Is saying "...so I know how science works." likely to convince people?

[FlatAssembler]

[Peebothuhlu]

Hello!

At work.

Well.... personally, in chats I've been having with colleagues. When I've posited that "The concensus of such learned folks is...."

I tended to get the reply of "Oh but those science types...." as a way of dismissing my side of said conversation.

Must adimt it was a tad disconcerting.

[bennyboy]

Richard Feynman, so far as I know, never said, "I'm a professor at Caltech, so you'd better listen to me." I'm pretty sure he brought forth the clear knowledge of science.

[Anomalocaris]

I would say, “so tell me, how does science work?”

[Dr H]

I trained in the "hard" sciences and technology (physics; electrical engineering) and worked in that arena for some years.  Later on, I acquired training in so-called "soft" sciences (developmental and educational psychology) and more recently have worked in that arena.  
Oddly enough, the most consistently useful training I have had is in philosophy, because it taught me to bullshit convincingly, which, properly applied,  I have found to be an asset in most arenas.

A great many non-scientists who think  they "know how science works", really have no idea.  One dead giveaway is the statement "science proves . . ." followed by pretty much any positive assertion.  The only thing science ever proves, is to prove something false.  In the pursuit of  falsifiability one may amass evidence in support of an hypothesis -- but the hypothesis is never established in an absolute sense; it must remain potentially falsifiable, else it is not a valid scientific hypothesis.  A lot of people find this concept very hard to grasp, or if they do grasp it, very hard to accept.

As you know, it can by significantly more difficult to set up properly controlled, blinded experiments in the social sciences than the "hard" sciences.   For starters, there are almost always more uncontrolled (and unknown!) variables when working with people and human interactions.  And you can't just isolate a person into a petri dish or a cloud chamber for six months to see what happens.  Even setting up a control group study means getting an acceptable human research protocol, with voluntary informed consent, etc., through IRBs and ethics committees -- no mean feat.  And one reason why so many social science studies use comparison groups, rather than control groups.

Still, it can be done; it is done; and meaningful results do emerge.

So, what do you think, is saying "I've published three papers about linguistics in peer-reviewed journals (about...), so I can safely tell you that's not how science works." more likely to be productive or counter-productive?

It's certainly relevant -- it establishes your credentials prior to pontificating in your field of expertise. 
Whether that's productive or not, is another question, and probably depends on context.

[The Grand Nudger]

Richard Feynman, so far as I know, never said, "I'm a professor at Caltech, so you'd better listen to me."  I'm pretty sure he brought forth the clear knowledge of science.

-but you can imagine him saying it, with that face and in his style, and how hilarious it would be, lol.

Flat....got a link to your peer reviewed work?

[FlatAssembler]

[Dr H]

1. Science is based on absolute proofs which are unscientific to doubt.
Sometimes, as in mathematics, it's possible to do that.

Math, in fact, may be the only field in which it is possible to have that kind of proof.

3. If people behave like all the scientists agree on something, that means they actually agree on that. In other words, that what journalists write about science and what our textbooks write about science is true to a very high degree.

The corollary to that assumes that valid scientific explanations are determined by popular vote.  The old "4 out of 5 dentists agree" approach, which always makes you wonder why that 5th dentist so obstinately refused to accept the truth.  If argumentum ad populum were an indicator of validity, we'd have perpetual motion machines, since so many people seem to believe they can build them.

One dead giveaway is the statement "science proves . . ." followed by pretty much any positive assertion.

Yeah, you get the point.

As you know, it can by significantly more difficult to set up properly controlled, blinded experiments in the social sciences than the "hard" sciences.

Well, yes, but think of what the physicists are currently studying. Studying subatomic particles is incredibly hard, both because they don't behave (as far as we know) completely predictably (the only laws we have about them talk about probabilities) and because of the complicated machinery needed to study them.

True, but complicated equipment and difficulty of experimental design aren't the same sort of problems as not being allowed to do certain experiments due to ethincal concerns.  If you want to study photon/pion interactions at one of the big accelerators, and have published plausible preliminary work and can raise sufficient financial backing, you get on the waiting list, and it will probably happen.

If you want to find out if hunger is a significant factor in motivating child neglect by selecting an at-risk population and withholding food stamp benefits from half of them, you're probably going to have a tough time getting that one off the ground.

Also, not all natural sciences rely on experiments. Doing experiments in astronomy or in meteorology is obviously either completely impossible or impractical, yet their conclusions are quite certain.

It stretches the definition of "experiment", for sure, and a lot of it comes down to creative meta analysis of huge amounts of data.  But experiments can be done, even in astronomy.  A lot of them rest on the axiomatic assumption that physical interactions observable locally would necessarily be the same under similar conditions, anywhere in the universe.  But once that's done, predictions can be made that are testable -- that's certainly an experiment.

Whether that's productive or not, is another question, and probably depends on context.

I mean, like, is it likely to convince people that I understand how science works better than they do? Because many people on the Internet seem to have this idea that social scientists don't really know how science works.

Some of them, maybe.  Not most, probably.  Those already inclined to think critically about things will appreciate the credentials, so they know you're not just plucking things out of the air.  Those who don't think critically... it probably won't make any difference with them.

I spent several years in a discussion forum largely populated by a crowd of true believers, with a spattering of scientists, engineers, technologists, etc. thrown in.  We debated everything from astral projection to homeopathy.  People held to some ideas -- "higher consciousness", comes to mind -- with literally religious fervor.  Nothing I could say was going to change their mind, any more than my reading from one of Richard Dawkins' books is going to convert and evangelical Bible-thumper on my doorstep to atheism.  That I knew from the onset.

What I hadn't counted on, but quickly learned, was that people who believed other forms of woo often thought they were supported by science and logical in their beliefs.  Homeopathy was an excellent example of this.  It's whole premise is ridiculous -- I'd even say, superstitious -- and it flies in the face of rationality.  It has been tested scientifically, many, many times, and been found completely wanting and bogus.  Yet true believers will cite their own "scientists", experiments, and journals, and any hard evidence I could bring up counter to it was dismissed as propaganda put out by a conspiracy of mainstream science.

In effect, despite my background -- and those with far more impressive backgrounds than mine, including doctors and research biologists -- and 150 plus years of peer reviewed experimentation, the homeopathists were maintaining that they were better scientists than the rest of us.

What can you do?  
After a while, one gets tired of arguing with the wall, finds the door in it, and leaves.

Anyway, what do you guys here think about the hierarchy of sciences? Most of you haven't said anything about it. Do you think that idea has any merit today?

I think it's something that concerns philosophers of science more than scientists, per se.

Certainly there are some issues with Comte's original construct that views astronomy as one of the simplest and most general sciences.  Trying to reconcile the enormous complexities of relativistic physics on the galactic or universal scale with the enormous complexities on the quantum scale is the holy grail of modern physics.  As we've moved down the hierarchy we've only begun to discover how much we missed on our way down.

Basically, I think it's an overly simplistic view.  It's like Newtonian mechanics:  good enough to get you to work every day, maybe eve good enough to get you to the moon and back.  But if you want to get to the stars your going to need Einstein, at least.

The sciences are organized more like a neural network than a hierarchy.

[Jackie]

It isn't a matter of having the personal ability to be able to be a neurosurgeon or nuclear scientist. I cant do a fraction to save my life. 

But, just like one can have the basics of knowing how to drive a car, without having to melt metal to build a engine block,  or frame, one can understand the principle of how a combustion engine works. The spark plugs ignite the gas, which expands, which forces the pistons to  move, in turn, turn the crank shaft, which turns the axles then the wheels.

Scientists are humans. Scientific fields are what those humans study. But scientific method itself is a very simple principle(tool) that all scientists use.

The principle(TOOL) is very simple to understand. Just like driving a car doesn't mean you need to know how to build one from scratch.

1. Collect data on prior established methods.

2. Plug data into established formula with a large sample rate and control group.

3. Observe data.

4. Repeat experiment multiple times to see if the outcomes produce consistent results.

5. If they do, you are onto something, if they don't, then check for a flaw in your data, model, formula, methodology. 

6. If you are finding consistent outcomes then you turn that over to peers within your field and let them try to replicate the same tests the same way you did.

7. If they come up with the same findings, then you are onto something. If they do not, then you start over.

^^^^^^^^^

Scientific method is a principle, not a field. It really is no different following steps, just like driving a car. You obey the traffic laws, and if you get pulled over by police(peer review) that means you didn't get something right.

[FlatAssembler]

Anyway, a little joke from my real life about soft sciences and hard sciences:
In the summer before my second year on the university, my father asked me which subjects I will have that semester. Then I named, among other things, object-oriented programming. And my father asked me: "How? Object-oriented programming? Really weird name! And, is there also some subject-oriented programming?".
Then, after a few weeks, we met with some friend of him. So, my father asked me: "How is that subject you have this semester called, again?". So I repeated "object-oriented programming". And then my father asked his friend "So, can you guess what that name refers to? What's it supposed to describe? Can you think of a name that's more stupid?". Then the friend said "Well, I guess it's called object-oriented programming because it's mostly the engineers and mathematicians who do the programming. If it was done mostly by historians and economists, then it would be called subject-oriented programming.".