Statisticians, is this true?
(midwest.social)
People TwitterIn that case probability theorists are the chiropractors of math
I have a graduate degree in statistics and use it heavily in my job. I always tell people that mathematicians study math, but statisticians just use math the study data.
To badly paraphrase Mark Twain..."There are lies, damned lies, statistics, and polls!".
Even the original quote isn't from Twain. He misattributed it to British PM Disraeli, but the saying originated after Disraeli's death.
A bit ironic given the quote.
https://en.wikipedia.org/wiki/Lies,_damned_lies,_and_statistics
https://quoteinvestigator.com/2022/06/22/lies-statistics/
Cool info. Thanks!
Terrance Tao is an absolute beast. If you debate him on this you'll get a rigorous proof back on how you're wrong.
Irl they become actuaries. It's usually applied mathematics and if by mathematician one means someone exploring mathematics alone then this is accurate. CompSci kinda falls half into this and half outside of it.
It reminds me of the statement by Clifford Stoll, (paraphrased) the first time you do something it's science, the second time you're an engineer, the third time you're a technician. link.
In essence most statistician work ranges in the engineer to technician level of mathematics.
Hot damn, that's a take!
I really don't want to take anything away from dentists, cause I got an emergency removal that quite possibly saved my life.
Most STEM scientists i know end up doing statistics instead because it's easier to have a paycheque
I was a music history major, and we had to take 2 math courses, and one had to be Acoustics, which we were all afraid of, and waited until we were seniors to take it, and then it turned out to be pretty easy. I ended up working for a classical record label, and used my acoustics knowledge pretty often during recording sessions. I became the guy they talked to about acoustic treatments in the recording halls and such. We even switched recording halls, when I figured out that the hall sounded better when there were pillows in the seats to simulate people. I got an A in that class.
The other math class I took was Statistics. I was terrible in Math during high school, and barely survived Algebra 2. I was not looking forward to more of that, and statistics sounded like it had a real-world use that I could understand, and I was right. I could understand the math because it had a reason that I could see, unlike the abstract nature of most math. I got a B in Statistics, which I was really proud of.
I still use my statistics knowledge to this day, whenever one party or another starts crowing about a new poll, and I see that it had a tiny sample size, and a margin of error of 5%+, and it is essentially useless poll.
Of course its true. What even is math? Do you know someone who actually does math as a career? Really? Math?
I guess its complicated and I'd say you lose the status of mathematician somewhere between proving the Riemann hypothesis and adding 3 to 3.
Most of them won't do mathematics research. And since realistically it is quite hard to both innovate in mathematics and in an applied field at the same time, the differentiation is valid.
It's like the quote by Clifford Stoll. The first time something is done its science, the second time its engineering, the third time its technicians work. Most people in statistics become actuaries and do the math equivalent of being a technician. They aren't discovering or applying anything new, more so just going through the steps.
I like to think of them as accountants. They sure use numbers, but I'm not sure what they're doing with them is in a'y way shape or form morally acceptable.
I don't think I will ever figure out the order to read these things.
Just... Why? Wouldn't you want it to be formatted in a way that's immediately obvious what order to read them?
I used bluesky lately (which i guess uses the same format?) and it looks like so because we are viewing a reply(3) to a post (2) that cite another post (1)
Thanks. I hate it.
Genuine question: what's confusing about this post?
There are some that are confusingly structured, but this one should be quite obvious
Obvious would be top to bottom. You know, like how every human reads.
For replies, yes. And that is the case. But the first one is a quote, not a reply
And it makes sense to put the reply before the thing they're replying to.
You have to read the posts and then put them in order yourself. I see these a lot and I'm still only ever like 90% sure I'm ordering them right.
The other commenter is right, its a stupid way to list a chain of events considering its out of order and doesn't even start with "1".
Same
If statisticians are mathematicians then so are physicists and engineers.
Using math is different from being a mathematician. Mathematicians discover new mathematical principles, not just make use of existing ones.
Are you a programmer if you program programming languages, or a program programming programmer?
And as everyone knows theorems in statistics are brought to this world by storks.
Yeah I don't think this is the burn on statisticians OP makes it out to be. Lots of technical disciplines use mathematics, like... all of them I think? I don't know of any field that doesn't incorporate math that isn't purely artistic.
Also why are dentists catching strays?
Pretty much all art uses math in some form. Usually not a super formal thing but math is everywhere in art. Mostly concerning ratios.
"Statisticians are the dentists of mathematics" like, they do a very specialised job that a generalist would struggle to perform correctly without their level of specialised training, where performing their duties inadequately or forgoiing their services can completely disrupt the function of various area of life/business?
Statisticians are typically lumped in with mathematicians because Statistics is typically treated like a mathematics course. This isn't really the case with other technical disciplines.
Dentists are catching strays because they're likewise kinda considered "doctors" in a medical sense. They're specialists in their own field that get lumped in with the more general field by a quirk of categorization.
I don't know how it is elsewhere, but in the US they aren't "kinda considered" doctors, they are doctors. They have terminal medical degrees and practitioner's licenses same as any other medical practitioner. They're kinda segmented off from the rest of medical practice because of how dentistry evolved alongside other historical healing practices, but they are doctors.
Second, is statistics not a branch of mathematics? The courses I took on probability and statistics were taught by the math department. I don't see how it can't be. Is it "pure" math? Depends on how you define pure but probably not. Is it "easy" math? Arguably some of it is, though I think people who think stats is an easy science probably aren't very good at it. All that I get. But the idea that it is (uniquely among technical disciplines) "not math" is... confounding to me.
I think the chasm really is within the statistics discipline, less so between statistics a d mathematics. If you're doing statistics research you're essentially a mathematician in the statistics subfield. If you're applying statistics in another field, you're a statistician in the professional worker sense.
I think that's so, but that relationship between theory and application exists for every technical discipline. Computer scientists vs software engineers, for example. The line is usually blurry; I tend to operate closer to the applied side of software, for instance, but I still think about and am informed by the theoretical side, just as theory is shaped by the experimental results of application.
"Mathematics" is an odd case because what people call "pure mathematics" is upstream of even the theoretical side of technical specialties. Like, what a theoretical mathematician might call an applied mathematician, I might call a data scientist, because they're closer to pure theory than I am, but still closer to technical application than the theoretical mathematician. It's a super-theory that underpins other theoretical domains.
Statistics results change based on the lens through which you interpret the data. Pure math doesn't do that. Assigning probabilities is arguably pure math, but assigning error bars is purely subjective. It's more a reflection of the subjective selection and definition processes than of the underlying probabilities.
assigning error bars is purely subjective
I don't know where you got this idea from but it is incorrect. Error bars are used to indicate uncertainty in measurements and they are used to indicate confidence (or lack thereof) in those measurements. Measurement is hard, and precise measurement is harder, so engineers of all stripes use error bars to indicate how precisely their data have been recorded. It's not just a stats thing.
also this:
results change based on the lens through which you interpret the data
happens in every field, including pure mathematics. Look up the axiom of choice if you would like a lot of further reading about the implications of interpreting mathematics through that particular lens. Much as we may long for a "purely objective" language of the universe, free from the limitations of human interpretation, we haven't discovered it yet. The best we can do for now is try to make good assumptions and build from there.
Assigning probabilities is arguably pure math
Not even that is pure math. It depends on your prior knowledge, for example if you think one event is more likely. On the other hand if you don't include prior knowledge/assumptions like one event being more likely you're implying that the prior knowledge behaves in a way that makes your combination of probabilities and data the way it is (for example a flat prior, aka every event has the same likelihood, but in some cases it gets even weirder and would effectively force an absurd prior, so you typically just avoid that by defining some prior knowledge beforehand).
And they make more money because their liability insurance is a hell of a lot cheaper
What's funny is statisticians are generally the ones determining insurance payouts.
kind of, but you're looking at the wrong end. we're talking like 1600 a month for a policy instead of 12000 (some reasonable numbers from 15 years ago i'm pulling out my ass). 20k a year instead of 150k, just for your practice's liability and malpractice insurance, all because you have the patience to stick your hands in people's mouths (which most medical practitioners don't).
also i don't know if dental drills have changed, but wear ear pro when you use one if you're a dentist, DA or hygienist. the brief time a patient hears it won't damage their hearing (much if at all), but all day every day? My dentist friend is profoundly deaf in the ear that faces toward patients thanks to his drill. that's a risk to one's self most medical practitioners aren't willing to assume.
i tell anyone who wants to go into medicine, doesn't know the field they want, and wants to make a lot of money to shadow a dentist for a week. tell them you want to go into dentistry and they'll get excited (they might be able to sell their practice to you and retire once you're out of dental school). at least that was my experience. I don't like putting my hands in other people's mouths.
And by extension, so is a cashier.
Statisticians are somewhere between Sociologists and Psychologists according Mathematicians and between Chemists and physicists according to Statisticians.
https://xkcd.com/435/
Prof Brian Cox once joked that Biology is basically a branch of the Arts
philosophers way to the right of mathematicians saying "oh! I didn't notice you there"
Between chemists and physicists is where I would place them too, as someone with a physics degree.
i'm an economist (basically advanced statistics with weird assumptions).
I'd rank them macro econ > sociology > psych > micro econ > stats > bio > chem > phys > > > math > > > > > > > > > > philosophy
And technical analysis is the astrology of the investing world
As someone who has taught both mathematics and statistics in his life the real difference boils down to proof Vs evidence.
The mathematician is uncertain because Gödel showed no system can prove its own consistency. Proof is (generally) rigourous enough that this is the main issue; once it has been proven (assuming your axioms are good), it's considered true.
The statistician is uncertain because they work with samples rather than the population. There is also the issue of inferring causation even if your sample isn't unrepresentative. With statistics you're always building evidence, but you can never have concrete proof via statistical methods alone.
Also fun fact, given that there is more than one type of mathematics (e.g. platonist Vs intuitionistic), some giving different answers to the same question (excluded middle/trichotomy on the reals), and all of which are equiconsistent, we realise that mathematics really is just a branch of philosophy (i.e. what axioms are you willing to believe).
Everything is philosophy, ultimately
"statistics" can mean two things: it's a field of mathematics, and it's also the application of that field to the real world.
There are many theorems in statistics: the central limit theorem, the proof of the t-test, for example. This is maths.
But if what you're doing is assuming a certain real distribution is normal, or testing for normality of real data, that's not maths any more.
Just like calculus is real maths, but once you're solving real integrals for real scenarios, you're doing science.
7 out of 5 statisticians say yes
Should've used a logit model instead of a linear probability model, eh?
This is a very very very good analogy
Then who are the chiropractors?
Economists, money math is pretty much make-believe
Economists are fantasy writers whose entire purpose is maintaining the falsehood that upholds the wealthy getting to have so much wealth. Much like lawyers whose entire job is upholding the rights and privileges of property owners
I am oscillating between "math is just applied philosophy" and this.
Statisticians in reality are programmers, typically using R or Python to run models. You only ever touch math in undergrad.*
There's a long tradition of skipping hard math, though-- ever have a stats class that has you looking at a t-table for a critical value? That's because it gives us a cutoff to use instead of calculating a p-value (which is hard).
*Note: statistics majors in PhD programs still need the hardcore math. Matrix algebra, calculus, etc. Who else is gonna make the packages we use?
I took statistics with Roger Purves. I distinctly remember him saying that stats wasn't "math" in his intro lecture.
Hehe, I mean I'm forced to teach by-hand statistics to undergraduates and we have to do... arithmetics. Multiplication. Division. Square roots!
It's a pretty established truth that we don't really do math. Lol
Or the surgeons, who started off as glorified barbers, and to this day don’t get the title of doctor, even though “brain surgery” and “open-heart surgery” are metaphors for tasks requiring extreme skill
What? That must be a thing in other countries. Here in the USA all surgeons are doctors.
True, although the ones I know did have the medical qualifications and associated title of 'doctor' and then renounced it when they qualified as surgeons, since it's traditional for them.
Are they constantly contradicting the advice of the rest of the mathematicians?
Mathematics is a search for absolute truth, as proven from axioms.
Meanwhile, this is said about statistics: There are lies, damned lies, and statistics.
And even when used in good faith, statistics tends to move toward approximate truth. Statistics can tell you the exact chance that you'll pull a red marble out of a bag of other marbles, but until you actually pull the marble, it still can't tell you exactly what color marble you'll pull. Run the experiment again, and it may turn out differently next time. You never get absolute truth, only percentage approximations.
Very different than other types of math, yes, where 2+2 is always 4, and you can know it for absolute certain in every case for all of time.
Stats is technically math but it's the softest math you can imagine. A huge amount of it is data collection and interpretation. It works different parts of your brain, requires different skills than pure math
Statistics is a sociological discipline that is based mostly on mathematical models. Choosing your interpretation, and data collection methodology, typically means much more than the math you do on the data.
That's why I was happy that my math course in uni was just 90% statistics
And then some
