Home Math Teaching About Me

Displeasure with Higher Education

Date Written: October 20, 2025; Last Modified: October 20, 2025

Especially in recent years, there have been a plethora of debatable policies surrounding colleges in the United States. At least two of these — the roles of diversity, equity, and inclusion and of affirmative action in colleges — revolve around balancing a social responsibility to address institutional inequity and the value of meritocratic integrity in educational institutions.

I think this is such a divisive topic because holding a college degree is strongly correlated (and almost surely causally linked) to income, health, socioeconomic mobility, and more. In a morally consistent society, the college degree is an indicator of virtuous traits and signifies that the degree holder “deserves” the societal privileges higher education entails. Thus to deny an individual the mere opportunity to earn a degree, whether it is in the name of equity or integrity, can be interpreted as a moral failure of our institutions.

Some arguments are indeed made along these lines. In a criticism of a criticism of the Supreme Court’s ruling on Students for Fair Admissions v. Harvard, an opponent of affirmative action sensationally exposits, “At Harvard, for example, Asian students in the top decile of academic performance had the same chance of admission as black students in the fourth-lowest decile.” This suggests a moral wrongdoing of Harvard on meritocratic grounds by highlighting a considerable difference in achievement. On the other side, an article appearing in the National Education Association’s online magazine cites flaws in measuring aptitude and merit to argue in favour of affirmative action: “Even what some people might consider neutral indicators of merit — like standardized test scores or extracurricular activities — are often influenced by factors that fall along racial lines, advocates point out.”

The point is that people in general consider American educational institutions to be fundamentally meritocratic,¹ and just as admission into a prestigious college is seen as a reflection of a student’s achievement, a degree from a recognised instution indicates the virtues of the degree holder. Moreover, the degree of accuracy with which this system recognises the merits of individuals is of moral consequence, as evidenced both by public discourse on the topic and by the aforementioned (probabilistic) effects of higher education on one’s livelihood.

It is this aspect of our educational system — how accurately it measures the merits it purports to represent — that generates much dismay and disdain in my wrinkly, poorly seasoned heart. The false negative rate has been examined by many before me, particularly through the lens of the admissions process: many virtuous (hard-working, talented, compassionate, etc.) individuals are held back by inequities in primary education and/or by socioeconomic factors. At least one of these individuals was a role model of work ethic and determination in my youth.

But having seen school from the other side of the classroom, I believe that there is an alarmingly high false positive rate as well.

Consider the Applied Mathematics BS at UCLA (as of 2021).² According to the registrar, an individual earning this degree has a strong content knowledge of calculus and differential equations, has analytical problem-solving and mathematical reasoning skills, knows foundational linear algebra and real analysis, and can perform basic C++ programming. In addition, there are university-wide course requirements for writing and general education. This is an explicit statement of the “merits” any degree-holder ought to have; of course, there are softer merits like intellect, eloquence, perseverance, integrity, etc.

Consider now the claim that an individual with this degree has a strong content knowledge of differential equations. One can earn the degree without taking Math 134 (linear and nonlinear systems of differential equations), Math 135 (ordinary differential equations), or Math 136 (partial differential equations), the only upper-division classes focused on teaching differential equations! Of the courses focused on differential equations,³ one needs only take Math 33B, endowing a knowledge of differential equations shared by computational biologists, physicists, and engineers.⁴

To pass Math 33B, one must complete a series of homework assignments and pass at most two midterms, one final, and sometimes a smattering of quizzes. A student with many scruples may attend every lecture and discussion, complete every homework thoroughly, and study hard for these examinations to earn a good score, leaving the class with a solid grasp of the material.

However, earning an “A” in this course does not always require this much work, especially in the absence of scruples. Homework is easily outsourced to ChatGPT or mathematically inclined friends. Exams are also poor proxies of content knowledge. Students are not rewarded for understanding how to derive the variation of parameters formula or why the Wroǹskian does what it does. Rather, rote memorisation and good test-taking technique are more directly assessed on these exams.

One can repeat the above analysis for the other academic requirements of the major.⁵ But beyond the incongruity of the stated academic merits this particular degree conveys and the actual curriculum, I hope this example illustrates a broader issue within the academic culture and incentive structure at our institutions of higher education, even beyond the scope of a single lower-division math class.

The constrained optimisation problem of “minimise time spent in a course subject to the constraint of passing” can be observed in the community directly. On June 8, 2025, someone asked on UCLA’s subreddit, among the general education (GE) classes offered last summer, which were the easiest to take? Someone responded to this question by stating, “Honestly unless you’re applying to med or law school just take the ge’s at a cc. It’s really not worth the money.” On June 28, 2025, another person asked for the “master list of all easy GEs.” I find it difficult to criticise this behaviour; most of the GEs that I suffered through as an undergraduate at UCLA were not worth the economic cost of tuition, not to mention the time sunk into them. Thus here is a clear reward in getting close to the optimum here — one spends less time whittling away at inconsequential coursework and can spend it elsewhere, and if one chooses to take the economically efficient route, there is a strict material gain as well. Degrees cannot distinguish between the optimisers and those following the spirit of a well-rounded education.

Going farther back to January 26, 2025, an anonymous TA bemoaned their frustration with generative AI being used to complete coursework; the poster and sympathisers warned of dire academic consequences and the difference in respect they held for one with more integrity. One popular response states, “Though I’m not a TA, I have more respect and trust for someone who fails a class after an honest unknowledgable attempt than someone who cheats.” In the same response, they contemptuously and exasperately say, “You see some people blatantly cheat here on exams and act like they are a genius. Then when you ask them a question on a topic, they don’t know shit.” Taking the defense of the generative AI users, another respondent states,

If you pay for education and you’re a humanities major, I get your point. But for most other majors, they’re not paying to take these GE classes, we are just forced to do it to get our degree. Most STEM majors I’ve talked to find it a waste of time to write an essay about something they don’t care about instead of using that energy to study for classes that are actually important for their future.

Another commentor shares a similar perspective:

to be clear, i don’t use ai to cheat on assignments, but most students these days are in college to get their degree…not condoning cheating, but i can definitely see where the motivations are; it’s really quite similar to the motivations in high school

Although one may believe that the students solving the aforementioned constrained optimisation problem are rather sparse, seekers of the paths of least resistance appear frequently, especially when navigating GE requirements. This last thread in particular highlights the prevalence of this min-effort max-degree attitude.⁶ Several of those contemptuous remarks even allude to the inferior merit of those that rely on generative AI, yet I must again highlight that a degree cannot distinguish which category the degree holder belongs to.

To reiterate, the qualities associated with being an exceptional student are not incentivised by our degree programs and class structures; these qualities likewise are not well reflected by simply holding a degree. Instead, one is rewarded for taking the easiest and least challenging classes and doing the bare minimum to get past course and degree requirements. I think the reward for taking these low-resistant paths has only grown with LLMs becoming more capable in popular; the risks for traversing these now well-treaded paths likewise diminishes steeply. Thus I believe the inaccuracy of the college degree as an indicator of certain virtues does appear in a high false positive rate: a significant number of degree holders have not in fact attained the academic mastery the degrees advertise, nor do they have the studiousness we would like to believe a degree implies.

This is a deep failure of our higher education system,⁷ and I think we need to completely restructure how we examine and assess our students. Mandated assignments in a course and mandated courses for a degree requirement in principle seem necessary, but their current implementation again has the wrong incentive structure.

Deep in my heart, I want to argue for assessments in the style of oral exams, presentations, and open-ended creative work like videos, especially since these avenues of assessment allow different means of the expression of conceptual understanding, of effort, of creativity, etc. An immediate objection is one of objectivity: how does one make consistent the meaning of the letter “A” on a transcript when the method of evaluation can be so disparate?

Thus perhaps we may identify a root of all I’ve described today. The language with which we evaluate students is as expressive as an opera singer that’s been run over by a truck. When a degree is communicated through just a handful of letters and a stray + or -, it’s no surprise then that it cannot distinguish between the next Plato and an ingrown toenail. And why not use “ardent” or “abhorrent” in place of “A”; “precocious” or “pissed me off” in place of “P”; “cracked” or “crustacean” in place of “C”? It pains me to say that so many valuable qualities in my past students have gone unrecognised, so many shortcuts and swindles gone unpunished by a system that speaks like its voice may run out, confined to a vocabrulary of maybe 20 pen strokes, conveying no more than a dozen bits of information in a grade.