities in the sample so that they can say something meaningful about black women age 40-50, and then reweight by turnout to get the overall number.
Do you know if that's the type of thing that BLS would do to its data set before release? Or would that just be the best practice for using it? The former means my concern isn't real, the latter means we're relying on the statistical literacy of the Heritage Foundation and...well...
Here’s a more detailed version with more explanation (and comparison for students who are parents) and it has similar numbers. So I *think* weighting is right (though they explicitly state it ignores summer and holidays - not sure if that includes spring break or random December days after finals and before Christmas.)
No-one I knew was working 14 hours / week and certainly no-one had to work longer hours than they studied. (Another illustration that the average college student differs greatly from the "debating staff" and myself)
On one hand I see the point - I had more leisure time in college than in high school or now as a salaryman. On the other hand, this smells a little off - I remember my quarter system university days consisted of four classes per quarter, typically two hours, two times a week; even if you average that over 7 days (which you shouldn't) its a lot higher than 1.2 hours per day... do people really skip that much class? Also did I actually get 8 hours + of sleep per night? No way.
I guess the other thing is there was a lot more fluctuation back then. I am confidently not working after 5pm and on weekends, the same could not be said during college; I was sometimes done with class and binging TV by noon on a Wednesday, sometimes writing a paper at 11pm on a Sunday. I prefer having structured and standard relaxation time as a working adult to the plentiful, but chaotic college schedule. My friends and I are all off tomorrow by 6pm and will be meeting up for cocktails and will not need to do any homework or projects on Saturday - that makes me happy.
Have you watched the recent 60 Minutes segment on Havana Syndrome? You've made a pretty convincing argument against that had me as a "hoax" leaner, but after the recent episode I'm back to leaning in the other direction. Thoughts?
Tying an otherwise unremarkable police chase over excessive speeding to a Russian spy operation is good TV at the very least. I'm also slightly more convinced.
I don't have a take on what the ideal proportions should be, it likely varies by student, and as long as students graduate, the precise amount of "work", however defined, I don't think matters a lot.
But beyond college, we should always strive for the next generations to work less than the previous ones, because that means we're growing richer and more productive as a society.
Cranky old Boomer here. I have four degrees, the first a B.A. in English, the other three in Mechanical Engineering, all from a large Research One public university in the Midwest. I think it's very important to look more closely at what the students are majoring in. I can guarantee that no one majoring in any technical field is devoting only an hour or two per day to homework. Perhaps for the first year, but after that they've switched to communications or social work or, yes, English. It is very frustrating to me that 95% of the discussion of higher ed in the US (especially in the NYT and WaPo) is focused on humanities majors at Ivy League universities. Dammit, the engineers, scientists, and programmers who drive our economy are overwhelmingly educated at large public universities. But how often does the mainstream media even acknowledge the existence of these schools? Personally, I don't give a damn where all you lawyers and journalist went to school, what about the engineers??
It's insane to me that no one's written the big expose on college GPA adjustment formulas yet. Every remotely competitive college has GPA weighting formulas that the use to adjust GPAs in an effort to correct for grade inflation and the profoundly different meaning of the same grades at different schools. They're proprietary and secret and probably play a profound role in who gets into college, which is a constant source of clicks. As far as I know they often involve curving for a given high school based on previous applicants from the same school, but nobody really knows. And there's as yet no big NYT or New Yorker or whatever investigation. They're secret formulas that decide who gets into college, and nobody cares! Baffling.
Freddie deBoer in the SB comments section?!?! What new devilry is this?
My own theory: reporters don't want to spill the beans because they think the current formulas will help their kids get into Harvard, the alternatives are just too awful to contemplate: https://www.youtube.com/watch?v=osJyfLATy1s (That or the universities are doing a "catch and kill" type of thing where the promise spots to the editors overseeing these pieces and they agree to spike them.)
I suppose my question is why would this be controversial? Grading likely varies quite widely from high school, and to be fair to applicants (and to try to identify the actual best prepared students), you’d want this.
In all likelihood, it’s probably just a regression model with the dependent variable being college gpa (of past students who have matriculated), dependent variables being high school gpa and a high school specific intercept. If they’re smart they are maybe using a random effect for high school (for regularization purposes). If they’re really smart, they’re accounting for truncation but I doubt it.
A regression model implies that you know a statistically significant number of “Y’s” for all of your “X’s”. They only know the grades of their applicants and maybe results for a few accepted students.
hence adjusting for truncation...
Also, there is no such thing as a statistical signficant number of "Ys" - statistical significance is a property of the result of an experiment.
"statistical significance is a property of the result of an experiment."
That's what we want it to mean. But the technical definition is a property of an *analysis*, which is not the same thing as an experimental result (you can do multiple analyses on the same dataset; this is why statistical significance is a useless indicator).
right, and I was commenting on the other guy referring to "a statistically significant sample size"
The experiment is predicting the equation, “Y=f(x)” from sample pairs of x and y.
Just made a similar comment. It's not particularly controversial (as opposed to pretending all high schools are equally rigorous) so the practice isn't really a juicy target for expository journalism.
Can you share more about what you know here? I work in the industry and have long assumed some kind of school-weighting algorithm, but I've never read anything or had any conversations that put a fine point on the subject.
High school GPA is just a terribly noisy indicator; there are schools where the valedictorian has a 3.8 and schools where the entire top 20 is above 4.2. Schools try to adjust for that sort of thing in various ways, and some college websites have explicit language about making those kinds of adjustments, but it's always vague. As I said above, a lot of people talk about schools tracking previous applicants from the same high schools and using that to set a curve. (Rank is valued so much because it's the only other way they have to adjust for a given school's grading tendencies.) This sort of thing has gone on for a long while, as I understand it, but it's taken on more salience as a) more and more kids are applying to elite colleges (even in an overall enrollment turndown) while schools refuse to increase enrollment, b) the SAT is being phased out for political reasons, and c) grade inflation is absolutely rampant and it just doesn't mean anything to have a raw 4.0 GPA anymore.
It's actually fairly obvious that there's a lot more elite college chasing than there was 20 years ago (where your high school's valedictorian MIGHT apply to one of the Ivies, at least if you went to a normal public high school) and it's hard to explain why.
Oh I don't think it's hard at all -- you can draw a direct line from rising income/wealth to elite-school applications. As jaw-dropping as tuition numbers are these days, there are more and more people every year who can afford them.
And since enrollment numbers basically never go up, actually getting into one of these places is perhaps the ultimate signifier of parent prestige. Drive around in a tony suburb for a day and count the number of top-25 university bumper stickers you see.
Well just judging from the dataset of slow boring interns you have a zero percent chance of getting a job unless you attend an Ivy League university.
It's a common knowledge effect. Anyone can go on reddit or one of the college advice forums and learn how to tailor their application, read stories about how somebody like them thought they would never get in but they did, etc. Students used to be limited to advice from their parents and guidance counselors, but they're not anymore. Everybody knows that everybody else is applying everywhere, so they might as well do that too.
I mean, it's not that complicated - more information and much easier to apply online than writing out the application and sending it in the mail.
Like, even the valedictorian at a random suburban school in Nebraska might know who all the Ivie's even were in say, 1986!
The SAT is making a comeback.
https://www.washingtonpost.com/education/2024/03/18/sat-test-policies-confuse-students/
How are they tracking previous applicants? Grade inflation at most of the Ivies has also climbing rapidly, so are they doing a curve on their own grading, or are they using other metrics. If other metrics, would be fascinating to know which ones.
In general, yes, these things are all happening, but I've never seen anything to suggest they're as heavily quantified as your original comment implied. My understanding is that a student's high school could move their application from "no" to "maybe," but it's not doing anything more emphatic than that. And basically all of this is up to the discretion of individual admissions readers -- admissions departments aren't ranking high schools on some sort of master list.
Happy to be disproven, though!
I know for a fact that a least a could schools are running GPAs through literal formulas but again they all are so secretive about it I can't say for sure how pervasive any of it is
My high school wasn't on a four point system so it would have to have a formula, so I'm not surprised college have them in general
I had long assumed admissions offices do this. I guess it's...an interesting topic? But it doesn't strike me the least bit "baffling" that their methodologies haven't been a major target for journalists. If anything, the existence of this curving—which indicates US post secondary institutions are well aware that not all high schools are equally rigorous—suggests colleges haven't quite abandoned meritocratic norms to the extent argued by the right.
If I recall from my college APs days the UC system will explicitly recalculate your GPA. It has a certain list of things it deems “honors” and will award a 5.0 for an A in those classes only.
I have to say that this is not my experience at all. I have done >100 engineer evals and school + GPA is a pretty consistent predictor of interview performance. As you mention below you are using GPA as a screening metric, which is precisely the point -> below certain levels it's not even worth an initial phone screen.
What do you look at?
We interview them for 8+ hours, 1 hour per interviewer and confirm that they have some fundamentals, are capable of solving problems, and don’t pretend to know stuff they don’t.
But you have to narrow them down somehow, if you have 50 candidates presumably you don’t spend 400 person-hours interviewing them all? What criteria do you use to choose whom to interview?
We phone screen down to a relatively small number of candidates. I’d guess we hire between 1/3 and 1/8 of candidates we bring (fly often) in for in person interviews. The variation is listed because sometimes you need to staff up, and other times you’re just looking for the superstars that every company wants to hire even when they haven’t got a rec.
That’s very thorough and informative, thank you!
As someone who's been a college professor for 26 years, I would say that my students are spending less time on their studies than 20 years ago. Even though much of the reading is posted online, few students actually complete it. Instead, they have become heavily dependent on notes, power points, and slides.
As an instructor, this makes it easier to separate the "A" students from the rest of the students. In their in-class essay exams, the "A" students have done the readings and can tell me (accurately) what Scholar X said vs. Scholar Y vs. Scholar Z. Everyone else is just spouting off generalities and summarizing stuff in the notes. They get low Bs, Cs, Ds, and even Fs.
It is what it is. I write a lot of letters for students who want to go to law school, and I always laud the students who have put the work in. For the rest, I have what I call my "mediocre but well-intentioned" letter where I damn these students with faint praise.
"spouting off generalities and summarizing stuff in notes"
They are preparing for life as commenters on discussion threads.
It's difficult work, the pay is low, but we endure.
Damn it this explains me a bit. I was always the one who did the readings but only when the readings actually came up in class. Online I also give minimum viable effort in my research.
"As an instructor, this makes it easier to separate the "A" students from the rest of the students. In their in-class essay exams, the "A" students have done the readings and can tell me (accurately) what Scholar X said vs. Scholar Y vs. Scholar Z. Everyone else is just spouting off generalities and summarizing stuff in the notes. They get low Bs, Cs, Ds, and even Fs."
This sounds exactly like it was when I went to school way back in the stone age. Are you saying that there are fewer people getting As? Or what exactly has changed?
Interested on your thoughts about the dearth of real textbooks as it relates good work. I find online reading difficult and not as conducive to actually pulling in the content. The cost of the physical books was already way to high decades ago when I went.
Why do you think this is? Is High School less rigorous or demanding? Or is it phones and social media? Something else?
I graduated college less than ten years ago and I don't think I knew anyone who slept 8.8 hours per night...
When you are hung over…
This is a good time to bring up that the average full-time college student probably attends a non-flagship state school and majors in accounting or some shit.
One of the annoying things about the Discourse about college is that the vast majority of college students are attending a Cal State-type school, not an Ivy.
interesting fact, the number of accountants is actually going down. We've lost about 300,000 accountants over the last couple of years. Personally I think I've done pretty decent in the profession though
https://www.wsj.com/articles/accounting-quit-job-security-675fc28f?st=o6hf8l0joka4ryv&reflink=desktopwebshare_permalink
I left audit for data science but am thinking of going back to accounting at some point.
I set up a power BI solution for our company. I very much enjoyed it.
I've been trying to do some Coursera on AI as well
I'd think that AI would shrink the field a lot.
Stem major at a top school = heavy studying
Stem major at a mid/low school = moderate studying
Liberal arts major at a top school = some studying
Liberal arts at a mid/low school = very little studying
This definitely used to be true but I think the unanimous push of all colleges into stem subjects has heavily diluted the distinction.
There are now so many mediocre students who are doing computer science because they think it’s marketable.
"STEM" is just so broad that I feel like it probably depends much more on what specific major you're in than what school you're at.
Computer science, specifically, seems like the kind of dumbed-down thing that would appeal to people who got told to major in STEM but aren't actually interested in STEM.
I agree. Individual majors like engineering, math or philosophy on the other side of things heavily select for the better students. In the past one could have made the assumption the guy above does.
But the monetary sweet spot for colleges is in easy, not actually so technical stem courses. The parents think STEM (copyright trademark) is a pathway to riches, the kids certainly don’t want to get their asses kicked in engineering and these are the archetypal adjunct taught courses so everybody wins.
Yeah, I wish people should be more honest about this. One of our school board members is a PhD biologist. He suggested recruiting science teachers at Masters and PhD programs. His colleagues said, "why would they go into teaching." He said - "are you kidding? A stable job in one place with excellent health insurance and retirement is much better than chasing postdoc after postdoc." We do live in a place where teachers get a decent wage and aren't subject to some of the nutso school boards we have been seeing lately.
Yeah, that's what I'm getting at. Colleges responded to the demand for STEM by creating the STEM equivalent of sociology majors.
Maybe things have changed in the last 20 years (god I'm old) but few people considered CS an easy major at my school. You could definitely reduce the difficult and only have a handful of really challenging classes, but the major had a significant number of the "hardest classes there are" as options and most people took 1-2 of them at least.
I think "CS" can mean a couple different things, and which particular one a given school prioritizes can dramatically impact how hard it is.
If CS is "coding" or "web development" then it's probably a comparatively easy major but still has decent career prospects (you just need college to get you enough skills to pass a coding screen and talk your way through an entry level interview then a decent dev can probably cruise through a basic career trajectory).
If CS is math and theory that's a good bit harder, but also probably tends to select for students who are good at that kind of thing, so I could see this path being harder without seeing a significantly higher workload in practice.
If CS is engineering-adjacent and heavily project focused, then that's where I'd expect to see heavy workloads: even if it's not quite as rigorous as some of the tougher STEM majors, but it does tend to have lots of work that consumes significant time even for strong students.
My program (grad 2011) was the latter: we didn't take quite as much math as the Electrical Engineers, but were encouraged to do a math minor that mostly brought parity there. We didn't have to do the Fundamentals of Engineering exam or anything comparable, but we did do a year long software engineering practicum capstone that was definitely equivalent to their senior design project. If you were a strong student who only needed to do graded homework it was probably a bit lighter than an EE load, but if you needed to do work out of class to master the material, we workes about as many hours, IMO.
20 years ago was about the lows in interest in CS as a major - the bubble had burst and then a few years for people who had already picked it as a major to graduate out. It was certainly hard then. Plausible that’s changed as interest has boomed and more programs have been added, although I don’t think it was easy in the 90s either.
I majored in computer science at a likely mid school. We didn't study so much as just demonstrated that our code worked. Didn't even have a formal thesis or final tests.
Surely you spent a lot of time coding/debugging?
Many of the classes were "here's dozens of programs we want you to build. Your grade goes up the more of them you complete.".
Those classes basically amount to a trade school education. Which is fine, I think. "What education should be" is a weird topic.
I was an English Columbia 2001. I never stopped studying, ever. I’d study in the library for eight hours, have dinner, and then feel guilty if I didn’t get back to it. I read hundreds of pages of books each week. And this notion that liberal arts majors are some sort of cake walk…I did not go to the college where that was so. College was so much work, I graduated in a shattered heap of broken pieces.
People complaining about liberal arts majors at serious colleges being less demanding were not a liberal arts major at a serious college. Majors like philosophy, history, english, etc. still attract many of the best students. I studied philosophy and chemistry, and O-Chem is significantly less callenging than Kant and Aristotle.
I still hate Kant for being so fucking long winded.
Students at the non-serious college, asking for a reading assignment: “Do you have anything light?”
Professor: “Yes, how about this leaflet, ‘Famous Jewish Sports Legends’?”
Yeah, the reading assignments for English back then were pretty intense. Ironically I was much more easily able to cram for science classes, whereas falling behind humanities reading was deadly. Though I was good at turning out humanities papers the night before, though could often be a nearly all nighter. I distinctly remember having to play with fonts for an English take home mid-term essay test because I was bigly over the word count to try and make it look normal length.
Or "The Wit of Margaret Thatcher"
Or "Great English Lovers"
Is that you, Basil?
I was an English major at a liberal arts college who graduated in 2022 and definitely did not have that experience; I'm sure different schools can vary a lot but think this one has definitely changed. Professors would sometimes ask us to read a book in a week but most students just didn't, and post-covid it got a lot worse. Even in my majors-only junior seminar you could tell a lot of people were skipping much of the (considerable but reasonable amount of) reading. I think in a weird way this is probably part of why humanities majors are declining--if a field of study isn't hard, people won't take it seriously. But if you just made an English major way harder overnight you'd lose a lot of students...kind of a vicious cycle.
People who don’t do the readings shouldn’t graduate.
They shouldn't bother attending at all.
Likewise. In 2004 I was an English major at a third tier state school that once claimed it had "one of the top six public university business schools in Alabama" and that still meant reading like 2 novels a week on average.
I was an English major because I like reading novels, so it wasn't like the hardest work in the world, but that shit still takes time! Writing papers is time consuming too if you have even the slightest care about them being good. At my school I'm certain I could have spent a lot less time worrying about them and still gotten As. But again, this place was a far cry from Columbia.
This is why I found law school to be much easier than undergrad. The reading in law school was so much lighter than in humanities undergrad classes because in law school you are expected to discuss every case that you read in class which really limits what can be assigned each week. I came to realize that the people who thought law school was a lot of work were not doing the reading in undergrad so relatively it was more difficult because with the Socratic method of law school you had to actually do the reading. But it was honestly never more than a couple of hours a day (at UC Berkeley law in the 1990s).
Ha, yeah, I was a political science major. I find the "real world" to be a fair bit easier than college. The very stressful thing about college was that if you were sick for a few days the clock kept rolling, and it was very hard to catch up
I hear you. When I went home at the end of each semester, my temperature would immediately shoot up and I had to spend the first couple of days in bed, for four years.
I recovered from my college experience—checks calendar—last week.
I definitely have still not worked as hard as I did during a few of my college quarters except for maybe 1-3 weeks at a time during the craziest crunch times at my startups. I don’t think I’d be physically able to.
Two comments:
1. This is quite seriously out of date. 8 years old data with the pandemic in the middle.
2. I’d like to see more about the methodology. Whatever else you think, close to 9 hours of sleep on average doesn’t strike me as plausible.
9 hours of sleep is insane. My daughter is 7 and she get's 10 hours and it seems like SO MUCH SLEEP. 9 in undergrad just doesn't feel possible.
if you're binge drinking, you spend more time asleep to make up for it since the quality of sleep is worse.
It’s kind of a weird thing to do per day since classes are like 4-5 days per week
Yeah, I'm assuming this is somehow factoring in the weekend into the numbers and screwing up all of the data? Otherwise, 1.2 hours of class? I distinctly recall many days having 4 1 hour classes.
Even if you use 7 as the denominator, the only explanation is that there is rampant class cutting. I was a square who went to class, so I can't say if the average person actually only shows up to half their classes.
As a professor, I think my small classes (under 30 students) have about 10% average absences, while my large lecture classes (300 students) probably average over 50% absences.
I didn’t always do the reading, but skipping class seemed crazy to me. That’s what I’m paying for!
Wow! I’m surprised to see that high rate in college!
Heck, I was a slacker that skipped more than a few classes. Still sounds low? Googling, I see a "credit hour" is supposed to be about 3 hours per week. Assuming a low target of 12 credit hours, you would still expect that to come in around 5? Which, yes, is closer to the 2.8 in this poll, but wow.
It seems like there must be something funky with the data. How are students defined? Is it only full time students? Is the data annualized or is this only when class is in session. (There really aren't that many full weeks of class during the calendar year.)
It says full time students. But I’m guessing this is from time use studies at random points of the year, so that it’s 365*1.2=438 hours a year in class. If we assume 30 weeks of actual class time (two 15 week semesters or three 10 week quarters) that works out to 14.6 hours a week in class, which actually sounds maybe a little high! So I don’t know.
Yeah if they're dividing by 7, work+education is 35 hours a week, which doesn't strike me as an indictment of the laziness of the generation.
I confess I would be surprised if anyone working their way through school is close to that low of a number? I think I would be fine with that as an end goal, but I was getting 35-40 hours a week just for work back in college. And I did not think of myself as coming close to struggling with work and school. Folks I knew that were struggling to work through school were working more than one job.
I suspect this is how time use surveys are usually reported, but yeah it’s definitely misleading (as it also is when you average across workers and retirees, or other heterogeneous populations).
My first observation is that the slow boring readership almost certainly studied quite a bit more than the average student. Two might not seem like a ton but I think you could have gotten by with two even in the old days if you were consistent about it.
Yeah, that works out to about 11 hours a week which sounds about right?
I think I’m changing my mind. Even including the wide array of institutions that are out there 1.2 hours of classes per day (yes including the weekend but still) seems crazy.
Do we know if they sampled only while school was in session, or if they are averaging in the whole year? 1.2 hours per day in class is 438 hours per year, which is 14.6 hours a week for 30 weeks and 0 hours for 22 weeks.
I found a study with more details and similar numbers. They didn’t count summer or “holidays” but they did count weekends. Not sure whether winter break or spring break are “holidays”.
https://www.bls.gov/opub/mlr/2019/article/pdf/balancing-college-and-kids-estimating-time-allocation-differences-for-college-students-with-and-without-children.pdf
Where the heck are they getting their data? I wish students were getting over 8 hours of sleep a night, but sleep deprivation is a big health issue on campuses. And even if you no homework whatsoever, you will likely be in class ca. 10 hours a week if you are a full time student. Based on those two stats alone, I can't take the rest of this chart seriously. I have two kids in college now - one at a very demanding school and one at a middle of the road one, and these numbers make no sense to me.
My guess is that they are sampling at random points of the year, including when school is not in session, and averaging it all together.
Turns out they don’t do June to September, and they also throw out “holidays” but there’s still probably a good number of non-school-days (at least weekends, also probably finals week, spring break, and maybe the week or two between terms).
Just realized that this is from 2016, so predates (1) time wasted on more attention-grabbing social media, (2) remote learning induced absenteeism issues, (3) [insert your favorite weird social fad over the last eight years]. As someone who was in college at this time I buy this chart as the experience for the large sample of college students who went there without a real plan and partied for 4 years.
Oh wait this whole thing is overweighting weekends anyway? "To ensure good measures of time use on weekdays and weekend days, the sample is also split evenly between weekdays and weekend days. Ten percent of the sample is allocated to report about each weekday, and 25 percent of the sample is allocated to report about each weekend day." Source: https://www.bls.gov/opub/hom/atus/design.htm
That makes no sense to me. Why would you not just use 1/7 for each? You don't need to estimate how much of a week any given day takes up.
Probably they are trying to get good size samples for each day, and then afterwards they reweight them to make the numbers add up right. That’s what exit pollsters do to get cross tabs on minority voting - overweight minorities in the sample so that they can say something meaningful about black women age 40-50, and then reweight by turnout to get the overall number.
At least, I hope that’s what’s going on.
Do you know if that's the type of thing that BLS would do to its data set before release? Or would that just be the best practice for using it? The former means my concern isn't real, the latter means we're relying on the statistical literacy of the Heritage Foundation and...well...
Here’s a more detailed version with more explanation (and comparison for students who are parents) and it has similar numbers. So I *think* weighting is right (though they explicitly state it ignores summer and holidays - not sure if that includes spring break or random December days after finals and before Christmas.)
https://www.bls.gov/opub/mlr/2019/article/pdf/balancing-college-and-kids-estimating-time-allocation-differences-for-college-students-with-and-without-children.pdf
No-one I knew was working 14 hours / week and certainly no-one had to work longer hours than they studied. (Another illustration that the average college student differs greatly from the "debating staff" and myself)
On one hand I see the point - I had more leisure time in college than in high school or now as a salaryman. On the other hand, this smells a little off - I remember my quarter system university days consisted of four classes per quarter, typically two hours, two times a week; even if you average that over 7 days (which you shouldn't) its a lot higher than 1.2 hours per day... do people really skip that much class? Also did I actually get 8 hours + of sleep per night? No way.
I guess the other thing is there was a lot more fluctuation back then. I am confidently not working after 5pm and on weekends, the same could not be said during college; I was sometimes done with class and binging TV by noon on a Wednesday, sometimes writing a paper at 11pm on a Sunday. I prefer having structured and standard relaxation time as a working adult to the plentiful, but chaotic college schedule. My friends and I are all off tomorrow by 6pm and will be meeting up for cocktails and will not need to do any homework or projects on Saturday - that makes me happy.
Have you watched the recent 60 Minutes segment on Havana Syndrome? You've made a pretty convincing argument against that had me as a "hoax" leaner, but after the recent episode I'm back to leaning in the other direction. Thoughts?
Whaaaat. Gonna have to make this a Monday poll question. Matt had me so convinced.
Tying an otherwise unremarkable police chase over excessive speeding to a Russian spy operation is good TV at the very least. I'm also slightly more convinced.
I don't have a take on what the ideal proportions should be, it likely varies by student, and as long as students graduate, the precise amount of "work", however defined, I don't think matters a lot.
But beyond college, we should always strive for the next generations to work less than the previous ones, because that means we're growing richer and more productive as a society.
Cranky old Boomer here. I have four degrees, the first a B.A. in English, the other three in Mechanical Engineering, all from a large Research One public university in the Midwest. I think it's very important to look more closely at what the students are majoring in. I can guarantee that no one majoring in any technical field is devoting only an hour or two per day to homework. Perhaps for the first year, but after that they've switched to communications or social work or, yes, English. It is very frustrating to me that 95% of the discussion of higher ed in the US (especially in the NYT and WaPo) is focused on humanities majors at Ivy League universities. Dammit, the engineers, scientists, and programmers who drive our economy are overwhelmingly educated at large public universities. But how often does the mainstream media even acknowledge the existence of these schools? Personally, I don't give a damn where all you lawyers and journalist went to school, what about the engineers??