Slashdot Mirror
Terrible Advice From a Great Scientist
Shipud writes "E.O. Wilson is the renowned father of sociobiology, a professor (emeritus) at Harvard, two time pulitzer prize winner, and a popularizer of science. In a recent article in the Wall Street Journal, Wilson provides controversial advice to aspiring young scientists. Wilson claims that math literacy is not essential, and that scientific models in biology, intuitively generated, can later be formalized by a specialized statistician. One blogger calls out Wilson on his article, arguing that knowing mathematics is essential to generating models, and that lacking what Darwin called the "extra sense" is essentially limiting to any scientist."
Math, intuition, and insight are all important. But they don't all have to come from the same person. I have worked on plenty of teams where the creative work and number crunching tasks were delegated to different people. I am currently working on a 3D educational game, using OpenGL. It involves lots of gnarly trig and vector math, which I am good at. It also involves lots of creative scene design and character development, which I am not good at. So I work with an artsy chick, and we make a good team. I don't see why splitting creativity and implementation shouldn't work for biology as well.
Say something wrong that people want to believe, then block the box for 30 years.
Fugue for Aaron Swartz
From that WSJ article: "If your level of mathematical competence is low, plan to raise it, but meanwhile, know that you can do outstanding scientific work with what you have."
I don't really see anything wrong with telling people to still keep thinking about things, find out what they like to study, and get more math. More 'don't let current lack of math get you down' than 'you don't need math at all'.
Code or be coded.
Sociobiology is theories about how observed human behavior and social structures have arise from evolution. Where does cooperation come from? Where does homosexuality come from? How are these traits beneficial for animals and humans, and why haven't they been selected against? Sociobiologists come up with plausible and reasonable sounding theories for many of these, but most of them remain just guesswork if there isn't hard data and hard mathematical modeling (many remain just guesswork even with data and models). Wilson is right that you don't need to be proficient at math to succeed at science. But that's perhaps more a testament to the poor criteria by which some areas of science measure success than to what a scientist actually needs.
Math is not necessary -- in fact it can be a serious liability -- in formulating hypotheses. For instance, much of sociobiology. On the other hand, it's indispensable for testing those hypotheses and sorting the valuable ideas from the attractive bullshit.
Which category holds much of sociobiology is a question beyond my own skills.
Lacking <sarcasm> tags,
That's like literature without words...
“He’s not deformed, he’s just drunk!”
Math is how I weed out my bad ideas from my good ideas. When I find something that seems like an insurmountable road block it's because the math points towards what a poor choice it is, more than anything else.
YMMV.
The article effective asserts that it is OK for a scientist to be unable to "rapidly alternate between experiment and quantitative analysis..." This is ridiculous. That the author was lucky enough to stumble on a fertile phenomenon, and be able to communicate well with a mathematically literate person who was not too bogged down in his own work to help (this is less likely today), and do it all before someone who did not need interpersonal communication ate his lunch and published first, is a freak event. Today if some phenomenon is getting funded for study, the quickest to iterate experiments to publishable results will be the doubly-literate.
The person who collects the data is not usually the same person who draws the statistical models from the data, despite what you may believe. Like engineering scientific endeavors are nuanced and multifaceted. Many roles that are critical to discovery, do not require math literacy. The fact that people get up in arms about the defense of math (like yourself) indicate a problem with education.
Are sensationalized bullshit. The original article did not make that claim, only that you shouldn't let a fear of maths or advanced maths prevent you from a career in the sciences. Obviously, don't plan a career in Physics, but there are plenty of interesting areas of study that don't require Calculus+ areas of math proficiency (sociobiology being one).
As an ECE, most of my studies were centered around differential equations. However, my sister, who did biology/chemistry(two hard sciences) with an intent to move on to dental school, hardly had to touch maths at all.
while(1) attack(People.Sandy);
... and that infamous father of sociobiology lie to us. Why should people believe what the blogger says? I'm very curious.
nop, nop, nop #VBLANK
What would Feynman say!?
Jiggity
Sure, the roles do require "math literacy" which is a lower standard than "sufficient mathematical and comptuational capability to independently produce results for a research journal."
Just like natural language literacy is a lower standard than powerful, skilled writing.
My 'aunt', who still works for a pharmaceutical firm analyzing statistical analyses by researchers, would snort tea out her nose reading this. Doing the research, finding a useful drug, doing minimal testing, and then concocting the analysis to fit the very limited empirical model is not uncommon in the drug industry. Her job was and is to study that 'analysis', identify any problems, send it back for improvement, and repeat until either the researchers give up and move on to something they can demonstrate is effective AND safe enough for the market, or succeed and are able to show provable, reliable results.
Wilson would not like herm, and for good reason - she would call his methods little more than guessing. She has proven repeatedly that well-meaning researchers can find some statistician to lend unwarranted credence to imaginary results.
Kinda sad that this passes as science at all. Wilson seems, to me, to be stating that research need not be proven, merely justified.
deleting the extra space after periods so i can stay relevant, yeah.
The math behind quantum physics and relativity is of secondary importance compared to the phenomena they predict and define. Einstein had the insight that everything must be relative, and the math followed from that. Mathematicians merely model nature based on existing insights. But it are these insights that create new science and discoveries, and not the math that models them.
My karma ran over your dogma
Collecting data without having a darn good grasp of how the data analysis works is a great way to waste a huge amount of time and money collecting mostly useless data. It may not be the same person doing both, but the data-collector definitely needs to be intimately "in the loop" about how their experimental work impacts uncertainties in the final analysis.
No he is not right. Research is not like programming. When coding a program the basic framework already exists: someone comes up with an idea and then someone else can write all or part of the code. Now imagine doing the same with research: someone does an experiment and then another person analyses the data. Chances are that this analysis will be worthless because they have not accounted for all the systematic errors and corrections due to nuances of the experiment itself. To analyse the data you need to have an incredibly detailed knowledge of the experiment and understand it well enough to figure out all the corrections needed during the analysis. In reverse you also need to design the experiment to minimise any biases and effects on the analysis.
Worse his singling out of a "few disciplines" clearly shows how ignorant he is of fields outside his own. ALL of physics, not just particle and astro, needs what he would call "advanced" maths: calculus was invented to describe newtonian mechanics and quantum mechanics requires that you solve partial differential equations to understand what is going on: indeed using intuition with QM is not likely to end well. Chemists need some level of understanding of QM since this is what governs reactions. Earth scientists need "advanced" maths for seismology and climate modelling and lets not even mention all of computer science (although perhaps he regards this as information theory).
Teams these days are really large, so much so that the data-collector is often not even the person designing the experiment. And that person is not the person doing the analysis of the data, who is not the person designing the mathematical model, who is in turn not the person implementing the simulation software. They all have to communicate in various ways, but they cannot each have all of those skills.
On smaller projects it may be the case that there's a more unified role of "experimental scientist", who does need to do all of understanding the model, designing the experiments, and carrying out the experiments. But on large teams the people actually collecting data need more technical skills, focused on operating various kinds of equipment properly. Someone else has drawn up exactly which experiments need to be run, but getting them run properly is not easy. Hence there are various scientific roles, like laboratory technician, that don't even require advanced degrees.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
Have gnu, will travel.
I would say it's actually less likely today that you will be able to "rapidly alternate between experiment and quantitative analysis" even if you want to. Roles are much more specialized, and labs much larger, than they used to be.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
Many problems are so big that no one person can fully grasp them. There's nothing wrong with not being one person who can take on the Whole thing. If some aspects need specialists, that's ok.
But of all the things to not be able to handle, math?! This is like a writer not being able to spell.
Before gathering data, you've got to design an experiment. Without understanding the measurements and statistics involved, the experiment design might turn out to be crap.
Have gnu, will travel.
Is his book, The Social Conquest of Earth, Wilson takes droves of biologists to task for espousing the theory of kin selection to explain altriusm, accusing them of both torturing their "relatedness" math and also essentially back-solving from a desired result. Wilson makes the case that the theory of group selection (one social group besting a neighboring social group) explains altruism more simply, and occam's razor applies.
We have a a professor emeritus at Harvard, two time pulitzer prize winner saying one thing, a blogger saying another, and the headline looks like the blogger wrote it. Bad slashdot.
Even the least math-y science of biology involves rates of change of growth. That means calculus to me. And, of course, you've got piles of data so that means statistics. And you've got structures so that means geometry. A first year course in each will let you understand what you are looking at and give you the ability to look up what you don't understand. Without that training, you may miss a phenomenon entirely, misperceiving it as randomness.
"Today's scientist have substituted mathematics for experiments, and they wonder off though equation after equation, and eventually build a structure which has no relation to reality" Nikola Tesla
issues here. One is mathematical thinking—this is intuitive, and very difficult to teach; some people display aptitude for this (logical relationships, congruences, dependencies, correlations across qualitative cases, a "sense" for probability that is remarkably in tune with formal outcomes) and others struggle with it even if they become very proficient with Two, which is notation.
Too often, we conflate the former with the latter and call the whole package "math." But in fact, it is a deep, intuitive understanding of mathematical principles rather than incredible fluency with notation and notation manipulation which is needed for innovation in science and research. I know people that have one in spades (incredible "math sense" but poor formal notation skills or vice-versa).
It isn't necessary to have the formal notation skills to the nines to be a good scientist (a good co-author/co-PI can help to fill the gaps that you have), it is absolutely necessary to have habits and patterns of thought that are "mathematically" sensible, and the best scientists that I know are the ones that can look at a dataset and—after an "eyeball test"—have the strong sense that something important is in evidence in this series, or in this column, or in that set of experimental results, etc.—even if they struggle to prove it. Colleagues often come along and, if they are able to listen and grok, can come up with the formalities.
STOP . AMERICA . NOW
Without understanding the measurements and statistics involved, the experiment design will most certainly turn out to be crap.
Here, fixed that for ya.
Ezekiel 23:20
Any 4-year degree from the same college costs you the same amount of time and money whether it is a degree in Art History, English or Electrical Engineering.
;-)
The value of the degree in the marketplace tho is totally skewed towards mathematics. The more math you have to take to get your degree, the more money it is worth in the marketplace. Compare Computer Tech degree to Computer Science degree to Computer Engineering degree.
E.O. Wilson is perhaps technically correct about -needing- math early, but he is socially incorrect as far as how the populace in capitalistic countries values knowledge of mathematics. (And frankly, I think the capitalists are correct
Quantitative analysis is easier and faster than ever, if you buckle down and learn a semester of statistics and something like R, Octave, or Matlab. If you are unable to learn these two things, while that does not make you a bad person or hopeless scientist, it makes you a relatively bad investment of research resources. Requiring such work to funnel through some other person on a team injects delay and possible confusion into the system.
The 'great scientist' article is telling people, "don't be afraid of studying science just because you aren't good at math." He points out there are plenty of fields that don't require much math (as opposed to physics). He doesn't say math isn't useful, good or important, he merely says that you can still be great even if you're not good at it.
The blogger is irate, angry, and irked. He lashes out with his words. Thank goodness we have bloggers in the world to be angry at great scientists.
There's no reason to be afraid of science just because you are bad at math.
"First they came for the slanderers and i said nothing."
Of course, all scientists need to conceptually understand basic concepts like the different measures of central tendency, deviation, why normal distribution arises, correlation vs. causation and the difference between predictive and explanatory statistics, robustness, and (this is a biggie) conditional probability. But there's no particular reason why they need to know about the Chi squared distribution or the precise mathematical formulas used to calculate these things.
I think the problem is grade inflation and ever more laughable academic standards have caused the sciences to protect themselves by treating math classes as a trial by fire. They want to make you do tough shit to prove you're smart enough to be a scientist, so they always make you go through the details of the calculation instead of making sure you intuitively understand how the concepts all fit together. Which is a goddamn shame, because I've met several medical doctors who've taken three or four semesters of calculus and several of applied statistics, yet they still can't grasp the simplest conditional probability problem. (Which should come up *all the time* re: error rates on medical tests.)
It seems to me that if one were to take this proposition seriously, it should appear as an article in a scientific journal, not the Wall Street Journal.
While I have no qualms with the Wall Street Journal, it does concern me when an article is published for a bunch of MBAs and CFOs that basically equates scientific research as nothing more than a bunch of individualized technicians. Research is not like web design where you have a design architect and a bunch of coders. But the article, phrased as it is, makes it seem that research can be handled in the same way.
The logical conclusion of such thinking in the WWJ is pay for a researchers who have all of these skills when we can just split them up (and save money). While that may be true in the short run, it is not how science advances in the long term. The simple fact is that if you want to be a reasearcher, you need to know the science and the math.
Teams are great and necessary, but the best teams are the ones where the members understand the major parts of the research and that means the math, too.
In the article, Wilson talked about how making it through Calculus ended up giving him all the math he needed to do his own work, and would suffice for much other important scientific work. I frankly thought that his target was not simply the population of smart but "merely OK at math" students who are being deterred from scientific fields, but the gatekeepers of the fields themselves, who would probably reject someone like Crick for his C grade in Calculus. He's not arguing for lower standards, but for more diversity in how we see scientific talent. If the litmus test for the "promising future scientist" were based almost entirely on the verbal SAT score, I can imagine that Crick would be railing against that. But as it stands, he simply thinks the pendulum is too far in the math direction, and this is doing a disservice to science. I find that quite reasonable!
The idea that specialization works isn't new, even to academia. It's a trend that we can observe in many fields; It's glaringly obvious, if you want an example, in industry. With that said, you need a basic understanding to interface. Saying you can be totally 'math illiterate' is saying that pointy-haired bosses are functional.
Only on
Sure, you can learn it, and probably should. But most large-lab workflows aren't set up for the same person doing the wetlab work to also be doing the data analysis, even if they want to. Their job is to stay in the lab and get more data.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
America is the land of pseudo-science, and pseudo-scientists who push the 'right' agendas can easily rise to the top of their profession, and be lavished with all kinds of prizes and recognition.
-The depraved monsters who created and executed the 'scientific' studies to inject healthy black Americans with syphilis, and watch them suffer untreated, were highly regarded doctors.
-The depraved monster who photographed generations of young men and women naked at ivy-league universities all across America in order to push his ideas on race and eugenics was a highly regarded scientist in the same vein as E.O.Wilson
-The depraved doctor who introduced female genital mutilation to the USA (a practice that was widespread up till the 1960s) was thrown out of the UK, but was given a tremendous reception by the medical community in the USA.
-The depraved monster that attempted (and almost succeeded) in having lobotomy as common as vaccination won the highest scientific awards in the USA.
-The racist filth that created the concept of eugenics, and pushed for programs that eventually led to forced sterilisation in countries all across the globe, were given the highest praise by the scientific community in the USA.
-Even today, male genital mutilation is universal across the USA, originally made popular by madmen like Dr Kellogg in the 19th century as a 'cure' for masturbation. Every 20 years or so the US medical community reaffirms the desirability of MGM by claiming it is a defence against whatever illness is currently significant in the minds of the public. It is notable that all the early studies in Africa discovered circumcised males suffered massively INCREASED rates of AIDS infection. When Jewish and Muslim and evangelical American propagandists took control of WHO research bodies a number of years later, magically the results of the studies reversed.
"Government scientist" is an oxymoron. You are either loyal to the fundamental principles of science, or loyal to a current political agenda. The 'scientists' that the general public hears from are not scientists at all, but propagandists. Sadly, many fields of science are very expensive to pursue, and the people that pay the bills frequently have strong ideas about the 'news' they expect to hear.
'Sociobiology' is just today's eugenics- another branch of pseudo-science strongly linked to religious concepts that are worked in order to create the circumstances for new wars on a global scale. 'Sociobiology' is designed to argue that 'war' is just an extension of evolution, just as eugenics and the theory of 'race' was originally created to give a scientific justification of slavery in the USA during the first half of the 19th century. Eugenics flourished in the USA after slavery was ended, in order to counter the concept of "all men are created equal", and ensure the spread of the 'Jim Crow' laws that existed until the 1960s.
Wilson states that to do good science and to be a good scientist you don't need to be a math wiz. Iddo states to be employable in the tech and science field the more math the better. Am I the only one who has noticed these aren't the same point? Iddo is worrying that if your C.V. doesn't show enough math you won't get the position to do the science at all. Wilson says you can find a place for yourself that uses the math you already know. Wilson is optimistic, Iddo is realistic/pessimistic. Wilson succeeded and is a giant. Iddo has watched his students struggle and have to wait tables to get by.
In the end Wilson is following closer to J. Bronowski in Science and Human values and Iddo is closer to my grandmother. Bronowski cared about humanity, grandma cares about me.
Math is just a description and before that can function, there must be something to describe. Ideas comes first, formal descriptions come later.
They go into the lab and discover circuits.
I have to admit, though, that I've never run into one who discovered a good delta-sigma analog-to-digital converter. Or anything else more than trivially complex.
Lacking <sarcasm> tags,
Biology has an advantage over many other sciences. You can apply close study to the subject. But when one crosses over into fields like physics there are situations where the degree of resolution has reached its end and now mathematics becomes almost the singular tool. The ability to use various unusual logics and to reduce the question into equations is perhaps as far as we can ever hope to go. Trying to approach subjects like the underlying fabric that supports the universe or much of quantum mechanics is quickly becoming a mathematics only type of situation.
The "double blind" method was specifically designed to weed out frauds like you.
In the information age, as in quantum mechanics, there is no such thing as useless data. There are only useless conclusions - like yours.
I was a natural sciences major in college and what you're talking about is one or maybe 2 classes worth of math. You don't need calculus or anything beyond that in most cases to design an experiment, obviously depending upon the particular field of study. Statistics itself is heavily derived from a set of formulas that you can look up in a book and the reasoning behind it requires at most intermediate algebra to understand.
I definitely agree that you need an understanding of statistics to design your experiments, but really, the amount of math you really need is surprisingly small given that you're going to want to bring in an expert that's experienced in the specific area you're working anyways. Now, were we to go back in time to days when there wasn't a huge team, that would presumably be a different matter. But, understanding doesn't really require that much math.
TL:DR, you're going to want an expert in dealing with modelling and data of the type you're looking at. It makes more sense than reinventing the wheel every time you do an experiment and forcing people to master not just one specialty, but several of them, and ultimately it's unlikely that they'll achieve a level high enough to compete with the best in both fields.
You make a very strong point. There are often statistical and mathematical modeling assumptions that the researchers are aware of ahead of time, subtle pitfalls in the experimental setup that must be avoided to produce the type of data needed, etc., that the technicians/engineers will be unaware of unless the researchers themselves are directly involved in the experiments. By the same token, it's a good idea to have an engineer involved in the data collection review the research prior to publication to catch any obvious flaws in the modeling assumptions or misuse of the data (even if he doesn't understand everything in the paper). 'Separation of duties' is something that comes from laziness or time/budget constraints rather than being a template for solid scientific work.
Wilson is right. He doesn't say, math is not needed, he just readjusts its priority within the process of scientific work.
I keep hearing this over and over from people. It seems like biologists are never educated in certain aspects of basic research. I know a CS grad student (who also has a degree in biology) who was talking about Google Scholar with some other biology grad students; they started taking notes because they'd never heard of it before. I asked how they managed to cite anything, and I was told that they do get the journals as they are published, and they do read, but they never SEARCH for anything.
Again, the greatest value of math is not the math itself, but the ability to abstract, extract metastructures and isolate higher order patterns from what might otherwise be just chaos or noise. Agreed some, fields are more math intensive that others. Whereas studying primates in the wild (what few are left) mostly needs only the math to get your time and GPS values properly recorded, I would be a little more concerned for the folks a the Large Hadron Collider armed only with algebra. The same goes for most other hard sciences and specializations of the softer sciences.
It may not be the same person doing both, but the data-collector definitely needs to be intimately "in the loop" about how their experimental work impacts uncertainties in the final analysis.
I couldn't disagree more. Ideally, the data collector should be completely out of the loop , and have no knowledge or bias about the hypothesis, or even the purpose of the experiment. This is what "double blind" experiments are supposed to achieve. The people that formulate the hypothesis, collect the data, and analyze the data should (ideally) be three different people, with different skill sets.
Read any psychological journal and you'll see why this is a bad idea. No one knows what any results mean. If you don't believe me, pick the background section of any random article and read to see whether they list any specific details about the results they cite. Most often, they say that there was a significant effect and leave it at that. No one knows how significant, or whether they are studying a phenomenon that is so faint that discovering the nature of the interaction they are looking at will take thousands of participants rather than 100-200. All findings are treated equally, and experimental methods don't take into account the expected scope of the effects which make up the reasoning behind the experiment. Now that this has become established practice, new research is being built on top of the pseudo-scientific findings of the past few years, probably leading to entire theories which don't hold any real weight. These people need an understanding of math badly.
Wilson is (or was) an observational biologist and naturalist. His book "The Ants" is great, but it's a picture book with essays.
For a science to lead to applications, it must have predictive ability. The hard line on this is from Sir Fred Hoyle: "Science is prediction, not explanation". Much of engineering is about prediction - being able to figure out what will work before you make it. Without that, you can't build anything big or complicated and get it to work.
The market for scientists in fields with no applications is small.
http://xkcd.com/435/
Comment removed based on user account deletion
The problem is that you have no clue if your mathematical collaborator is a complete fraud. I've seen this happen in AIDS research. The worst Ph.D. candidate in my graduating class slept with the candidate's advisor so the advisor would blackmail the committee into approving the dissertation. There were so many errors in even basic calculus that passing it bordered on fraud. Said math Ph.D. switched fields to biology and now has a lab named after the Ph.D. at a biology department at a leading university.
'nuff said.
As a somewhat practising experimental scientist, without significant mathematics training, I wouldn't feel at all comfortable gathering/analyzing data and come up with possible improvements in the experiment design or be able to write a paper and defend it.
No, "double blind" doesn't mean the people performing an experiment don't know what hypothesis they are testing for or how the data is being analyzed. In fact, precisely determining and understanding the protocol for data analysis *before* doing the work is a critical component of proper "blind" experiments. "Blinding" means that the experimental protocol is designed so the researchers will not be able to tell which way particular results will skew the outcome while doing the experiment, not that they are ignorant of what they're doing.
Blinded protocol example: researcher 'A' randomly fills numbered bottles with either a solution being studied or an identical-appearing control substance, recording which bottle contains which solution. Researcher 'B' takes the bottles, not knowing which contains which contents, sprays them on a series of petri dishes, and measures how stuff grows on each one. After 'B' collects the data, 'A' reveals what was in each bottle, so analysis (according to a rigorously predetermined procedure) can indicate what effects the solution under study had compared to the control. Blinding has nothing to do with 'A' and/or 'B' being ignorant about what their hypotheses or analysis methods are.
Wash, rinse, repeat
Double-blind experiments are a tiny subset of the types of studies which happen in engineering and science. Yes, they do serve the useful purpose of mitigating 'fudging' of the results by the subjects and the researchers themselves, which is important when the chances of conflict of interest or bias are high (such as with drug trials).
As has been pointed out though, there are significant benefits to having researchers be more hands-on with their experiments. These range from making sure the experimental plan is being properly followed, sanity checks on the measured data (before weeks are wasted gathering bad data), revisions to the plans or equipment if required, coming up with additional tests in a slightly modified setup to verify any 'surprises' in the data, etc. You might be surprised at how often overlooked issues are discovered just by the person being there.
Of course, if you're one of those people who thinks every researcher is going to fudge data, compute a thousand different test statistics to fit their preconceived outcomes, modify the experiment in senseless ways until it gives them the outcomes they want, hide data that doesn't fit a certain model, etc., then I doubt any system of experimentation is going to be good enough for you.
Hey - I am going to take an unpopular position here..
In the past 150 years, truly remarkable advances were made in physics (and chemistry) using math. Please recall that many what I will call "continuous methods" were worked out in a time before iteration and massive sampling, as is so common today. SO Is It Surprising that those studying in the tradition of math, would look to new and important worlds to "conquer".. biological models..
Well guess what, the biological world is orders of magnitude more "messy" than pure gravity and light and such.. You can say good *general* things about ecosystem interaction, but *math does not rule the messy biomes* as it does in the physics lab. At university, I am highly suspicious of many models, and they smack of a hammer in search of a nail to me..
I support EO Wilson in general, and I dont know this speech but I suggest that computer and math people are not so quick to see things only their way..
You're a moron.
Anyway, a simple example of what "wasted" data would mean: suppose you are measuring some small signal with a background of similar or larger magnitude. You can measure signal+background or just background. So, you tell your lab tech to spend the next couple weeks measuring the signal and background, and at the end of the time they come back and say "I measured the signal+background for 100 hours, interspersed with 10 hours of background data." D'oh! If the signal is smaller than the background, then the statistical sensitivity of 100h Sig+Bkg and 10h Bkg is barely better than 10 hours of each; you should have measured ~55 hours of each. Your signal measurements are statistically crippled by uncertainty in the background, and you've wasted a lot of lab time for suboptimal results. You needed people with expertise on all parts of the experiment --- what the expected signal and noise components are, how to allocate time for maximally useful statistics --- to be in on the planning.
He's absolutely right. There are some biologist (some bioinformaticists) who need to be real math pros and the ones that are have a distinct advantage. However, most biologists aren't and they do fine.
For instance, to do qPCR (a way to quantify gene expression) requires a lot of mathematical calculations, essentially calculus and linear algebra. You don't need to know them though because there is great software which does it for you. You do need to understand what its doing though to use it. I've seen people use it poorly because they don't understand it.
So you need "intuition" about how calculus and algebra work, bu you don't need to do it. I know what an integral is. I know what a linear transformation is and how it can be used. I could not though for the life of me integrate or derive anything myself on paper.
30 years ago you could say the same thing for an accountant. They needed to be an absolute whiz with not calculus but also arithmatic.
Now they have Excel, and they don't need to be a master at doing math, they need to be a master at understanding it.
I think most people don't like what EO said because they think he's against math or math education, and he's not. Do you know how you get math intuition? You take a lot of math classes.
"Faraday was an excellent experimentalist who conveyed his ideas in clear and simple language; his mathematical abilities, however, did not extend as far as trigonometry or any but the simplest algebra."
http://en.wikipedia.org/wiki/Michael_Faraday
This is not a science.
Come up with some model of how you want it to work, then pay someone to try and make the math fit.
Troll is not a replacement for I disagree.
A synthetic organic chemist needs pretty much no mathematical ability beyond simple arithmetic and fractions. A challenging math day is one where you get to make a buffer, forcing you to hit the squared button on your calculator.
My sister opened a computer store in Hawaii. She sells C shells by the seashore.
You seem to have some trouble with statistics. I guess you're of these "math challenged" folks everybody is talking about. I dunno if they'll let you into that science-y place or not, but I gotta point out just one little flaw there...
Without understanding the measurements and statistics involved, the experiment design might indeed accidentally be useful. It might even do exactly what was intended and generate a useful result referenced for ages.
And just for future reference, we can tell you're math-challenged as soon as you substitute an absolute in place of a probability. If you're going to try to hide your different-ability and hang with nerds, I recommend to avoid absolute statements in the vast majority of circumstances.
Or, to fix your statement, we might say, "Without understanding the measurements and statistics involved, the experiment design might turn out to be crap." Implying of course, "It might not."
Watching people arguing that one must understand a great deal of math before one can deal with science makes me wonder --- What is "Science" ?
Is the definition of "Science" a static one --- that is, there is only ONE WAY of define what "Science" is, --- or, is there more than one way to define "Science" ?
What I mean is, while it is true that a person who understand a great deal of mathematical concepts (while not necessary a mathematician) may arrive at a particular "enlightenment" faster, it does not mean that a person without great deal of knowledge in math can't discover something new
Muchas Gracias, Señor Edward Snowden !
> Any scientist needs to understand basic maths, notably statistics.
Bullshit. Statistics is not "basic" and science is not dependent on math. You're fitting an argument to your own bias.
> Science (from Latin scientia, meaning "knowledge") is a systematic enterprise that builds and organizes knowledge in the form of testable explanations and predictions about the universe
This enterprise is not predicated on minimal amount of data. I only need a sample of 1 for scientific analysis of many situations. Look both ways before crossing the street or more practically, to not ingest a bottle labeled "cyanide".
You are the kind of annoying cunt that's criticizing Wilson.
the nice thing about science is that it is a sprawling, diverse subject
you can do Nobel Prize class science without knowing more then basic algebra
And, there are fields where to do Nobel Prize Class Science you need a lot of math; as one eminent gravity researcher said, if I wanted to work on string theory, i would first have to read a 1,000 page book on math...
However, as a matter of logic, since it is possible to do great science without math, the proposition that math is required is therefore FALSE
Did anyone actually read Wilson's article... including the irate, myopic blogger who is projecting his own bias while criticizing Wilson for the same?
Well, I have a professional secret to share: Many of the most successful scientists in the world today are mathematically no more than semiliterate.
In my fifteen-or-so years as an academic scientist, I have found this observation to be 100% correct and I have worked with some incredibly famous and well-respected scientists not unlike E.O. Wilson.
Far more important throughout the rest of science is the ability to form concepts, during which the researcher conjures images and processes by intuition.
In other words, math skills have nothing to do with creativity and science is driven, at its most fundamental level, by creative thinking.
Pioneers in science only rarely make discoveries by extracting ideas from pure mathematics. Most of the stereotypical photographs of scientists studying rows of equations on a blackboard are instructors explaining discoveries already made.
Math is a descriptive language, not an engine for discovery, duh.
Ideas in science emerge most readily when some part of the world is studied for its own sake. They follow from thorough, well-organized knowledge of all that is known or can be imagined of real entities and processes within that fragment of existence. When something new is encountered, the follow-up steps usually require mathematical and statistical methods to move the analysis forward. If that step proves too technically difficult for the person who made the discovery, a mathematician or statistician can be added as a collaborator.
Modern science is too complex for one generalist to do everything (and to take credit for it). These days everyone is a specialist, with a PhD in a very specific subject, and they all work together to bring ideas through to discoveries and eventually to technology. Would anyone argue that the POTUS runs the entire federal government by himself, being a world-class expert in everything from speech writing to foreign policy? Then why is it so hard to imagine that great discoveries are supported by the collaborative efforts of many, with one generally receiving the lion's share of the credit for the actual discovery?
The response of the blogger focuses on the idea that Wilson is an outlier and that, like Bill Gates dropping out of college, his resume should not be used as a template. But Wilson is not arguing that he was successful because he was semi-literate at math, he is arguing that you can be successful by focusing on what your good at, and complimenting your abilities with fruitful collaboration. His reason for making this argument is simple; too many people that would otherwise make talented scientists shy away from the sciences because they aren't good at math.
My two cents: there are, very broadly speaking, two principle kinds of scientists (with many exceptions). There the creative types, who are rarely good with math, often lack attention to detail, but who are astonishingly good at creative problem solving. Then there are the analytic types, who are too skeptical to be creative, are often detail-oriented, but who are astonishingly good at analyzing and understanding raw data. The best science is performed by teams comprising both types of people who respect and trust one another.
Actually, I wrote my thesis on life experience.
The creative side can still be reasonably competent in math, but to emphasize it at the expense of everything else may be a mistake.
At least try it in biology. Science is about experiments, so give it a try. If he's wrong, he's wrong, but you don't really know until you try.
The cost of failure may be a few hundred thousand dollars, but the value of success could be a revolution in science work practices. Seems like a bargain investment to me.
Table-ized A.I.
How can you understand the statistics behind continuous random variables without a knowledge of calculus?
Some of the world's great scientific ideas were thought of first by well-educated mathematically literate science fiction writers. Others still were dreamt up by fantasy writers just trying to get a story to work so they could sell a book for a living. These are all sources of valid ideas and, if someone works to develop them even a bit, that is one less bit the scientist has to do. Essentially, eventually, you need the mathematically (and eventually the mathematical logically) literate to take things further, but there are only so many truly literate mathematicians and mathematical logicians (who truly love their subject for what it is and thus see its true beauty, not its utility as a tool). Divided we are weak, together we are stronger.
John_Chalisque
re: No objects, the syntax is not orthogonal (octave is a clone but seems to have done indices right, at least) :>)
.
A couple of questions for you. I tried to look up "orthogonal syntax" on wikipedia, but the only mention of "orthogonal" on "Programming Languages" is on "weak and strong typing" rather than in the "syntax" section. What exactly do you mean by "orthogonal syntax"?
.
I've played with Octave, but I've never had the chance to play with Matlab proper. What's the difference in how they deal with indices?
.
And does your "no objects" statement mean that you can't define a type and create new instances of it, or is it about "object oriented programming" style availability in writing the matlab/octave programs? Thanks ahead of time for replying to this!
Or did you describe a class of various types of work?
The latter, wasn't it.
Yes it was.
"The things you describe are more the product of being a skilled technician than a scientist..." but you then pretend that you were describing the entirety of the work of EJ Corey...?
If his work was able to be described so completely in such a short space, then he's not worth the prize he got and I rather think that it is YOU he'd be in most vehement disagreement with.
Disclaimer: I have a degree in mathematics. I also have an MD, but not a PhD.
I must disagree with the sentiment and content of your post. Perhaps some of your criticisms were true in the past, or perhaps they are true at unreputable university X, but IME, and in the collective experience of people with whom I work who come from a variety of places, they are not true, both the aspersions regarding MDs and labs, as well as criticisms of MD/PhDs.
During my MD degree, I received specific training in statistics (which in the context of my maths degree I was able to judge as high-quality) and in the critical evaluation of laboratory and clinical trials. During residency, we received specific training in criticism of conduct and statistical methods in clinical trails. In fellowship, we received specific training in statistics specifically (again, it was high-quality and delivered in a classroom setting on the undergraduate side of campus), as well as in the design and conduct of clinical trials and laboratory trials. During fellowship I joined a large laboratory run by an MD and several PhDs. There were other postdoc PhDs and postdoc MDs and their work was equally excellent.
Regarding your aspersion that MD/PhD is "a joke" with "no more rigorous science classwork than they incidentally receive in training for their MD," that is patently false. Again, perhaps at your joke of an institution that is true (although I doubt it), but it is categorically NOT TRUE at any of the institutions at which I have studied or worked. The MD/PhD students take the same course work as any other PhD student and it is separate and distinct from the coursework for the MD degree.
There are also many career paths for MD. Some think little of science after medical school and labor to care for you and your family in the community. Others never see patients again, and after a postdoctoral fellowship quantitatively and qualitatively equal to a postdoc done by a PhD, they work in bench science for the rest of their career. Should all MDs be in a lab? Most certainly not. But to paint with the broad strokes that you have frankly demonstrates a powerful ignorance.
Finally, why should physicians be at all involved in research? The answer is 'translational research.' You don't think that PhDs design and run experimental therapeutic trials do you? They do not. They [PhD scientists] are tremendous, fantastic collaborators in the laboratory, but ultimately to bring information from the clinic to the bench and then return the discoveries back to the clinic requires physician-scientists.
So, instead of ignorantly promoting cross-discipline hatred we should instead nurture young medical students and residents and encourage the brightest to embrace research and the culture of science so that we might all benefit.
Ah, now it comes out. Envy.
It might even do exactly what was intended and generate a useful result referenced for ages.
Yes, because despite the fact that we're trying to solve more and more complicated problems (in such areas as biology, climatology, Earth science in general, psychology, sociology etc.), it still happens all-too-often that the wrong solution generates the correct result, all the way from school exercises (where it's never suspicious on a written exam when student obviously doesn't understand what he's doing and he still comes up with the right answer) up to Nobel-prize winning research. Right.
Look, if I'm "math-challenged" by understanding how wrong you are on this, you're reality-challenged.
Ezekiel 23:20
That sounds more like a poor communication of exactly what to measure.
Physics, Chemistry, and Computer Science re heavily math laden disciplines. Using the computer is not the same and there on some occasions the math can be separated from the computer work. CS ans CIS are quite different when it comes to math. Even the first year in graduate school takes more math than a minor or it did for me. I have enough math from under grad for a minor and grades good enough to become a graduate Assistant, but I still needed more math. Many times the math can be separated out on the job, if there is someone sharp enough to coordinate it or as in graphics, some one can tell you what they need. In one graphics course alone we started with linear transforms, worked up through Fourier transforms and on to discrete math and matrix Algebra.
You need to know calculus; not the formulae (except for maybe integrating and differentiating e^x) but you do have to grasp the concept of slope and area. Schools that try to teach science lite without calculus and try to get around the concepts in other ways just make it harder.
On the other hand, after doing just fine with classical physics and E&M and relativistic physics, when I got to subatomic physics and quantum mechanics and wave functions and so on, I found I had no intuition at all, even do I could do the math well enough to land an A. And that was too uncomfortable a place for me to be, so time for a career change. Now I wonder if the folks who make a living at it have any more intuition than I do, but maybe they're just more comfortable with that.