It depends on the school. From my quick survey, it looks like around 15 CS-related classes outside of math for a CS degree is pretty typical at either LAC or larger research universities.
What I strongly disagree with is the idea that larger research universities tend to produce better or more qualified CS graduates than those from LACs. I understand why some current students at larger research universities want to believe it - it’s good for their egos, but it’s one of those things that you quickly find isn’t true once you’ve entered the workforce and have the chance to compare CS grads from various schools. I’ve never found LAC CS grads to be inferior to CS grads from larger schools.
I found the opposite in the LACs I looked at (I have relatives going to college next year, so I have been looking at some). They were 12 CS-related classes at most, with most requiring 8-9. I think there is a discrepancy in the schools we are looking at.
100 "What I strongly disagree with is the idea that larger research universities tend to produce better or more qualified CS graduates than those from LACs."
I on the other hand firmly believe that larger research universities tend to produce better and more qualified CS graduates than those from LACs.
“Mind if I ask what your CS experience is? I’ve been going through your past comments and see no indication you have any programming background.”
Sure, I graduated with a Computer Engr degree which at the time was a combination of EE and CS with more of a focus on digital and programming (taking data structures e.g. rather than electromagnetism). I’ve been involved with high tech like you in mainly software and internet companies.
Agree and I didn’t bring it up, but when someone says that Stanford and Amherst have the same number of AI courses, that needed a response, you learn AI by more than courses was my point. And if Stanford has a gap in AI, they can get a visiting professor from Carnegie or MIT for a year or two, pretty easily.
“What I strongly disagree with is the idea that larger research universities tend to produce better or more qualified CS graduates than those from LACsA. I understand why some current students at larger research universities want to believe it - it’s good for their egos, but it’s one of those things that you quickly find isn’t true once you’ve entered the workforce and have the chance to compare CS grads from various schools. I’ve never found LAC CS grads to be inferior to CS grads from larger schools.”
Well I’ve not run into too many LAC grads so the research universities definitely produce more CS grads out here in silicon valley. I have run into more ivy league grads than LAC ones, outside of Harvey Mudd. And I don’t really ask where they did their education, unless in an interview situation. LOL on the ego comment, - this entire thread is about LAC grads protecting their egos.
@yikesyikesyikes : I’m not sure why you seem to be looking at LACs in terms of their minimum CS requirements (~11 computer science courses at the schools I’ve personally checked). In terms of program range, LACs may offer 15+ suitable courses for interested students. For computer hardware/general electronics courses, check the physics departments at colleges of potential interest.
That’s funny. None of your many previous comments give any indication of actual engineering or CS knowledge.
Who have been the LAC CS grads on this discussion board?
That doesn’t make sense. I’m here in the Bay Area. Mudd is extremely tiny, and I only know two graduates from there. Yet I know several dozen Ivy League graduates. I’m scratching my head trying to figure out how you know more Mudd graduates than Ivy League graduates. And if you work with people long enough, you almost always find out where they went to school, even if you don’t ask.
I believe that not all BA/BS degrees in CS are created equal, but that is OK. It depends upon what you want to do with that degree. Let me explain:
When my son and I were doing college visits a year or so ago, he attended a Tartan Overnight at Carnegie. We went to the college of computer science presentation on admissions and curriculum. During that presentation I learned that, unlike many other schools, CMU makes their freshmen start with an understanding of parallel/multi-threaded programming from the very start, with single thread programming as a special case of parallel programming. Now when I did my MS in CS thirty plus years ago, that was often only taught much later in a systems programming/operating systems type class for juniors or seniors. I suspect many other college CS programs have a course sequence more like the one I attended than the one at CMU, which gets me to my second point:
I think programs like CMU, MIT, Berkeley, and Stanford better prepare CS degree holders for jobs where the domain in which the programmer is writing his/her code is CS related, like programming languages, operating systems, autonomous vehicles, etc. On the other hand, at schools like LACs where the student is required to take more general ED type courses, those students are often better equipped to teach computers (through their programming) about other domains, such as Finance, Accounting, Geology, etc.
So I think it boils down to making sure the curriculum taught at a particular school is more closely aligned with your career aspirations than whether this program or that program is better in the abstract.
I have to LOL just a little bit, here. You do realize Wesleyan graduated something like 20 CS baccalaureates last year versus ~400 for Berkeley? In a sense, this whole thread has been about <1000 students spread across a dozen or so extremely selective LACs versus tens of thousands at comparable RUs. Which ones do you think are going to get more attention, have more opportunities to work closely with an actual CS professor and be guided in the right direction, if they want to do advanced work?
I am looking at minimum requirements, because that is what is required for students to graduate, and thus is the most appropriate way to benchmark the curricular depth and breadth of a degree program.
Even if we used your method, and compared by courses offered, you will see the vast majority of RUs will beat out the vast majority of LACs in terms of CS offerings in both depth and breadth.
Which are the best LACs for CS, and why. Maybe if this was know then comparisons to research universities would be easier. Is Bucknell really better than Amherst? Why?
Within it, LACs can be compared to each other as well as to universities such as WUStL. (By the standards of the analysis, Bucknell and Amherst would appear to offer equivalent programs, though particular course topics at Bucknell would be available more frequently.)
Can you give the Readers Digest version of that eye chart? Is it simply based on frequency of courses? Do those courses represent the basics or something more? How does the LAC rated #1 for CS differ from the one ranked #15?
The table appears to record the availability of foundational and elective upper-level courses across a range of CS topics that may be of importance and interest to students, as well as noting the frequency of these courses. Personally, I’d attach lesser weight to courses offered less often than bienially. Also, computer hardware courses may be offered through physics departments, and might not be registered in all cases.
It’s not a great chart. From my own experience, some colleges just have very confusing websites, making it well nigh impossible to tell what courses fit what category.
I know this is not satisfying, but, LACs do not, for the most part, excel by having nationally ranked departments. They excel by the product they produce which is well-educated individuals with passions for certain subjects. CS is certainly a popular and well-supported major at nearly all the top twenty or so.
Bucknell, unlike Swarthmore, offers a CS degree under both its B.A. and engineering programs. In that sense, it is more like some of the bigger research universities. As other posters have mentioned, if you have a passion for EEG, that may be your best bet.
I don’t think I have anything other than a few inchoate thoughts on this topic, and have appreciated reading the viewpoints of other posters. However, these would be a few of many possible forms of analysis:
Significant weight should be given to relevant CS course topics up to a count of about 15.
Dimiished, but still some, weight should be given to course availability beyond this level.
Thesis courses, in which the expectation will be that a student will produce originally presented work under the guidance of a professor, should be valued highly.
The desired educational experience outside of specific CS factors should perhaps be the most heavily weighted consideration.
Makes sense - I was explaining my train of thought (focusing on the depth and breadth of a CS education within LACs compared to their RU counterparts).
My main point is that LACs are great, but will not provide the same depth or breadth in CS coursework (except for schools like Harvey Mudd). The amount of CS knowledge an average CS undergrad will have coming out of a LAC will be less than the amount of CS knowledge an average CS undergrad will have coming out an RU at the same level.
Does this mean that you are doomed if you go to an LAC for CS? Certainly not. Top LACs seem to have their CS grads doing alright. We could argue until the cows come home on how they compare to their top RU student peers. However, I would have many reservations in majoring in CS at a less selective LAC. A big state public school is probably a better option by far for academic or professional opportunities.
In surveying CS programs, the Association for Computing Machinery observes both a trend toward intradisciplinary specialization and a trend toward interdisciplinary programs. Its response:
If we all could agree somehow that the correct number of required CS courses is precisely 14, but observe that some CS majors are graduating with only 11, it might be nice to know how the extra time is being used. Is it being spent on courses in physics? Mathematics? Basket weaving? If a college is admitting talented, well-prepared, highly motivated students - and also is conscientiously advising them - then it may be justified in allowing more freedom to explore interdisciplinary connections even at the expense of intradisciplinary specialization.