One of the more disappointing aspects of this-- it's not even a good service to these businesses to operate this way. Some of the least effective colleagues I've had the pleasure of working with attended these brand name schools. Now, I certainly wouldn't say that represents all graduates, but some of them are absolutely being carried because they look good on paper. That's no way to build out your workforce.
I've never heard of NC A&T so I went to Wikipedia [0] and learned their average undergrad SAT on intake is ~1,000. Average SAT at Duke is something like 1,500.
One of those is top 5%, the other is sub-50%. It seems a stretch to say the biggest problem the NC A&T group faced was not being invited to the right parties.
SAT isn't the be-all and end all, but those difference are far too large to be brushed aside. It is not at all obvious that race is the major factor here.
There is a case here that the recruiters were being realistic; the smart black candidates would presumably be working to get into some university other than NC A&T. The CS program entrants would need to look like legitimate geniuses compared to the average to qualify for red carpet treatment.
Actually, that's how lots of graduate recruitment is moving to, looking at those contextual elements now (such as value add, and performing better compared to relative median).
Sometimes this means that someone graduating from a university with a decent degree when the average intake grades were comparatively low shows they are and are capable of outperforming to a greater extent. They don't have to evidence themselves as legitimate geniuses, just that they have the capacity to develop faster.
Obviously it then comes down to bias about whether the university is just scoring people higher, but that's a whole different ballgame.
Brand name universities still get the preferential treatment though, often because they have more developed alumni links with the hiring companies.
I'm sorry, but I didn't get into Duke as a 17 year old. Does that mean my career prospects are shot for good, and that "Hooli-esque" companies shouldn't recruit me?
I'd say the software engineering side of things are a bit different, the interview process and testing "is the gatekeeping". It might impact your ability to climb the ladder when you're competing for Sr Manager positions with a bunch of MIT/Standford/Harvard/etc type of graduates, depending on where you live....
...but by then, you'd have crushed your early career and gone off and gotten a prestigious Master's Degree, or founded your own company and sold it to one of the bigger fish, earning your way into Sr Leadership...
This is certainly the prior I bring to this story: that all this elite recruiting is horseshit anyways. But, of course, that horseshit also gatekeeps the most lucrative jobs in one of the most important white collar professions in our economy.
I just don't get which world the people who make these decisions live in. I've worked as a developer for almost 2 decades now & some of the best programmers I worked with were high school or college drop outs. And some of those that went to elite schools were mediocre.
I'm sure there is some positive correlation between school ranking and quality of education but it seems too weak & there are so many exceptions for this 1 data point about a candidate to be given such weight. I'd absolutely take someone who went to Colorado State University & has 3 years of relevant job experience than someone fresh out of Stanford (all other things being equal).
Also almost every time I was involved in hiring we had more competent candidates to choose from than we had jobs to fill (with exception of the highest tier jobs, like searching for a Tech Director or CTO which nobody is giving to people fresh out of school anyway) so why do you even need to butter up the Duke grads? Are recent grads with no job experience so desirable in the US?
According to this study, you're correct. The funny thing is that when it was discussed in HN at the time is that the thread had many comments from students who went to MIT, Stanford, etc. trying to deny it.
I think one thing is that people factor the college they went to in terms of the jobs they are seeking (not just in terms of pay, but also prestige of the company they are looking at, flexibility with moving, etc.), that may not matter as much at the high-end of the pay range or after years of experience, but I think right out of college this is important.
My argument is basically the low-end of the jobs that someone who went to Princeton and excelled in computer science is willing to settle for is higher than the low-end of jobs that someone who went to a lesser-name school and excelled in computer science is willing to settle for.
In theory, if the difference in quality of education is enough, then the level of education that someone who excelled in computer science at Princeton vs someone who excelled in computer science at a lesser-name college would equal out this difference in expectations, but I don't think that is the case, at least when we are talking about high-end school vs mid-range school.
Moreover, in computer science especially there is a lot of self-learning. Someone who has done consistently better than their peers probably has a good talent for self-learning, even if the average of their peers is lower.
I think you get the same discount for quality when hiring weird people or people with odd job histories.
My general idea here is that if someone is less desirable in a way which you do not care about but where other companies might care about, they are probably going to be on average of higher quality than someone more conventionally desirable asking for the same salary range. Not saying this is an iron law, but I think it is a tendency.
> I've worked as a developer for almost 2 decades now & some of the best programmers I worked with were high school or college drop outs. And some of those that went to elite schools were mediocre.
Within the software industry there should be a negative correlation between school quality and performance as an employee. See https://en.wikipedia.org/wiki/Berkson%27s_paradox . People who are unqualified and unfit to do good work have been removed from the pool. Anyone in the pool who is unqualified must be fit to do good work.
During my undergrad years, there were people who made it out without programming in any language besides Java, and barely did any tech hobbies or independent programming outside of class. There were others who were constantly learning other languages, playing with random tech, and building small opens source tools and webapps.
I've met people who went through community collect programs who didn't understand basic BigO notation or algorithm optimization (no fault to them; it's just not taught in a lot of two year courses) and I've know people with only high school degrees who've picked up advanced CS books and would learn about compile theory and different types of parsing on their own for fun.
Absolutely you get out what you put in. My company has thousands of programmers. I was in charge of helping a group of programmers self learn a new framework/language. It started off with 30. It ended up with me alone in a room. This was easily 3k worth of free classes. It was not even that hard you just had to watch some videos and maybe do some simple coding, maybe 1-2 hours a week, 9 sessions over 3 months. Most just did not do it, which means they would not show up to the sessions. By the last 2 sessions I just did not bother to try to get them to come. There were 3 I could kind of coach along and get them to sometimes engage. But mostly they just were not interested in helping themselves. They wanted me to sit in front of the 'class' and teach it. Now that I think about it this could be a good way to filter for people who are motivated to do work. But it would bias against people who have a full schedule.
Nah, your instructor's/TA's time is limited. I can't schedule 6 hours with them. I can't request as much time as I want. Colleges have other prerequisites which must be filled. The American systems don't let you go to college and do nothing but computers, and you must take time away from programming courses eventually. I can't maximize the time for programming, so, I'm limited to what I can put in. And if it's a tough college, it takes real effort to pass those other courses and you likely can't just coast. It's a forcing function with guidance.
If I put in a ton of time on my own to study stuff from the internet, that's the internet returning my investment of time and focus. It's got nothing to do with the college at that point.
I found that if you want to talk about the professor's research area, they will often have time to talk, at least at the beginning / end of classes or over emails.
I'm not as sure with TA's, partly because I often skipped the TA-sessions for various reasons.
I think on the engineering side it's different. the interview process is the gatekeeping and FAANG companies are totally open to interviewing anyone who thinks they can get through the process. Your ability to climb into Sr leadership might be impacted, if that's the direction you want to go, but you can absolutely prove yourself in engineering.
Also, I like your point about CSU. I do a lot of hiring in the Boston area, and we have so many "second-tier" schools who produce absolutely excellent SEs. It might be trickle-down because of MIT here. Don't get me wrong, definitely hire MIT, but we also cherry-pick from Northeastern, WPI (my favorite), etc. with equally as good results.
For someone who was just starting out and went to a no name college (no insult intended - I did), what signals would a FAANG company have that you are even worth their time interviewing?
I graduated in the mid 90s so I really don’t have a frame of reference. But, hypothetically speaking, if I had graduated in 2012 from the same no name college but spent a lot of time “grinding leetCode” and had the technical altitude to do well in the interview would they have even given me a chance?
I also went to a good but unknown school and was recruited by Amazon and Google. As far as I can tell it was purely due to applicable work experience (which I got working in companies you never heard of).
No one is arguing that once you have your first job, what college you went to matters. If you can get that first job, then only experience matters.
Heck on my very sparse LinkedIn profile, it basically showed me as an enterprise C# CRUD developer and recruiters were reaching out to me from Facebook and Google even though I wouldn’t get through the first fifteen minutes of a technical phone screen for either without at least six months to a year of prep. Recruiters throw out a wide net.
That being said, after the local market tanked post Covid, and my unknown company with less than 50 people all in did an across the board pay cut, I threw a Hail Mary and applied for a remote job as a consultant for AWS. I had the experience and I knew it was going to be a remote interview so I took a chance and got in.
But would they have recruited interns from my no name college - even if I did try to give them some type of signal like side projects and an AWS cert?
