I know quite a few (12+) rationalists and CFAR graduates who are entering University soon or have just recently started University.

There was a lot of advice I wish I had been given or heard before I entered University and I think having a good repository of rationalist-contributed knowledge/advice/suggestions/information/links/DireWarnings could be very helpful to people in that situation.

1. What advice do students starting University need to hear?

2. What advice did your past self need to hear or what advice would have benefited you at that point in time?

3. Many people fail to ask the right questions. What questions do students need to ask themselves and other people?

Any links or guides on any related topics would be helpful. I will be posting some of my own ideas and links in the comments.
Possibly Relevant Repositories
New Comment
94 comments, sorted by Click to highlight new comments since:
Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings

Your parents have probably been helping you resist superstimuli (drugs, alcohol, video games) and fight procrastination. If you will be living away from home for the first time, you will need to be especially diligent against letting these forces harm you.

Political correctness is an unfortunately fact of most college campuses, but it's still a fact. You can hurt yourself with your teachers and fellow students if you say or write something politically incorrect even if your statement is true.

Figure out if you will need a high GPA to accomplish your post-college plans since the type of courses you take can greatly influence your GPA.

Not showing up for class because you overslept (especially your first year) is a sign of failure.

If you like LessWrong you will probably enjoy microeconomics.

You can often negotiate your financial aid offer, especially if you have a better offer from a college of equal or greater prestige.

You can take classes over the summer at a local community college to graduate early and hence save tuition.

Be very clear in your writing. Your professor will likely count any confusion he experiences in reading your paper against you. ... (read more)

Conversely, don't assume something politically correct is true just because no one is challenging it.
I've only ever worked as a tutor but I'll add a couple of "should be obvious but apparently isn't" bits of advice. Actually go to classes. If you're bright there's a good chance that the first few classes will cover very basic stuff which you may know already. Do not screw yourself over by deciding that it's so simple that you don't need to go to class. It's a surprisingly common failure mode for smart individuals. If there is any form of continuous assessment or marks that can be gained before the test do the work. 2% or whatever other small fraction each assignment is worth may sound small for something that takes 4 of 5 hours but lots of people screw themselves by not doing the work and thus making it very hard for themselves to pass the module. Actually go to classes. All of them. Yes, even if you're hung over or feeling really tired. Even if they're at 9am on a monday. I cannot stress how common a failure mode this is. Turn up to classes on time. Not 20 minutes after they've started. Professors can get very sick of people clattering into the hall and interrupting their lecture due to not being organized enough to turn up on time. Do try to make sure your questions are good ones. Professors can get sick of individuals who ask a lot of spectacularly stupid questions who then don't absorb the answers. Take advantage of the clubs and societies your campus has. College is about more than your chosen subject and you can get some great experiences.

In many residential universities, there's a tremendous amount of social inertia--the people you're friends with a year or two in will often depend heavily on who you met your first few weeks, and who you met in the first weeks of classes. So turn this to your advantage: introduce yourself to people you don't know early on, and try to deliberately figure out who's a good fit and who's a bad fit instead of just trusting to chance. Making homework study groups for all of your classes is 1) a good way to meet people 2) a good way to ensure that you're on top of your classes and 3) a good way to ensure your work is spread out across the week / semester, instead of bunched near deadlines.

I like this. Much of the value in college is from the networks you form. Start thinking about that early when the situation is most fluid.

This advice may go against other advice, but it's a tactic that has served me well: in making early-career decisions, such as your choice of major, always ask yourself which choice preserves future options.

For example, let's say you are considering a major, and you are equally interested in Architecture, Literature, and Engineering as careers.

Under my analysis, I would ask, which of these choices preserves the most options?

If you choose to pursue a degree in Literature, it is unlikely that you will be able to parley those skills into any kind of job in Architecture or in Engineering.

If you choose Architecture, you will find it very difficult (though not entirely impossible) to switch into Engineering for a graduate degree. However, you may find that you can try to pivot into some kind of Literary existence more easily.

If you choose Engineering, you'll find that Architectural schools will be eager to accept you for a graduate program, and the difficulty of switching from Engineering to a Literature program will probably be equal to the difficulty of switching from Architecture.

So, under this analysis, Engineering is the choice that preserves the most future options. At the point of... (read more)

Paul Graham has similar advice:
As a complement to this advice (which I think is good), it's important to make sure you still explore. Don't be so worried about making sure you do the thing that maximizes optionality that you're afraid to fail and don't try things. So if you think you should study math rather than econ (as per Kaj's comment), then start with math as your default, but make sure to also take an econ class to see if you're so much more interested in it / better at it that it's worth it to specialize.

If you are a technical person, concentrate on metaskills (learning to think algorithmically and learning to do and read proofs is much more important than learning python and linear algebra).

I think more theoretical majors are better for metaskills (even if you are practically minded). But I know some will disagree.

Learn both data analysis and programming -- the world will be your oyster.

I agree in principle, but am not sure of how to focus on the metaskills in practice. In my experience, I mostly learned these kinds of metaskills by trying to get better at concrete object-level skills such as "python and linear algebra", and wouldn't even know how to practice just the metaskill. (Legend is that the University of Helsinki's Computer Science department decided that the one mathy thing that CS majors really needed to learn was understanding proofs. So they asked if the Math department might possibly provide a course custom-tailored for just this purpose. The Math department replied that this already existed, and it was called "doing a minor in Math".)
I agree. Mathematicians haven't learned how to teach proofs abstractly, they teach by subject. My point was, the subject isn't super relevant. Although if you have specific aims you might want to tailor the subject to that. For example if you want to deal with data, linear algebra and probability theory are probably useful to know.
That really depends on what you want to do later on. If you are going into math/CS/etc. grad school, focusing on proofs is a great idea. If you'll get a bachelor's and go get a job, I'd much rather be skilled in Python and linear algebra than in doing proofs.
I think you are wrong about this.
I have no idea whether to downvote. On the one hand, you don't explain why. On the other hand, neither does Lumifer. But: I have never in my career needed proofs, and never expect to; I've needed linear algebra on multiple occasions, and expect to many times in the future. I work as a programmer, and have been programming for around twenty years now. I have developed something that twenty years ago might have been called AI. I have managed dozens of projects, and worked on dozens more. So, where am I going? Proofs are an absolutely worthless skill to have in terms of "getting a bachelor's and going and getting a job". There -might- be a job where you need to write proofs, but if there is, I haven't seen it.
What did you develop 20 years ago? ---------------------------------------- You should know I ignore karma, btw. ---------------------------------------- Think about the form of the statement you are making: "I don't know X, and it doesn't seem like I need X." Well, how do you know you don't? You have to compare current world to a counterfactual world where you did know X. How do you know you wouldn't be vastly better off? See also: "I don't need this fancy book learnin' I am doing fine in life."
That's not the statement I am making. I do know X; proofs were required coursework for every CS major at the educational facility I attended. I've never needed it.
Which specific work-related (meta) skill do you think doing proofs develops? It's not going to raise anyone's IQ, I don't see why it would be particularly effective at improving, say, the ability to focus or critical thinking or something like that.
I don't care about IQ, I think it's a fairly uninformative number. Doing proofs eventually gives you a nebulous thing called "mathematical sophistication" (what I sometimes call "metal struts in your brain") that I think helps enormously for adapting to and solving novel technical problems. When Heinlein said "specialization is for insects" I think he was making a similar point about metaskills.
I don't mean IQ as a number, I mean the underlying g. And people who graduate college and start working neither do, nor are expected to "solve novel technical problems". The closest to that are programmers who do have to solve problems daily, but for them courses in e.g. data structures or just experience with radically different languages will develop much more useful intuitions than "mathematical sophistication". If you are going to become a mathematician or a logician, by all means go study proofs. Otherwise I don't think they justify the opportunity costs. Aren't you suggesting specializing in a particular metaskill?
I think g is sort of a mathematical artifact, not a real thing (but don't really feel like getting into a big thing about this). Factor analysis doesn't tell people what they think it does.
The first principal component of scores on various tests is, of course, a mathematical artifact. But it's not the real thing, it's just an estimate, a finger pointing at the real thing. I agree that people can be both stupid and smart in very different ways, but at a certain -- and useful! -- level of aggregation, there are generally smart people and generally stupid people. There is a lot of variation around that axis, but I think the axis exists. I'm not arguing that everything should be projected into that one-dimensional space and reduced to a scalar.
Here is how this game works. We have a bunch of observed variables X, and a smaller set of hidden variables Z. We assume a particular model for the joint distribution p(X,Z). We then think about various facts about this distribution (for example eigenvalues of the covariance matrix). We then try to conclude a causal factor from these facts. This is where the error is. You can't conclude causality that way.
Do you think IQ has to be a causal factor to be a good predictor/be meaningful?
No I do not. I think IQ can be a useful predictor for some things (as good as one number can be, really). But that isn't the story with g, is it? It is claimed to be a causal factor. If we want to do prediction, let's just get a ton of features and use that, like they do in machine learning. Why fixate on one number? ---------------------------------------- Also -- we know IQ is not a causal factor, IQ is a result of a test (so it's a consequence, not a cause).
I know how the game works, I've paged through the Pearl book. But here, in this case, I don't care much about causality. I can observe the existence of stupid people and smart people (and somewhat-stupid, and middle-of-the-road, and a bit smart, etc.). I can roughly rank them on the smart - stupid axis. That axis won't capture all the diversity and the variation, but it will capture some. Whether what it captures is sufficient depends, of course. It depends on the purpose of the exercise and in some cases that's all you need and in some cases it's entirely inadequate. However in my experience that axis is pretty relevant to a lot of things. It's useful. Note that here no prediction is involved. I'm not talking about whether estimates of g (IQ, basically) can/will predict your success in life or any similar stuff. That's a different discussion.
??? ---------------------------------------- To the extent that you view g as what it is, I have no problem. But people think g is (a) a real thing and (b) causal. It's not at all clear it is either. "Real things" involved in human intelligence are super complicated and have to do with brain architecture (stuff we really don't understand well). We are miles and miles and miles away from "real things" in this setting. ---------------------------------------- The game I was describing was how PCA works, not stuff in Pearl's book. The point was PCA is just relying on a model of a joint distribution, and you have to be super careful with assumptions to extract causality from that.
I think of g as, basically, a projection from the high-dimensional space of, let's say, mind capabilities into low dimensions, in this case just a single one. Of course it's an "artifact", and of course you lose information when you do that. However what I mean by g pointing a finger at the real thing is that this high-dimensional cloud has some structure. Things are correlated (or, more generally, dependent on each other). One way -- a rough, simple way -- to get an estimate of one feature of this structure is to do IQ testing. Because it's so simple and because it's robust and because it can be shown to be correlated to a variety of real-life useful things, IQ scores became popular. They are not the Ultimate Explanation for Everything, but they are better than nothing. With respect to causality, I would say that the high-dimensional cloud of mind capabilities is the "cause". But it's hard to get a handle on it, for obvious reasons, and our one-scalar simplification of the whole thing might or might not be relevant to the causal relationship we are interested in. P.S. PCA actually has deeper problems because it's entirely linear and while that makes it easily tractable, real life, especially its biological bits, is rarely that convenient.
I don't think we disagree (?anymore?). ---------------------------------------- Also in practical informal talk, people overemphasize IQ because it is so fun for hierarchy-minded primates to arrange people from best to worst. edit re: PCA: Yes, PCA is a super-parametric method, with the usual super-parametric method problems. However, the issue I have with PCA in this context is different, and also occurs in very flexible, fully non-parametric methods. Basically the issue is, no matter how you massage it, the joint distribution simply does not have the causal information you want in it, in general.
Yep. You just have to pick the correct metric: the one where you come out on top ;-)

Science grad student perspective:

If you like books, it's often a good idea to read a textbook. This skill can be extremely powerful. If you're thinking about taking a class (that does not otherwise contribute to your graduation) because it seems interesting, try just reading a highly-rated textbook on the topic instead, and not spending all that time on the class. Like, classes aren't the end-all be-all of learning stuff. If you're taking a class that will require studying, reading a textbook on the subject ahead of the class will make everything much easi... (read more)

I highly commend reading textbooks in your chosen field as well, before taking classes (starting in high school if possible). The less time you have to spend reviewing and pondering basic issues during the class, the more time you have to hear, understand, and follow up on interesting side-notes in class.

You will get a lot of advice. Much of it you will agree with. What will probably surprise you is the scale of the advice.

Paul Graham talks at some point about knowing that startup success follows a power law, knowing that this meant most of the value of the Y Combinator fund would come from their single best company, and then still was surprised when, what do you know, most of the value in their fund came from their single best company. It wasn't real until it had actually happened to him. As much as possible, treat the advice that you agree with as if it had already happened to you.

80000 Hours
The 80000 Hours Career Guide
An impressive career guide that helps people maximize future impact and future earnings. It gives lots of strong advice on a variety of career choosing topics as well as looking in-depth into a few specific ones. (This website was created for Effective Altruists, but can be used by others very easily.)

Adulthood Fallacy?
This is purely me talking. Do not trust someone to be wise, emotionally mature, responsible, or trustworthy just because they are old. This applies to everyone you meet in the future and everyone you al... (read more)

For pure lectures (no real discussion, just the lecturer speaking for a long time) in courses where there are good written materials covering the same content and presence at the lectures is not required, don't bother going to the lectures if they don't feel obviously valuable. Self-study from the written materials is likely to be a more effective use of your time; lectures are generally a terrible format.

I used to think this, but I now suspect that this is mostly just for "our kind." The failure of online education to replace traditional lectures might be because typical college students gets something significant out of live performances. It's also probably related to why video conferencing hasn't replaced live conferences.
It's my impression (moderate confidence; have only skimmed a bit of the research here but that's the impression I came away with, see e.g. the link in my earlier comment) that "lectures are terrible" is the general consensus of education research.
Isn't "education research" the standard example of cargo cult science?
I think a big reason people are uninterested in video conferencing is the fact that eye contact does not work correctly (if you look at the other person's face, the other person does not see you looking at them.)
I might agree with your rationale, but not with following the conclusion. It can be very important to attend lectures in order to hear about most likely topics that will be on tests, any possible changes in test or project details and deadlines, to keep good rapport with the professor (so you aren't one of those "students who never show up to class"), and to keep yourself focused on following the material at the rate from which you will be tested on it. This may also vary by University and professor since some may care more about attendence and some are better than others at emailing information about test and project changes to students rather than just announcing it in class.
Probably. There have been a couple of courses where I'd probably done better on tests if I'd attended the lectures, but these have been rare exceptions. For the most part, skipping the lectures has only been beneficial, and I'm far from the only student who has found this to be the case. I also never got the impression that most professors particularly cared about lecture attendance, if they were pure lectures. Classes that involved actual discussion are different, of course. (There's probably a correlation with the fact that in order to have useful discussion in a class, the class size can't be too large, so pure lectures tended to be mass lectures where the professors were unlikely to notice your presence or absence anyway.)
I don't care about attendance, University isn't high school, it's not a prison.

It has happened more than once that a professor has assigned a textbook, which I bought, only for the professor to say in the first class that the only reason they assigned a textbook is because they were required to, but will never use it. Holding off on buying textbooks until after the first class (or, I guess, emailing the professor to ask if they plan on using the textbook) would have saved me several hundreds of dollars. (Having textbooks to study from is nice—they are, to me, the most efficient way of getting up to speed in math or science—but the ones professors assign because they need to put something down tend not to be the best ones.)

Library Genesis provides a way to access pdf of textbooks. In many cases PDFs are superior than textbooks. They aren't heavy objects that you have to carry around. You can search in them.
Location-specific advice Libgen is blocked by court order in the United Kingdom, but if you're a student, you can usually access it through Eduroam.

As with a lot of things, there is probably not a lot of good generically useful advice here. For example, the advice "do the absolute minimum amount of school work and spend all your time trying to build a startup out of your dorm" is probably terrible advice for 98% of people and great advice for 2%.

I recommend: 1) find a role model who is older than you, but similar in other respects, and whose lifestyle/career/situation seems appealing and 2) imitate his/her college trajectory.

I think this is good advice. Although I think a lot of successful people similar to (generic) "you" were also very lucky.
That number seems to be off by orders of magnitude. The problem is that the world changes. At one point nuclear engineering was an excellent high-potential major.
Spending effort on coding a program for a startup can be useful even if you don't have a successful startup.

Do go to your professor's office hours if you go to a school where professors care about spending time working with undergrads. Smaller schools seem to be good about this.

This helps make the professors like you if you are pleasant and ask informed questions. In theory, this can also help with networking later on.

Two important questions to ask yourself about the job a major will get you:

  1. What is the unemployment rate on these jobs?
  2. Will this job be automated in 5-10 years?

-Your health is very important for your success
-Getting enough sleep, having a good diet, maintaining energy levels, and being healthy will contribute to your long-term happiness and success.

-Melatonin is a supplement that many rationalists take to get better sleep at night
-Sleep Cycle is a good app that monitors your REM sleep cycles and wakes you up in a 30 minute time peri... (read more)

This actually goes both ways: I always heard the Freshman 15 as "you either gain or lose fifteen pounds your first year." (I know I missed a number of dinners due to inattention.) Especially if you're used to eating home-cooked food most of the time, eating cafeteria/restaurant food most of the time will most likely lead to weight gain. You may want to prioritize getting access to a kitchen and preparing food for yourself.

The most important decision you will make is your major. It is easily in the top 5 of all decisions you will make in your lifetime. Yes, you can move on to a different job later on if you choose the wrong major, but your major will limit your options when you do decide to switch careers if you picked the wrong one. A psychology major simply won't make nearly as much if they later decide to become a programmer as somebody who majored in computer science from the beginning. A wrong decision can cost you hundreds of thousands of dollars over a lifetime.

Pe... (read more)

My advice is probably better suited for a liberal arts major (compared to a STEM major, say).

Learn more than you know now about the jobs that your field of study might support -- especially salary and life style. This seems like a big blind spot to a lot of students.

Go to professors' office hours. They are fascinating people and know way more than you do. (P.S. I'm not a professor.)

Audit classes that you wish you had time to take.

Actually do the homework before the class in which it is due. (This is less of a problem for STEM majors than in humanities and ... (read more)

This sounds weird. Obviously you don't want to be constantly addicted to video games, but everyone needs to relax, too.
Let's go with a minimum of: * Immediately delete or throw away all MMOs and games that require regular time investments * Stick to games that can be picked up or put down easily so that they don't cause harm to your study schedule, sleep schedule, or social commitments
Would you also recommend giving up all other hobbies that required regular time investments?
Depends on the amount and specific interval of time investment. MMOs create demands like playing 3-4 hours every tuesday and thursday evening without fail for raids. This now requires a 6-8 hour time investment a week minimum with strong social pressures from online friends. That's not going to be helpful if you have a test on wednesday or friday. If you had a normal RPG or other hobby you could pick it up and put it down without social pressures and regular time investments at intervals in the middle of the week.
Given that I know plenty of people who play MMOs and participate in raids while still managing to pursue their studies/work/parenting/etc. successfully, this description of the absoluteness of the MMO time demands seems a little overblown. Granted, I don't actually play them myself, but it seems hard to believe that you couldn't have a good time playing an MMO while also finding an in-game social group that was reasonable about participation requirements.
From our different observations of anecdotal evidence on this and the other comment thread I think that the university environments and populations you and I were exposed to were very different from one another. My environment was not with exceedingly intelligent people (likely below LW average) and was at a decent but not great university. My observations were from when I was a Freshman in college and observing other people that age though. I've seen and heard of people who were older (grad school or working) and had much better experiences managing their time. However, I also remember meeting plenty of people when I was in high school who were much older and seemed to be playing far too much. Statistics on the topic would likely be useful at this point and people who are better at managing their time and dealing with At this point for giving advice to Freshman aged students, I'd rather put out a warning and see how they handle it than not say anything at all. If someone isn't already adept at managing their time at that age then I honestly think that playing an MMO on a regular basis could be detrimental and far more addictive a hobby than they should be testing themselves with at such a crucial time in their life.
That's reasonable. And admittedly, I've personally avoided MMOs precisely for the time management reasons, so one might say that my words were in conflict with my actual actions... but then I've also gotten the impression that other, less addiction-prone people than me have gotten a lot of genuinely valuable things (e.g. friendships, management and organization skills, etc.), so I'm inclined to object if people present what seems to be an unfairly negatively slanted view of the genre.
MMOs are optimized to be superstimuli in ways most hobbies aren't.

You are paying for the classes, i.e. the attention and time of your teachers. Make sure to get your money's worth: if you don't understand something speak up, or contact the teacher after the class. If your class has teaching assistants contact them (for example by email) if you get stuck on the homework/exercises or don't understand something from the lecture. All of these people are literally being paid to answer these questions, be aware that this is a resource you have at your disposal at all times. A common failure mode is thinking: "It's embarra... (read more)

On the same theme, note that if you don't understand something then almost certainly there are other people there who also don't understand and will be glad you asked.
When it comes to asking question, take note that asking questions is a skill. Asking good questions is a skill worth developing.

Ask for help when you need it. If you're struggling with a class, ask the professor or your advisor where you can find help. If you're struggling with life, find a counsellor. If you're struggling with a paper, find a writing tutor.

Take introductory Calculus, Chemistry and Physics in your first year*. At least at my school it was somewhat difficult to complete a science major in three years, so best to start off as though you are going to do one (unless you really don't want to).

Find a way to contact and talk to people who are where you want to be in the f... (read more)

It seems to me that one reason why some people behave irrationally is that they start implicitly thinking about themselves in terms of a particular identity, particular archetype. If people of that archetype tend to be bad at a X and one is also bad at X, one might not feel the irresistable urge to fix it, even though intellectually one might agree that it would be better if they fixed it.

In a university setting, at least at the beginning, two such archetypes are "hard working (but not necessarily talented) student" and "talented, but lazy s... (read more)

Very impressive article by Sidrea on the real reasons and value behind university

There are very few majors / areas of study where a single focus isn't significantly improved with a minor - and frequently, if it's not your major, Comp Sci is a great additional skillset. This is especially true if you need to take the credits anyways, and can choose between random course, or completing a minor with just a bit more work.

You want to do science? Almost no area doesn't need programming as well - it will help you get into grad school. You want to work in business? You'll spend half your day working on spreadsheets, and a CS background is inva... (read more)

On the science of how to learn: Make It Stick.

On the margin I think its usually better to take less and/or easier courses. It is better to do very well on an easier schedule than do "ok" on a hard schedule. If you apply for a job or grad school everyone will look at your gpa. If a semester is too easy this is not a serious problem. You can always read an extra textbook or do some coding projects in your spare time. Next semester you can up the difficulty. If a semester turns out to be too hard and you do badly the penalties are real (though survivable).

I was uncertain, coming from high school to university, whether or not I would adjust well. So I did the cautious thing and hedged: I only took 12 credit hours (the minimum load) the first semester. About three weeks in I was talking to one of my professors, complaining that I was bored out of my mind. They offered me a research position in their lab, and getting involved in student research my first semester was much more helpful than a class would have been. Similarly, if you find yourself with spare time and energy, find a productive outlet for it. Get used to building in pad time for yourself and then using that pad time flexibly. (This is a key factor in avoiding burnout during your adult life, I think.)
This is one of those "depends" pieces of advice. At some schools, the entire Freshman year GPA is completely forgiven, ignored, or given specific exemptions in how it's factored into the final GPA. In this case, why not go for broke? I always suggest starting out with the hardest thing and then ratcheting down instead of vice versa. In reality, I think people usually underestimate themselves. It's better to try the hardest possible path and then find your limits, than be cautious and never discover them.
That, I think, is false (though universities encourage the undergrads to believe this). Employers don't care about GPA at all.
I found that the first employer cared, and it was tricky to get a first job with a bad GPA. After the first job, they don't even ask about it. Unless you're trying to work for McKinsey or something, yeah, I wouldn't worry to much about it.
As a professor looking for promising graduate students, I don't care about GPA very much. Well, I suppose a super low GPA is a signal (but even then does not rule you out). ---------------------------------------- Admission committee might care.
This might be cultural/regional; in Finland it's common for employers to not care about grades, just whether or not you have a degree. (Though grad school will definitely care about the grades.)
In many programs you may also be able to take fewer classes during the full semesters and take fluff classes over the summer semester. If your program requires, for example, music appreciation or a physical ed class, these are good summer class; they are easy, you don't loose much by having the class shortened/modified, and they aren't distracting you when you need your attention for more important classes.