Top academics and officials at computer science programs rated the overall quality of undergraduate programs with which they were familiar on a 1-5 scale. A school’s undergraduate computer science rank is solely determined by its average of scores received from these surveys. To be included in this standalone peer assessment survey and ranked, a program must either have been accredited by ABET or have recently awarded 20 or more bachelor’s degrees in computer science
That’s the most worthless methodology there is. There is ZERO objective information. It’s well known that people way down the chain are assigned to fill those out and that reputations are based on research output. Just like USNWR Engineering that has Columbia at 13, Penn at 18 and Harvard at 21…garbage in, garbage out.
I would like to see the questions, the sample size, and also to know the position of every single person who filled it out…so many of the “top academics and officials” (what does that even mean?) give these surveys to low level admissions people, institutional reporting staff, academic dept staff, etc., to fill out. Of course, I expect like the overall ranking surveys, most are thrown in the garbage.
Found the gory details. 537 Surveys went out, 28.5% completed. So these rankings are based on the subjective opinions of 153 people.
This year, U.S. News & World Report is publishing its third annual Undergraduate Computer Science Programs rankings. These rankings are based solely on the judgments of deans and senior faculty members of computer science departments at institutions around the country.
To have received the survey and be ranked, an institution’s computer science department must have met at least one of two below criteria:
House a computer science bachelor’s degree program accredited by ABET, the Accreditation Board for Engineering and Technology.
Be a regionally accredited institution that awarded at least 20 bachelor’s degrees in computer science, according to the Integrated Postsecondary Education Data System (IPEDS) Data Center.
In the spring and summer of 2022, U.S. News surveyed deans and senior faculty members at each of the 537 undergraduate computer science programs that met at least one of the criteria above. All programs surveyed were ranked.
These deans and faculty members were asked to rate the academic quality of peer institutions’ computer science programs on a 5-point scale: outstanding (5), strong (4), good (3), adequate (2) or marginal (1). Individuals who were unfamiliar with a particular school’s programs were asked to select “don’t know.”
Average peer assessment scores for each computer science program were determined by computing a trimmed mean – eliminating the two highest and two lowest scores given by respondents. Average scores were then sorted in descending order. Of those surveyed, 28.5% responded.
In the overall computer science ranking, programs with average peer assessment scores less than 2.0 are listed in alphabetical order in a ranking range based on the number of programs that also earned scores below 2.0. U.S. News will supply schools or programs that are listed in the ranking range with their numerical rankings if a school official submits a request to official@usnews.com.
U.S. News also asked those same respondents to nominate up to 15 of the best programs in the 10 computer science specialty areas listed below. Specialty rankings are based solely on nominations by survey respondents.
Specialty areas are:
Artificial intelligence
Biocomputing / bioinformatics / biotechnology
Computer systems
Cybersecurity
Data analytics / science
Game / simulation development
Mobile / web applications
Programming languages
Software engineering
Theory
The 10 computer science specialty areas are ranked in descending order based on the number of mentions they received. A program needed seven or more nominations in a specialty area to be ranked. This means that schools ranked at the bottom of each specialty ranking received at least seven nominations.
Schools offering any courses in a specialty are eligible to be ranked in that specialty. Schools did not need a listed program or major in a specialty area to qualify.
People will ‘vote’ for the same schools every year.
That’s sad it’s just a popularity contest.
Perhaps some high ranked schools aren’t delivering - don’t know. But surely some unranked schools who are delivering big time will never have a chance.
And all these kids chasing rank are spending many thousands of dollars based on nothing but a Dean or administrator ranking their buddy highly.
Wow, great work! Thanks for doing the lifting. Kind of like the Wizard of Oz…pay no attention to the man behind the curtain. How many people swallow hook, line and sinker and don’t even know it?
I do not claim to know CS so usually do not comment on chance/match threads about CS (or engineering)… but this ranking even has me, the great uninformed, gobsmacked.
I’ll just chime in to agree with this and say that my kid is a senior at Hope College majoring in CS, and Hope is AFAIK completely unranked in CS and has only 5 full time CS faculty. I’m sure a very advanced kid would not have picked it and would’ve maxed out the classes way too soon. But it’s worked for him. He had no difficulty getting internships including one sponsored by Hope at NASA, plus an NSF REU at CMU, and finally this past summer an internship in our hometown in AZ that has led to a full time job offer after graduation in information security. It’s not Google or Amazon but he is very happy with the outcome.
So it isn’t what you make its what you keep and where they are employable. Also you could just take a cs course, learn some programming and get a job. Many pay like $10,000 for intense programming online to get jobs. College is not just about learning CS programming etc.
There are many flavors of programs to suit different people. Many great schools out there regardless of the name of the school.
This is a good example and I don’t think we see enough of ones like this. There are a lot of good jobs with government contractors for starting salaries like this. They’re not winning contracts by paying new graduates $125k. $77k will go a long way in Huntsville.
On the other hand, the greatest test of how well undergraduates are taught in their field is in their success in more advanced degrees.
Caltech win at that, hands down, with something like 40% of their graduates going off and doing PhDs. MIT is a close contender, but is only #3 for CS. At #2 we have another college which is not on the list, Harvey Mudd. Then, #4 is Franklin Olin. In fact, only MIT and Stanford are in the top 10 producers of CS PhDs per graduating students in major, and Stanford squeeks in at #10.
Overall, liberal arts colleges are the big winners for quality of education, at least when wide areas of that education is needed, and the ability to synthesize that education. That is not only true for CS, but across the board.
As for quality of CS grads as employees in industry, that depends on many things, but the most important is likely whether the student had an internship or internships. A student from aprograns ranked #117 who has done two internships in industry is far more prepared for employment than a Harvard graduate who spent their summers vacationing.
The fact that access to internships is not in any ranking system points to just how useless these systems are at saying something about undergraduate programs.
PS. Of course, Harvard students who spend their summers vacationing will likely get a high paying job after graduation at one of the companies where their parents’ have connections, and then Harvard will point out that their newly graduated CS majors are making 7-figure salaries within 6 months of graduation…
If you are going to use a ranking system, it’s good to consider what criteria is used in that ranking rather than just choosing a college that is ranked high in criteria that you don’t care about.
The USNWR ranking is based entirely on only one criteria – a survey in which admins rank colleges’ computer science program on a scale of 1 to 5. Historically research powerhouses tend to do well on this type of survey and smaller colleges or colleges that are not as well known for academic research do not do as well. The additional criteria used in the Brown rankings are faculty publications, faculty best publication awards, and placement of grads/PhDs in to CS faculty positions, Obviously colleges that do not emphasize research would not do well in the Brown rankings. This fits with no LACs appearing in the rankings of top 200+ colleges.
Salary was mentioned frequently in the thread. Salary and employment are not considered in any of the listed rankings and should not be assumed to correlate well with the listed rankings.
Rather than focus on rankings, why not focus on the criteria that is important to you… and look at that specific criteria rather than a magazine/website’s numerical ranking? For example, suppose the Brown ranking’s rate of grads obtaining CS faculty positions is very important to you. If so, then look at which colleges do well/poorly on that metric, rather than just look at the overall ranking number based on the average of all Brown’s criteria. If you restrict it to just what portion of undergrads eventually become CS faculty, then many LACs will do much better than suggested by the Brown aggregate average.
As I said, if you’re looking purely at employment opportunities there are many pathways. But my point was - at top CS schools (and I’m not limiting “top” only to those schools that appear on these lists, but schools that have strong programs), students who really want to go above and beyond will have more advanced classes and research opportunities available to them. At other schools they will exhaust their options earlier.
Not every student who has a passion for computer science wants to pursue a MS/PhD.
But that’s a topic for another day.
This is true. I am not sure the MS adds much to a BS. You can take whatever courses you want to take in an MS while you are in a BS anyway without even adding an year. The PhD is a far different beast. And hard decisions are made due to the possible large opportunity cost of being out of the labor force for 5 years.
And that’s why I don’t love rankings. What’s important for one student is not for another.
I just had this discussion in a different group because an Ivy grad was struggling with finding a job post graduation. That schools sends 40% of their engineers to grad school. Their career services/industry connections are not as robust as many state flag ships where 80%+ go directly into industry.
It’s super hard to predict what a student is going to want when they are applying to undergrad at age 17/18, but if they are leaning one way or another, it makes some sense to do a deeper dive and make up individual rankings of what’s important to that particular student and disregard the popularity contests.
This is so important and students/families don’t give it enough weight in the search process IMO. It’s not only flagships with outstanding career centers, but schools like Dayton, DePaul, Elon, Bentley, and American (there are more I’m sure) are placement machines with great connections and more (good) opportunities than there are students.
