The Most Regretted Majors All Had One Thing In Common

@Nhatrang excellent suggestion I wouldn’t have thought of.

@sciencenerd Great to know this can be done, for the college student who may want their resume to show skills that are not obvious from a simple identification of major and minor, who can’t fit in minors in multiple subjects.

Majors associated with especially high earnings tend to be overrepresented at highly selective colleges, particularly ones that are privates and not LACs. If you look at changes over time, it’s likely that CS will become the most popular major in the near future. For example, a comparison in primary concentration enrollment at Harvard between 2008 and 2017 (most recent available year) is below. I also included Stanford. Both Harvard and Stanford saw a 5x increase in CS enrollment during the reference period over the past ~decade. At Stanford CS has a similar enrollment to the next 4 most popular majors combined – a larger enrollment than any other major at any time in Stanford’s recorded history (2nd highest was history major enrollment during Vietnam).

Harvard Most Enrolled Concentrations
2017

  1. Econ – 660 (decrease)
  2. CS – 453 (5x increase)
  3. Government – 340 (decrease)
  4. Applied Mathematics 305 (3x increase)

2008

  1. Econ – 742
  2. Government – 485
  3. Social Studies – 296
  4. Psychology – 288

Stanford Most Enrolled Majors
2019

  1. CS – 745 (5x increase)
  2. Hum Bio – 240 (decrease)
  3. Econ – 190 (decrease)
  4. Symb Systems – 179 (3x increase)
  5. Engineering – 170 (small decrease)
  6. MS & Engineering – 159 ( increase)

2008

  1. Hum Bio – 390
  2. Econ – 225
  3. Bio – 218
  4. Engineering – 181
  5. IR – 169
  6. CS – 141

@Nhatrang 1,000%. You will find many LA grads in the highly paid field of management consulting. And firms are not paying 300-500 hourly for someone to program. The highest level skills are often soft skills and combining soft skills with technical understanding or any other knowledge skillset bumps you into those jobs.
@roycroftmom Not was I was talking about at all. Skills learned by someone in the fashion of online learning for accreditation are far stronger than taking a class in marketing 202.
The STEM bias is not going to work for anyone unless they have soft skills as well. There are many programmers, Xray techs and more just looking to replace US jobs. This is ongoing and likely will continue with the advance of robotics and AI in the next decade. So it’s not STEM vs. liberal arts, it’s really having solid skills for a changing job market and the ability to pivot and learn new things. Wise people realize that they need to be open minded about what skills are demanded in the market and might have to learn many sub skills over time. Many of the latest degrees weren’t even around 20 years ago.

No @happy times, you will not find many LA grads in consulting for the simple reason that the top tiers of consulting aren’t hiring huge numbers of people. The ones they hire are largely HYP, and thus largely liberal arts grads, so yes, the relatively few consultants there are do have that background, but from a class of say, 1200 at Princeton, McKinsey will take fewer than 10, Bain and BCG the same. Same at other schools. Not a big source of employment for the whole class, and you better show extraordinary quantitative talent in those case study interviews to even have a chance

The importance of major varies dramatically with field of work, as well as what specific job within that field. For example, in my post above, I mentioned that CS enrollment has had a dramatic increase at highly selective colleges, and is on pace to become the most popular major. One of the reasons for this trend is median starting salaries for CS grads at highly selective private colleges tend to be in the 6 figures and well above any other major.

Higher salary entry level CS and engineering jobs usually explicitly state something about major in the job positing. Some job postings explicitly state that they require applicants to have a computer science degree While others say phrase it more like “degree in Computer Science, similar technical field of study or equivalent practical experience.” Such positions also usually test CS skills within the interviews and expect the applicant to be knowledgeable about CS when they are hired, rather than be taught key CS related skills on the job. A non-major who only takes an intro to CS type class would be unlikely to do well on this type of technical interview, even if the job posting does not specifically say just CS majors.

However, there are also plenty other fields that expect the applicant to learn key skills during job training rather than in college, so a wider variety of backgrounds are hired. It’s difficult to generalize.

Many colleges have databases showing outcomes for grads by major including company, job title, and salary. Some also track long term outcomes many years after graduating. For example, Cornell A&S lists some basic info at https://as.cornell.edu/majors-minors-and-grad-fields .

For Computer Science majors, they report 99% employed or in grad school. Most working grads have the job title “software engineer.” And the most common employer was Google.

For English majors, they report 83% employed or in grad school. Job titles are widely varied. The most common job title is teacher. Editor and analyst also appear several times. Employers are also widely varied, encompassing several very different fields. The only employer that has more than 2 was Teach for America (5)…

@roycroftmom, about 10% of Harvard students go into consulting upon graduation. Yale reports 13%. The only field with higher rates at these schools is finance. Princeton reports similar numbers.

However, most college graduates, regardless of major, do not get hired into management consulting. So choosing a major on that basis may not be too helpful to those not at targeted colleges for those employers.

With respect to companies and training, one of the companies I started has three lines of business. One line of business offered a high-value training to companies. Over the last five years, that part of the business has declined. Companies have sought to replace in-person training with virtual modules. Clients have been shifting to virtual training across the board and there are arguments that it can be equally or more effective, though it doesn’t work well in our field. We’ve replaced the lost business with another that focuses on helping them to achieve better results. Training is a part of what we offer, but it is a smaller part of a much bigger package (with a correspondingly longer sales cycle) and the budget for this kind of program comes from the business and not the shrinking and reallocated learning and development budget.

What I also hear is that many companies want new hires to be able to hit the ground running. Bank training programs used to be staples for bringing liberal arts-y folk (non-quantitative) into the company. Now, these programs seem to have become largely extinct (Chase, Goldman, etc.). Companies want the new hires to be trained in some area that can be applied on Day 1.

At quite a number of our clients, both general training and new hire training budgets have declined. According to Peter Capelli, a labor economist at Wharton, in 1979, “the average young worker received 2.5 weeks of training per year. By 1995, training time fell to just 11 hours.” (See https://www.entrepreneur.com/article/314468.)

@Nhatrang wrote about people vilifying LACs. People can learn to think well at LACs and learn some technical skills (usually not engineering, but stats, a mild version of CS, etc.). No reason to vilify LACs, except to the extent that some are selling a data-free social justice warrior version of education.

My firm works cooperatively with the top management consulting firms and I think the conclusion that they are hiring humanities-oriented majors to be a bit of an overstatement. I have heard that new grads are screened first on their math SATs and that there is a cutoff of something like 750 (I don’t know this as a fact). Moreover, one of the partners at a top consulting firm described the new hires out of undergrad as “data slaves” for their first two years. Their job becomes collecting data, analyzing, massaging, and preparing presentations. They are also looking for candidates to also have effective people skills, as they have to get the data and the cooperation of employees at their clients, but a good humanities education without some quantitative talent probably wouldn’t cut it at this point.

Finally, someone talked about looking at the majors of current CEOs. I don’t think that’s relevant because when they were hired a) there wasn’t resume-screening software; and b) the businesses were much less data-driven. By analogy, I had dinner with a friend who is on several major corporate boards. He says he would never be invited to those boards now as the search committees are explicitly look for anything other than white guys. The screening that was done then is very different than that which is done now both for new employees and for board members.

@Sue22 The percentages are accurate but most aren’t going to McKinsey, Bain, or BCG. There are of course many other firms 1-2 rings down the prestige ladder plus boutiques. The odds are very slim for breaking into MBB from undergrad from any school. And while major doesn’t matter, quantitative and data analysis skills are absolutely necessary. But those skills can be acquired through adequate study in preparing for interviews. Of course, the student needs to get an interview first and so will need to stand out on paper.

Firms a rung or two down are still excellent jobs (assuming the consulting lifestyle is something one finds interesting and attractive :slight_smile: ).

Many liberal arts majors can also have good data analysis skills. It’s not rocket science and, if one is going to a selective college, chances are one is also decently intelligent.

@doschicos True as to both. Although, below MBB, prestige of undergrad plays a much smaller role in getting hired. Anyway, I am not in the anti-liberal arts camp at all. I think it’s short-sighted. Smart students can teach themselves enough accounting and econ to ace consulting interviews.

Getting a data free (or very light) education (not necessarily SJW) can be done at many, probably most, colleges, LAC or otherwise.

Perhaps the reason that some look down on humanities majors is because that is commonly possible and commonly done by those who want to avoid statistics or math as much as possible.

Why is that something to look down on?

My favourite article on how college kids should choose a major is the Anderson article I posted earlier. This is going to be my last comment on the topic, which is an article by another prof on the Anderson hypothesis:

https://www.outsidethebeltway.com/why-more-americans-dont-major-in-the-math-and-science/

My favourite article on how to pick the best candidate for the job is this one:

https://www.nationalreview.com/corner/how-elite-business-recruiting-really-works-jim-manzi/

I see the SAT cutoff and quant requirements as “hook busters”. I see the casting of a wide net egalitarian. The system is elitist, but fair. I don’t see anything better out there.

@roycroftmom is correct.

Very poor understanding of logic, math, and statistics can limit what one can do in areas where basic logic, math, or statistics knowledge is useful. That is analogous to having poor reading and writing skills.

Examples on these forums include thinking that college admission decisions are independent events and believing that admission rate is all that needs to be know to determine how difficult it is to get admitted. Another example is difficulty understanding the difference between necessary and sufficient (e.g. high GPA and test scores are typically necessary, but not sufficient, to get admitted to the most selective colleges).

Using SAT scores is judging a college senior or graduate based on his/her high school credentials. High school credentials (and college name) are more affected by advantage or disadvantage given by parental circumstances and choices than record in college (which is also more recent and therefore more relevant).

Basic logic can come without majoring in STEM. Many, many majors require some modicum of math/stats/STEM anyway. These things don’t needed to be tied to math majors. I was known for having a mathematical mind in my career in finance and never took anything beyond Calc. I’m not alone in that regard either. I think many posters in this thread overemphasis STEM degrees’ importance in having analytical and computational skills.

Two recent grads I know (yes, I know anecdotal) from strong LACs.
Philosophy major getting a PhD in Biometrics
English major working financial consulting (also high stakes poker player in spare time)

Yes, it can, but most colleges do not require that as part of general education, like they typically do for reading and writing courses.

One does not have to major in a STEM field or philosophy to learn basic logic. But many college students avoid learning it (after forgetting what they learned in high school geometry), so it is not surprising that employers that care assume lack of such unless shown otherwise.

Regarding CS jobs and major, that’s one field where I’ve seen capable students do extremely well from pretty much any college they attend (small religious colleges to large secular ones and everything in between). They have to be capable though - not just graduates. And really, they don’t necessarily even have to go to college, but most do at this point. They just know a lot before they leave high school.

I can understand why more are choosing it as a major. It has me wondering if the field will oversaturate at some point. Plus, there will still be those who graduate who never were cut out to be that major IME. If one only chooses it for the money and doesn’t have innate talent I’m not sure they’ll get (or keep) jobs, esp with more competition in the near future.

What do you define as “basic logic” anyway? the assumption that stem study requires “basic logic” and LA study doesn’t require logic is absurd.

Doesn’t everyone need “basic logic” to convey their ideas? It would sound utter nonsense without “basic logic”.

Don’t tell me that writing an essay about a historical event, analyzing it, breaking it down, applying your own interpretation, taking the reader through your own lens etc., doesn’t require “basic logic”!

A guy who used to work for me, who thought of himself as so logical. Yet everything came out of his mouth doesn’t make sense to people. He told the user to “type in ip config” when he did support work. He told me it’s so logical bc that’s how one troubleshoots a computer problem. I had to explained to him step by why that wouldn’t work with a regular user. After so many complaints I had to let him go. If I could put him in a basement and had him do what he was told with no interaction with another human being, that would have worked.

Saying LA students don’t require to learn “basic logic” is the same as saying Stem students don’t require to write.