Hacker Newsnew | past | comments | ask | show | jobs | submit | more throwawayjava's commentslogin

> I realize my experience is anecdotal

FWIW I'm not totally convinced that the paper isn't also anecdotal. They selected for people who would agree to have their internet activity spied on in exchange for extra credit, and then from that group selected only for people using the internet -- not their computers, but the internet! -- in more than half the lectures. And did all of this in a large enrollment undergraduate gen ed intro course with lectures that lasted two hours.


These processes are called a self-selection bias and a sampling bias. Please don't use the word "anecdotal" in such contexts.


The use of "anecdotal" here is a rhetorical device, mirroring parent's language to emphasize that strongly biased data isn't necessarily any more valuable than mere anecdote. This is a conversation, not a peer review.


Please don't tell someone what words they are allowed to use in what contexts.


1. "Please don't" is hardly offensive or authoritative. 2. If you disagree, you were laughably hypocritical in your post.


Please don't define what phrases are "offensive" or "authoritative" in a conversation, especially after those phrases have been used. Context will have already defined them for you.


is there really any reason to expect the students who didn't take free points in exchange for privacy to outperform the observed group?

Sure, the results pertain more to excluding the internet from the classroom, but really that's just splitting hairs imo


That could depend on a two part hypothesis that I just came up with off the top of my head:

1) people who are more informed care more about privacy

2) you become more informed by being the type of person who seeks out and successfully obtains information and not being distracted

If 2) and 1) hold, your methodology will be biased to those liable to undertake distracting activities or be distracted. Furthermore, perhaps some would not even partake in the study because it's a distraction relative to why they're in the class in the first place.

Not saying that is demonstrable or true. Don't downvote the hypothetical messenger and all :p


Is there any reason we should be guessing at this when the experimenters could've chosen some other incentive (or none at all)?

And yes, I'd expect weaker students to be the more likely to spend their time / privacy on extra points.


The paper is entirely anecdotal. It's a small, self-selected group who also selected when they would be tracked. I do believe that the group's worse performance was related to their laptop usage, though. So a third variable, perhaps a quality of the students, caused them to do the observed behavior (have a laptop, sign up, log into the process and then) and do worse in their courses.

Finally, doing slightly worse on two hour gen eds is also not the same thing as doing worse in general. Those classes are ones students are forced to take, often with professors who are forced to teach them. I know my gen eds were lower quality classes, with more people, less participation, worse teaching, and worse assignments.


> ...Students who were enrolled in an introductory psychology course...

Ugh.

Also, from the paper:

> Five hundred seven students enrolled in an introductory psychology class in fall 2014... Each lecture lasted 1 hr and 50 min with a 10-min break in the middle.

Alternative headline: "500 person intro courses with meat-space lecture periods lasting TWO HOURS are stupid, and everything else is water under the bridge"

edit2: Also from the paper:

> Participants who logged in for less than half of the sessions were excluded from analysis

Why?! You don't think that frequency of use could correlate at all with type of use?! It's like they're willfully introducing unnecessary threats to validity...

And it keeps going like that for the entire paper. I smell experiment design hacking.


>And it keeps going like that for the entire paper. I smell experiment design hacking.

This sort of thing is exactly what pops up in my mind as well. I think they just looked at complete set of data, which might have shown that there is low-to-moderate correlation (assuming most people only use laptops when there is a part of course that is completely irrelevant/boring/already known); once they dropped the most significant use-case, all was left is the butt-in-seat users.

Also, I don't know why they never produce figures of a scatter plot for the data they are analyzing. A simple grade vs laptop time (academic, non-academic, total) figure would explain the results so much better than the raw tables. Maybe I'm just used to other kind of articles though.

P.S. There is also a really funny correlation in table 3, showing negative grade influence from playing games, where table 1 has 0 minutes spent for that across the board.


I believe that we may know a lot about how students at top western universities that take introductory psychology courses think and behave. As for the rest of us, who knows.


This is one of the classes I CLEPed while in highschool. I don't think I spent more than a couple of hours actually studying for it.


> This was tried. It was called "New Math".

No it wasn't, at least writ large in the USA.

And that's not what New Math was. Explaining what multiplication is doesn't demand an introduction to set theory.

If we read Feynman's CRITICISM of New Math, we actually find that he ADVOCATES for exactly what your parent is suggesting ("cobbled together.. in order to solve real-life problems"). So clearly, your parent isn't describing "New Math". Or perhaps Feynman is just a raving lunatic.

So I'm no advocate for "New Math", but I do oppose the argument you're making here, in which "New Math" is taken to mean "anything other than memorizing times tables" and is then denigrated on face. Without regard to the fact that the most vocal opponents of "New Math" were in fact advocating for exactly what your parent post is suggesting.

> Spectacular failure

So brief was the new math intervention that, to this day and despite all of the hoopla, we don't have a good empirical basis for claiming new math worked or did not work.

New Math was barely attempted, and its primary opponents were mathematically illiterate parents and teachers. This is just true, even if there were mathematically literate opponents to New Math, e.g., Kline or Feynman.

(But also note Meder’s reading of Kline. It's also worth noting that Mathematicians are maybe not the ultimate authority when discussing secondary pedagogy, especially in the mid 20th century. I have no basis for this belief, but IMO lots of mathematicians who weighed in on New Math were very possibly waging a sort of proxy battle as part of a larger war over the future of their own field -- pure vs applied.)

> Do you want to know what worked? Memorizing times tables.

Is this satire (honest question)? For all the things we don't know about math ed, we know that this doesn't work. Students who memorize times tables are routinely incapable of multiplying 12 by 13 or 55 by 55.

> There's just no way around the fact that drilling is key to early mathematical learning.

No, there isn't. But there's also no way around the fact that drilling without understanding is why a whole bunch of students who are "good at math" can't get through even the most dumbed-down versions of proof-based courses in college, or in some cases can't even get through a full calculus sequence. But they're "good at math" because they can rattle off 12*7 real fast!

New Math advocates (and their opponents!) were all at least correct about one thing: we REALLY SHOULD seriously ask what good is learning "math" if the student does not become a better problem solver. It's not 1417 anymore -- problem solving is important, but human calculators don't pull down living wages.

Probably the answer is that we should all be equal opportunity critics: memorization without understanding is intellectually lazy and limits growth potential, while understanding without practice is for most learners a contradiction in terms.


Also Kline spends a good part of his book shredding the standard math curriculum (especially rote memorization and mindless application of standard algorithms) and agreeing that it needs reform, at the end advocating constructivist alternatives, with word problems, use of physical manipulatives, and motivation via applications to other fields.

His beef with the New Math is with an emphasis on axioms, deductive reasoning, rigorous abstract logic, linguistic purity, and symbol manipulation, rather than with teaching conceptually or letting students think for themselves. He also doesn’t like the specific content of the New Math (set theory, inequalities, alternate number bases, boolean algebra, modular arithmetic). [I haven’t studied the New Math curriculum enough for myself to know how fair these arguments are.]

His key criticism: “Psychologically the teaching of abstractions first is all wrong. Indeed, a thorough understanding of the concrete must precede the abstract. Abstract concepts are meaningless unless one has many and diverse concrete interpretations well in mind. Premature abstractions fall on deaf ears.”


Tying students' progress through the curriculum to their age is silly. If you think about it for a single second, the whole thing is totally absurd. Vague pre-puberty/tween/teenager distinctions make sense, but within each grouping/building? totally unmotivated.

Yearly age-to-competency distinctions continue by sheer force of tradition, and are harmful to all but the totally mythological "average" student.


But it works at-scale, which is where the focus is. Bang-for-the-taxpayer-buck.

Further, socially-speaking, being with your actual peers is extremely valuable, obviously.

But yes, the ideal would be a tailored-to-each-child education. Personally, having seen the power of a good Montessori education, I think it's frickin' genius and should be the template for all education (specifically the grouping of 3 yrs together -- e.g. 3,4,5-year-olds together), so kids are perpetually working through the cycle of "look up to someone, mimic someone, mentor someone".

... but I digress.

Fact is, grouping by age works socially and it's much, much cheaper than a tailored education, which is all we an afford (get taxpayers to pay for).

At some point, I expect a software company to make headway in this space and you'd see a bunch of kids staring at iPads all day and a bored teacher playing Minecraft at the front of the room, only engaging when someone gets stuck.


My kids are homeschooled, and I'm in charge of teaching Math, because my wife is one of those "I suck at math" people. Her parents focused exclusively on memorization when they helped her with homework, and then she got a few bad teachers, so her mental models are completely off. She's getting better, but that's not really my point.

The math program I use with my kids is this one: http://www.defimath.ca/ecole-maison/ (in French)

It was developed by two mathematicians who actually studied in classrooms what worked on kids when teaching math. They've come up with a method where kids learn negative numbers and multiplication _before_ learning positional numeration (numbers greater than 10, with units, tens, hundreds, thousands, etc.). I've seen first-hand how these concepts just click in my daughter's mind, and how she often comes up with the new concept herself when you introduce the prerequisites in an order that makes sense to her.


They learn about negative numbers and multiplication before they learn to count past ten? Or they just learn counting without learning how to write the numbers?


The latter. My daughter actually learned it because we mixed different learning methods early on (we don't anymore, it confused her), but you have to admit that writing numbers greater than 10 kinda involves multiplication (by powers of ten, but still), so it makes sense to learn multiplication first. Negative numbers were seen while learning addition / subtraction (the concept was basically adding a negative number), which made a lot of sense too.


That sounds wonderful. My kids are about the right age to do this, and I'm curious, are there any English-language versions of this text?


Cheaper? I also don't know if that's necessarily true. If it's ineffective for many, is it a cheaper solution or just simply cheaper?

Personally, I think Khan Academy is one of the few forward-thinking organizations doing something right in education. It tries to skew learning towards a 'mastery-based' model which keeps kids from falling behind by not advancing them too early. Of course, teachers and parents still need to implement the usage of Khan Academy in this way for it to be effective.


>Further, socially-speaking, being with your actual peers is extremely valuable, obviously.

I would disagree... I think that there is huge value to being around a variety of people, and a strong negative value to only being around people in the same phase of life as yourself, especially when you are in the nasty phase. I think I made it through that part of my life because I had an after-school job where I fixed computers for a local office. I even enrolled in the vocational program at my highschool that let me out of school early to go to work (except every second Monday, when we learned how to spot shoplifters, short-change scams, and other hazards of the retail life.)

I think the number of people who felt good about their social lives, the things they did socially and the things, socially that were done to them in high school is... small. As far as I can tell, for most people, college is important because it is a kind of recovery from this, and prepares you for a workplace where conflict is muted, where yes, if you are good enough, you can still be an asshole, but where being an asshole has a pretty heavy cost that must be made up in other ways.

For me? Being expected to behave like an adult while being treated like an adult around a bunch of adults was amazing. It gave me a reason to keep going in high school, and when I came of age, I was all set to get a really nice job.


Not being argumentative -- is the current system actually cheaper? I'm not enormously familiar with Montessori, but is the student/teacher ratio vastly different, or are the various learning materials any more expensive over time than the books, computers, and reams of paper that traditional schools use?


I went to a public elementary school which had only mixed-grade classes (2–3 grades at a time), and it was just fine, and didn’t cost any more than having one grade per class with kids all precisely the same age.


>Tying students' progress through the curriculum to their age is silly. If you think about it for a single second, the whole thing is totally absurd.

Sounds quite logical to me. If we assume a gaussian distribution (which tests seem to verify), most kids of the same age will have the same skills/level.

So at worse you mismatch what's taught to some kids towards the edges, whose level there are ways to accommodate anyway in most school systems (skipping a class or two for extra smart kids, or staying behind/supplementary teaching for less than average smarts).

And it's not just about learning and who can cram more into one's head (akin to e.g. preparation for the Olympics), but also about sharing the same teenager and adult-making experiences as other kids of your age, which is probably even more important that what's actually taught (the majority of which most people will forget anyway).


> Sounds quite logical to me. If we assume a gaussian distribution

In this context, Gaussian is a pretty useless assumption without fixing a variance. Proposed alternatives range from "already implemented" to "totally infeasible" depending on variance.

> which tests seem to verify

Not really. For each individual subject area, maybe, and again, Gaussian is pretty uninformative.

But the odds of a student being "average" in every subject area != the odds of a student being "average" in a given subject area.

> whose level there are ways to accommodate anyway in most school systems

Except the whole point is that there are not currently ways to accommodate this in most school systems! From GP:

>> Schools (at least around here) do not hold kids back anymore. No matter what.

Also notice that holding back a student in math is possibly net detrimental if the student is not also behind in English and Science.

> but also about sharing the same teenager and adult-making experiences as other kids of your age, which is probably even more important that what's actually taught

Again, the prepubescent/tween/teen division is much less granular/restrictive than the competency-by-age-X division.


Sounds a bit circular: when we test kids taught in single age groupings they have "narrowly" grouped grades. Therefore we should teach then in single age groupings because they have narrowly grouped grades.

That doesn't really help us to know if the top achievers are being held back, for example. Perhaps we can speculate that lower achievers are being pulled up?

Outside school you're almost always going to be in age diverse groups; I think a larger element of that in public schools would be better. IMO it helps to emphasise that children are there to make their own education and not simply to be part of an age defined peer group where it appears you're doing something just because of your age, not because of the educational opportunity.


Encourage her to give it another try when she gets to late high school or college. I know a lot of math phds who fell out of love with math around middle school and then fell back in love during their first proof-based course in college.


Yeah, this.

I hated math in middle-school and early high-school. It was boring and rote.

But proofs based math, and learning how to think of math as a language, instead of a collection of overly specific "solutions" crammed into my head by teachers, was the key.


I never got proofs. I wish I did.


The thing that made it click for me was getting a "first principles" book for something relatively simple like Linear Algebra (or some basic CS-flavored discrete math book). These books prove everything, and reading the proofs the author wrote made me recognize what you needed to do to prove something.


Would you recommend one?


> No amount of explanation and reasoning will help the students and instructors get around the fact that they just have to spend the time to develop fluency with the basics.

Conversely, no amount of memorization and drilling will provide students with the intuition necessary to solve problems just beyond the grasp of naive symbolically-driven proof search algorithms.

In a typical non-proof-based calculus sequence, tricky u substitution and integration by parts problems are the two places where you really see "I was good at math in high school" students start to struggle because they lack a good conceptual foundation in trig and logarithms. Up until that point the students with trig identities etched into their skulls by practice and drilling do just fine. But from that point onward, they really start to struggle because they need a bit of geometric intuition and some practice with more effective problem solving techniques than "blindly apply equations until something works".

Really, both are needed -- drilling and practicing, as well as a solid (and preferably proof-backed) intuition about elementary properties of trig funcations and logarithms.


I agree, you need both. I am not optimistic about the state of mathematics education. So many things have to "go right" for a student to master mathematics.


Interestingly, my experience was that your two sentences are totally contradictory. Physics and engineering was all about using memorized formulas to solve word problems, and mathematics was where you learned how those equations systematically fall out of some obvious assumptions.

I think the real take-away is that both physics and mathematics can be taught poorly, and both can be taught well.


i get a feeling you're not talking about elementary school. but i agree - both can be taught good or bad. bad is dumb memorization, good is understanding real world phenomena.


Reminds me of this quote:

"The view that machines cannot give rise to surprises is due, I believe, to a fallacy to which philosophers and mathematicians are particularly subject. This is the assumption that as soon as a fact is presented to a mind all consequences of that fact spring into the mind simultaneously with it. It is a very useful assumption under many circumstances, but one too easily forgets that it is false. A natural consequence of doing so is that one then assumes that there is no virtue in the mere working out of consequences from data and general principles."

Alan Turing


You might want to edit that to fix the name. It's "Turing", not "Turning".


Thanks. That's what I get for posting on my phone :)


Basically, in most college courses the problems are either 1) easy enough that implementing a program to solve them isn't super insightful, or 2) difficult enough that complete automation would mean "do research".

I'll limit myself to Math since that's the topic of this article:

Calculus sequence: CS 1 is not a pre-req. And there's not enough time to teach both CS 1 and Calc 1/2/3 in a single course. "Implement it" works well for derivatives but not integrals. You're not gonna teach Risch, and implementing integration tricks isn't particularly insightful IMO. The cost/benefit ratio explodes in Calc 3, and the physical intuitions become as important as than the calculations.

Everything past that is proof-based and now you're kind of in "your homework is an open research problem in combining NLP with theorem proving" territory. Maybe with the exception of particularly bad Linear Algebra courses and a bit of the early stuff in Algebra.

From a "pragmatic skills" perspective, this approach is still highly suspect. E.g. no one's going to invent their way to Risch by implementing integration tricks.

Point is, every field teaches useful life skills / knowledge, and programming gets in the way as often or more often than it helps.


> So apparently giving up and having the wrong answers (or no answers at all) is what professors would prefer. Cool :)

It's much easier to correct known unknowns than unknown unknowns.

The best advice I was ever given as a student was "Pretend like you'll get an automatic A on your report card and treat your grades as a feedback mechanism for figuring out what you do and do not understand. Just focus on learning." As a simple corollary, cheating is kinda silly.


>It's much easier to correct known unknowns than unknown unknowns.

This assumes you'll get any kind of personalized feedback and instruction in a lower division maths course. Good luck with that at a public university.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: