Fact, Theory, Hypothesis, Law

[Vekseid]

A basic definition list.

A fact is a piece of observed empirical data. This is subject to falsification, so the best data is of course that which can be verified. Generally, scientists tend to offer the benefit of the doubt to people who gather a piece of bad data, because the more sensitive the instrument, the greater the opportunity for errors, and they can come from weird places.

A scientific law is, to put a banal description, a pattern observed in all available data that is, at worst, a good rule of thumb (This is an edit: MHaji posted a better description below). Whereas a legal law is declared to get people to stop doing something, a scientific law is pointed out when they are adhered to.

Laws are, as a rule, rather simple and elegant. Someone mentions "The law of conservation of mass-energy", and further states the definition - mass and energy may be converted between each other, but never created or destroyed. They also get converted into pithy phrases a lot, like "You can't win, you can't break even, you can't leave the game" for thermodynamics.

Elegance does not always equate to having an easy time comprehending why. The speed of light is the same for all observers, everywhere, no matter how 'fast' they are moving. Worse, the number seems to be arbitrary (or more appropriately, the fine structure constant seems to be).

Laws are at the end of the day descriptions. They are not explanations.

A scientific hypothesis is a testable explanation for an observed phenomenon. The need for testing is key - if it can't be tested, it can't make predictions, and if it can't make predictions, it doesn't actually give us any additional tools to understand the world with. You can claim that God made the Universe in situ last Thursday. Your assertion cannot be disproven. However, your assertion has no predictive value, therefore we cannot draw conclusions based on it.

This distinction is important - you can't prove a scientific hypothesis or theory, only disprove it. This means that its value is entirely dependent on what it can teach.

A scientific theory is a collection of one or more hypothesis that explain a portion of our Universe. To be considered a part of 'scientific consensus', it needs to

1) Explain all previous data
2) Do so in the most concise manner possible
3) Make predictions about future observations that would be surprising without the enlightenment provided by the theory.
4) It should be conservative - staying within its realm and not making wild assertions.

"To explain all nature is too difficult a task for any one man or even for any one age. `Tis much better to do a little with certainty, and leave the rest for others hat come after you, than to explain all things." - Isaac Newton

The Theory of Evolution says nothing of how life actually began, and an astute observer will note that Darwin wrote The Origin of Species, not The Origin of Life. He knew nothing of DNA, even though the most basic definition of evolution we now have requires its presence to even be understood - the change in allele frequency over time. That his theory has stood that test of time is a testament to his conservatism - he did not wildly speculate on how life began, just that life on Earth must have had a common origin.

Yet, despite that, it is still a bold claim. If it is wrong, then somewhere, somewhen, we certainly should have found a creature - a bacterium - somewhere in the world - with decidedly non-normal DNA. It makes a statement about what we should find when we look inside any and every cell native to our Earth, and should we find one single cell that is different, its presence will demand explanation.

Even when exceptions are found, just because a new theory replaces an older set of laws does not mean they fall out of use. We still teach that the Earth is round, even though we have a far superior understanding of the Earth's shape. When you calculate how long it is going to take you to drive home, you don't bother with relativistic modifiers.

Likewise, if Relativity and Quantum Mechanics break down under some ludicrous level of energy density, this does not mean we will stop using Relativity and QM in processor design. We might - I certainly don't know what form a unified theory would take - but it isn't given that it will be simpler.

On the other end of the spectrum, theories can sometimes have their processes observed from start to finish. You can take a microscope and watch germ cells invade a host, you can take a bacteria sample and direct their evolution. Germ theory and evolution remain theories, but these observations are facts, which means that any competing theory must also encompass the parts of the old theory that have been so directly observed. This can be a rather tall order if someone's sole goal is to just toss it on its head.



Something that I find a lot of people who object to 'science' misunderstand is that the sole, overriding goal of science is the discovery of Truth.

The scientific method, at its core, is about making observations, attempting to form hypothesis that make predictions, testing them appropriately, and repeat, until all of Truth is known.

And as Newton said, this is a task that is beyond the capacity of any one person. What we are learning today is the result of centuries of tests performed by our predecessors, and millions of people with the same simple goal.

Learn.

There are two main points I want to express, here.
1) Yes, it really is that simple. Most of the difficulty people have in understanding things is coming to the realization that they can actually understand.
2) You are not going to understand everything. At the end of the day, you have to trust someone at some point in your life. You have to trust a lot of people in modern society. Making sure that trust is merited and failing to do so has consequences far beyond the realm of science.

Thank you for reading.

[Sweet lil shakti]

WOW!

I confess, this is the sort of discussion I would be most passionately inspired by and responsive to, adding my own wealth of knowledge and an awareness that I cannot explain where it came from, but know it to be 'on the verge of the all-powerfully validation by 'scientists.'

I agree with your well thought out and expressed definitions of 'fact, 'theory', hypotheses', and 'law', however, I would like to add that all of these are based on our perception of the workings of the physical world...matter, the dynamics that impact and affect that which is, at a primitive level, able to be conceived of, perceived, and observed, tested, theorized, etc...by 'scientists', who are limited to the subjective observation of what they think they see, understand, observe, and postulate about.

There is one MAJOR factor that has caused us centuries of limited and slow advancement in understanding the forces at work that manifest into our observable physical existance:  those are the laws defined by the WILL of the spirit, which is each soul's inherent 'birthright' (and by that, I mean the soul at birth, not the body housing it in the physical world.)

I have had a passion for quantum physics and have been keeping up to date with the latest developments, revisions, hypotheses', newly conceived ideas, thoughts, observations, etc.  I follow this because it is the ....je ne sais que....ironically, the "inobservable and therefore, irrelevent to consider' forces at work that cannot be explained by scientists....at present, in their limited, narrow-minded, and arrogant attitudes to dismiss anything that cannot be presently measured or tested.  It is 'Neandrothal" thinking that still continues to limit the enlightened physical mind to consider "such whimsical concepts' such as the "Big Bang representing the birth of, or Will of a Super Conscientiousness that became self-aware ('Let there be light!') from the emptiness of a vast darkness that previously existed without a single sentient self-awareness.

I have so much to say on this, but I will only comment that just as cavemen would never have been capable (truly) of conceiving of thoughts, concepts, perceptions, inspiration born of 'inexpicable 'light-bulb' sudden awareness, and just as scientists of the middle ages were dehumanized and incarcerated, if not killed outright for daring to suggest theories (for which they had measurable, consistent data to offer as proof of the mere possibility!) that the "primitive tribalistic minds of a society" could not even unerstand enough to perceive of their 'heretical ideas', so is man today afflicted with the same stubborn, narrow-minded dependence on what can be PHYSICALLY measured, observed and tested.

But, thank GOD....before Armeggedon returns us to a 'reborn physical world' that has risen from the ashes of it's previous corrupted and narrow-minded, judgmental, social tribalistic existance, science is on the verge of uniting that which was previously dismissed as 'mystical mumbo-jumbo' with learned and accepted forces at work in the physically manifested universe of matter (think about your dreams....your dream image has NO PHYSICAL matter, but it exists...etherically, in a place science is just beginning to conceive of).  THIS is the obsessively sought after 'unified theory' that will unite the proven laws of matter with the inexplicable and chaotic activity of the sub-atomic 'world.'

The biggest breakthrough, in my humble opinion, was the conception of the possibility of two things being true at the same time, as postulated by paradox of 'Schrodinger's Cat.'   THAT inspiration expanded the minds of scientists so that it became possible for them to then conceive of the fact that mere observation of a wave-particle collapsed whatever infinite potentiality it had of becoming manifest, into an observable, physically manifested part of the observable world, a point in time, and space.

Understanding the very simple principles of what this suggested had lead science to consider that WE, by our willful action, affect and constantly change what was not manifest into physical manifestation.

THIS is the basic foundation of the Divine power of 'God", (whatever it pleases anyone to call 'Him' or 'it'), and remember....man was made in 'God's image.'

Thank you for the opportunity to share my 'epistimology', which is my life work at the soul level.  I have NO one to ever talk to about my expanded awareness, because most people cannot understand, or they do not want to even consider such 'non-sense'.  They KNOW better, and yet they do not have the decades of knowledge I have acquired by ceaseless and serious dedicated search for the Truth of the meaning of life in this painfully restrictive world we call the Physical Universe.

Respectfully,
~Sweet lil shakti~

[Jude]

And now for another part of the scientific dichotomy:  pseudoscience.

[HockeyGod]

I'm sorry if you didn't mean this to be a debatable topic, but the way it is "stickied" generally gives higher priority. Most stickies are meant to be rules or guidelines of Elliquiy. I feel your opinions are dismissive of quite a number of instances where facts were obtained from knowledge-generating methods human beings have used throughout our evolution that are not included in your paradigm.

What you propose is an important construct in the evolution of human thinking, but your definition of Scientific Method is too limited. I would suggest that you're only offering one side of the epistemological debate and one that often is touted as THE ONLY method of developing fact and that there is only one truth. You use interesting terms that have quite obvious and alarming implications. To put it bluntly, what you've outlined can exist in a physical world where we ignore or dismiss difficult questions. Many researchers abide to rules of science only to be dismissed with a narrow definition of Scientific Method because they didn't produce predictable data.

I consider myself primarily a quantitative researcher and WANT to drink the quantitative kool-aid. It's very comforting and easy to know that 1+1=2 and to dismiss any other observations to the contrary. My area of expertise is social welfare and health policy and as a result I find that I am constantly referring to a qualitative world. It would be both/either unethical or impossible to conduct the type of experimentation on the variables I manipulate to determine causality or prediction. I cannot conduct a randomized control trial on poverty, race, age, gender, education, socioeconomic status, geography, etc. It seems from your definition, I do not develop scientific theory - I have broken your rules 1 and perhaps 2 and 3 (depending on your definition of "concise" - my dissertation was definitely not concise at over 300 pages with a significant amount dedicated to explaining the theory that resulted from the data).

The very nature of the research I do is, by Scientific Method definition, unpredictable. Does that mean it's less important? I can explain a number of phenomenon in great detail and can even make some general statements about data and observations. I cannot tell you, like the Law of Gravity, what goes up must come down. Ever heard of a Law of Poverty?

Sweet lil shakti brings up another interesting point and one that can be uncomfortable to many. Faith. Many who follow the Scientific Method you outline dismiss faith as unmeasurable/untestable. The vast majority of the humans on this planet believe in a number of unexplainable phenomenon. Whether or not knowledge is comprised of testable, predictable, or "proven" has very influence.

I would put forth less stringent definitions:

Facts are statements about observable phenomenon. (That table is brown; That table weighs 32 pounds; That table is Mission style)

The Scientific Method is utilizing appropriate techniques to collect data and make statements based on the data collected. (Observation; RCT; Survey; Focus Group; Time Series; Longitudinal; Cross-Sectional; etc.)

A hypothesis is a value-added statement about observable phenomenon that will be tested. (When you don't eat, you get hungry; When you put the table one way the room looks smaller; etc.)

Scientific theory is a premise of how certain phenomenon react in various situations it can be observational and can be predictive.

Just my little observation on your posting.

[Zakharra]

 But faith isn't science and often is used to dismiss current scientific and provable fact. Science is advanced by observation of things that happen and are proven and duplicated in labratories. Faith doesn't require any proof, it's mainly just an overriding belief that one's view is true even when confronted by facts that can disprove it. Such people often dismiss science in favor of their faith.   

I sometimes disagree with Veksied on things (politics and stuff), but his post was accurate and definitive according to scientific forms. It is accurate and cannot really be disputed unless you like to warp the meanng of the language.

[Jude]

I'm sorry if you didn't mean this to be a debatable topic, but the way it is "stickied" generally gives higher priority. Most stickies are meant to be rules or guidelines of Elliquiy. I feel your opinions are dismissive of quite a number of instances where facts were obtained from knowledge-generating methods human beings have used throughout our evolution that are not included in your paradigm.

Science isn't the only knowledge-generating method that human beings have, it's simply the most reliable when it comes to investigating discrete, rigid phenomena.  Being racist can produce startlingly good results at times (if you're dealing with people who adhere closely to stereotypes for example), but that doesn't mean we should accept it as a proper modality of thought.  Furthermore, science is all about precision, not efficiency, ease of use, or applicability.  Science is not the only way of inferring information about the world around us, and it can't be the only method we use -- day to day functioning requires sacrificing inaccuracy for expediency just to get by -- but it is a higher standard that can only be applied to matters which are capable of dissecting, isolating, and thus analyzing.  Thankfully, as our body of scientific understanding grows, we can apply science to more things; the margins that science cannot enter are shrinking with every new discovery.

What you propose is an important construct in the evolution of human thinking, but your definition of Scientific Method is too limited. I would suggest that you're only offering one side of the epistemological debate and one that often is touted as THE ONLY method of developing fact and that there is only one truth. You use interesting terms that have quite obvious and alarming implications. To put it bluntly, what you've outlined can exist in a physical world where we ignore or dismiss difficult questions. Many researchers abide to rules of science only to be dismissed with a narrow definition of Scientific Method because they didn't produce predictable data.

I think you have a fundamental misunderstanding of what science is.  There's a philosophical debate on whether or not truth exists with any permanence, and science does come down on the absolutist side in spirit, but it makes no claim to be the only vehicle by which one can approach truth.

I consider myself primarily a quantitative researcher and WANT to drink the quantitative kool-aid. It's very comforting and easy to know that 1+1=2 and to dismiss any other observations to the contrary. My area of expertise is social welfare and health policy and as a result I find that I am constantly referring to a qualitative world. It would be both/either unethical or impossible to conduct the type of experimentation on the variables I manipulate to determine causality or prediction. I cannot conduct a randomized control trial on poverty, race, age, gender, education, socioeconomic status, geography, etc. It seems from your definition, I do not develop scientific theory - I have broken your rules 1 and perhaps 2 and 3 (depending on your definition of "concise" - my dissertation was definitely not concise at over 300 pages with a significant amount dedicated to explaining the theory that resulted from the data).

The very nature of the research I do is, by Scientific Method definition, unpredictable. Does that mean it's less important? I can explain a number of phenomenon in great detail and can even make some general statements about data and observations. I cannot tell you, like the Law of Gravity, what goes up must come down. Ever heard of a Law of Poverty?

If by qualitative, you mean that your research delves heavily into 'value judgements' then you're right, your research isn't scientific.  That does decrease the certainty of your findings as well, because subjectivity coupled with the complexity of human psychology can easily result in a whole multitude of errors.  That does not mean that it is impossible to perform a "quantitative" (as you put it) analysis of race, socioeconomic status, etc; statisticians do it all the time and they use scientific principles in the process so that the information they derive is accurate.  However, that information will never give them the kind of rich complexity and flavor that your research produces.

I imagine you employ statistics heavily in what you do; this is good, it helps ground your professional opinions in fact (even though statistics can be misused or de-contextualized), but ultimately you're dealing in opinions once value judgments enter in the equation.  This doesn't denigrate what you do, it is still important and vital to our society, but what you do is fundamentally different from science.  The fact that it isn't as rigorous or solid means it can't be relied upon with the same certainty, but that extra degree of freedom allows you to tackle issues science cannot currently delve into.  It's a trade off.

Sweet lil shakti brings up another interesting point and one that can be uncomfortable to many. Faith. Many who follow the Scientific Method you outline dismiss faith as unmeasurable/untestable. The vast majority of the humans on this planet believe in a number of unexplainable phenomenon. Whether or not knowledge is comprised of testable, predictable, or "proven" has very influence.

Faith is nothing more than belief; there's no need to relegate it to a special status.  Some beliefs can be tested, like the idea that the earth is only 10,000 years old.  However, a lot of religious beliefs in particular are structured in a way that they are completely untestable.  That takes them firmly out of the realm of science.  Philosophers and theologians can argue what that says about religion, that in particular is not science's 'area of expertise.'

I would put forth less stringent definitions:

Science is by its very nature stringent; get rid of that and you're diluting its power by reducing it to the same level as other methods of acquiring knowledge.

Facts are statements about observable phenomenon. (That table is brown; That table weighs 32 pounds; That table is Mission style)

The Scientific Method is utilizing appropriate techniques to collect data and make statements based on the data collected. (Observation; RCT; Survey; Focus Group; Time Series; Longitudinal; Cross-Sectional; etc.)

A hypothesis is a value-added statement about observable phenomenon that will be tested. (When you don't eat, you get hungry; When you put the table one way the room looks smaller; etc.)

Scientific theory is a premise of how certain phenomenon react in various situations it can be observational and can be predictive.

Just my little observation on your posting.

The bolded sections are extremely subjective.  It may not make a big difference in proving the generalities you're discussing, but science doesn't deal in such wide areas on a case by case basis; rules and theories are the result of previous painstaking, specific efforts.  Experimentation is all about making very specific predictions then through directed observation (open-ended observation is incredibly dangerous -- it lends itself to anomaly hunting) confirming or denying that prediction.  The prediction itself needs to be rigidly defined and include straightforward criteria for determining whether the prediction was corroborated or disconfirmed.  The clearer the line, the better.

This simply isn't how the humanities, policy research, or social critics operate.  They don't have that level of impartiality or precision to what they say and do.  So much of what they weigh in on is interpretive, not analytic.  They tell you what data means to them.  We need these voices in our society, but elevating them to the same level as science denigrates science and denies the fundamental imprecision inherent in such estimates.

The core of my response can be summed up easily:  science is not just another voice in our society.  When it is done correctly science produces statistical fact that is more solid and reliable than the opinions of other institutions and individuals, and the restricting, rigorous criteria of what is and isn't science is the very thing that gives it this validity.  It doesn't encompass everything, it can't encompass everything, but in the areas it can weigh in on, what science has to tell is far more trustworthy than anything else.

[Vekseid]

Jude, you do realize what alx meant when he mentioned his dissertation, right?

See here, Jude: http://wilderdom.com/research/QualitativeVersusQuantitativeResearch.html

I'm sorry if you didn't mean this to be a debatable topic, but the way it is "stickied" generally gives higher priority. Most stickies are meant to be rules or guidelines of Elliquiy. I feel your opinions are dismissive of quite a number of instances where facts were obtained from knowledge-generating methods human beings have used throughout our evolution that are not included in your paradigm.

No - I keep topics like this unlocked because I want people like you to be able to present your concerns. I come from primarily a hard science background (computer science, electrical engineering, etc.) so I can see the issue with regards to explaining the scientific method.

The primary purpose of this thread is to address largely quantitative flaws made by people trying to denigrate hard sciences.

What you propose is an important construct in the evolution of human thinking, but your definition of Scientific Method is too limited. I would suggest that you're only offering one side of the epistemological debate and one that often is touted as THE ONLY method of developing fact and that there is only one truth. You use interesting terms that have quite obvious and alarming implications. To put it bluntly, what you've outlined can exist in a physical world where we ignore or dismiss difficult questions. Many researchers abide to rules of science only to be dismissed with a narrow definition of Scientific Method because they didn't produce predictable data.

I consider myself primarily a quantitative researcher and WANT to drink the quantitative kool-aid. It's very comforting and easy to know that 1+1=2 and to dismiss any other observations to the contrary. My area of expertise is social welfare and health policy and as a result I find that I am constantly referring to a qualitative world. It would be both/either unethical or impossible to conduct the type of experimentation on the variables I manipulate to determine causality or prediction. I cannot conduct a randomized control trial on poverty, race, age, gender, education, socioeconomic status, geography, etc. It seems from your definition, I do not develop scientific theory - I have broken your rules 1 and perhaps 2 and 3 (depending on your definition of "concise" - my dissertation was definitely not concise at over 300 pages with a significant amount dedicated to explaining the theory that resulted from the data).

Correct me if I'm wrong, but my understanding of qualitative research is that it's not used in a field void of quantitative study - it asks questions like "Are we asking the right question? What are we actually measuring? What constitutes quality of life?" - certainly a critically important part of science, but it's not the entirety. And there are lots of sneaky ways, once enough qualitative research is done, to start applying quantitative methods to it, and get results from this sort of thing.

Take socioeconomics, for example, and the guy who is trying to requantify Africa's wealth based on actual material and labor value, because when e.g. a Somalian makes two or three dollars a day, there's obviously more to their life to that. My understanding is that that sort of question - and the initial examination - is heavily qualitative. But the very goal is to develop a framework for better quantitative analysis. And this occurs throughout the socioeconomic studies I've seen lately - putting better numbers to what it means to live a good or bad life.

The purpose of the conciseness is to exclude extraneous explanations that serve no purpose but to complicate the issue. That can probably be better stated. I'm not sure why you consider the first point an objection - most hard science dissertations don't develop new major theories, they expand on what is currently known. I fail to see why your dissertation would be any different. Would I be assuming too much to say that you built on a lot of other people's work, and that you advanced your field a bit?

Similarly, the Theory of Relativity builds on Poincare, Lorentz, Maxwell, and others. And people have built on the Theory of Relativity. Your research doesn't exist in a vacuum.

The very nature of the research I do is, by Scientific Method definition, unpredictable. Does that mean it's less important? I can explain a number of phenomenon in great detail and can even make some general statements about data and observations. I cannot tell you, like the Law of Gravity, what goes up must come down. Ever heard of a Law of Poverty?

Your statement here is self-contradictory. If it provides no insight, then you wouldn't be able to make those general statements.

The first Law of Poverty I found said, basically, that one cause of poverty involves people withholding their own effort and labor because they expect others to do it for them end up so impoverished. This might be difficult to put numbers do, but you can still make predictable statements about it, even though there are certainly a lot of muddying factors involved. Just because something is difficult or even seemingly impossible to quantify does not make it unpredictable.

Sweet lil shakti brings up another interesting point and one that can be uncomfortable to many. Faith. Many who follow the Scientific Method you outline dismiss faith as unmeasurable/untestable. The vast majority of the humans on this planet believe in a number of unexplainable phenomenon. Whether or not knowledge is comprised of testable, predictable, or "proven" has very influence.

Whether or not a belief is influential has no bearing on whether or not that belief describes that aspect of the world, however.

Facts are statements about observable phenomenon. (That table is brown; That table weighs 32 pounds; That table is Mission style)

This isn't much different than what I said - easier to read, maybe, but the idea of empirical facts is central to all science - facts must be derived from experiment and observation.

The Scientific Method is utilizing appropriate techniques to collect data and make statements based on the data collected. (Observation; RCT; Survey; Focus Group; Time Series; Longitudinal; Cross-Sectional; etc.)

There are a lot of definitions of the scientific method, but it's still a process, even in qualitative research. You are trying to gain a better understanding of your world. The scientific method is largely about testing hypothesis. If you want me to remove predictable refinement from the definition, though, could you give an example of research that is actually unpredictable? I think you might just be focusing on too narrow of a concept of 'prediction'.

"Predictable" is "What we expect to find when we look at x". Social sciences offer a hell of a lot of predictable statements,

As I see it, qualitative research is basically what it means to do initial data gathering. Do you expect to explain all of human behavior in three hundred pages? Three million?

A hypothesis is a value-added statement about observable phenomenon that will be tested. (When you don't eat, you get hungry; When you put the table one way the room looks smaller; etc.)

This isn't much different from what I wrote, except for there's no real guarantee that a hypothesis will be tested. It just needs to be testable.

Scientific theory is a premise of how certain phenomenon react in various situations it can be observational and can be predictive.

Scientific theories represent what we know of the world.

[HockeyGod]

Wow, wow, wow! I love this thread and it's been up for many months. LOL. I need to digest all the comments more before providing a coherent response to both Jude and Veks. At initial read, I think Veks...we're definitely thinking more alike than different. I think some of your statements perhaps led you to the same supposition.

Anywho, I do want to comment now on one of your points.

There are a lot of definitions of the scientific method, but it's still a process, even in qualitative research. You are trying to gain a better understanding of your world. The scientific method is largely about testing hypothesis. If you want me to remove predictable refinement from the definition, though, could you give an example of research that is actually unpredictable? I think you might just be focusing on too narrow of a concept of 'prediction'.

"Predictable" is "What we expect to find when we look at x". Social sciences offer a hell of a lot of predictable statements,

As I see it, qualitative research is basically what it means to do initial data gathering. Do you expect to explain all of human behavior in three hundred pages? Three million?

You probably did hit it on the head when you said I was using a narrow definition of prediction. In the realm of research methods it is a fairly significant faux pas to have any connection between "prediction" and "qualitative methods." The very nature of qualitative investigation makes it contrary to prediction. It is also the bastion that quantitative researchers hold to when defending the primacy of quantitative research.

In terms of research methodology, that is, when you use the Scientific Method definition of prediction you have the classic formula of A+B=C. You can only determine A+B=C in very sterile randomized control trials where you can control for extraneous variables. In qualitative research the terms reliability and validity exist very differently from quantitative research (credibility and transferability) and it becomes very difficult to account for extraneous variables in the same manner you can with statistical methods. My underlying point is that qualitative and quantitative have a complex relationship. Qualitative is not always the first step in knowledge generation and testing of theory isn't always done via quantitative.

You make a statement that all research is predictable. So an example of unpredictable research? In 1996 the Federal Government passed sweeping controversial welfare reform. Aid to Families of Dependent Children (AFDC) was radically transformed into Temporary Aid for Needy Families (TANF). The research that is conducted on TANF for the most part is not predictive. It is evaluative and descriptive. Research has evaluated the effectiveness of the policy, cost-benefit analyses, trend data on enrollments, demographic data on participants, etc. However, it does not make predictive statements. Don't get me wrong, there are plenty of quasi-predictive statements. For example, a child born on TANF has a higher propensity of being on TANF as an adult. That's fairly predictive...sad, but predictive.

Research and the Scientific Method aren't always about prediction, but they are about knowledge generation.

Whew...gotta go read more now to answer other points

[Jude]

Totally caught with my pants down -- I in fact did not know anything about qualitative versus quantitative, as referenced in the paper.  I need to do some more reading before I weigh in.  I apologize for my error.

EDIT:  After reading up on the subject, I actually feel like my original points which were made somewhat in ignorance say pretty much what I want them to say.

[Vekseid]

In terms of research methodology, that is, when you use the Scientific Method definition of prediction you have the classic formula of A+B=C. You can only determine A+B=C in very sterile randomized control trials where you can control for extraneous variables. In qualitative research the terms reliability and validity exist very differently from quantitative research (credibility and transferability) and it becomes very difficult to account for extraneous variables in the same manner you can with statistical methods. My underlying point is that qualitative and quantitative have a complex relationship. Qualitative is not always the first step in knowledge generation and testing of theory isn't always done via quantitative.

You make a statement that all research is predictable. So an example of unpredictable research? In 1996 the Federal Government passed sweeping controversial welfare reform. Aid to Families of Dependent Children (AFDC) was radically transformed into Temporary Aid for Needy Families (TANF). The research that is conducted on TANF for the most part is not predictive. It is evaluative and descriptive. Research has evaluated the effectiveness of the policy, cost-benefit analyses, trend data on enrollments, demographic data on participants, etc. However, it does not make predictive statements. Don't get me wrong, there are plenty of quasi-predictive statements. For example, a child born on TANF has a higher propensity of being on TANF as an adult. That's fairly predictive...sad, but predictive.

Research and the Scientific Method aren't always about prediction, but they are about knowledge generation.

Well in terms of research, I mean predictable as in - were you to go out into the field and collect the same data again, I would certainly hope you'd expect
1) To find similar results, or
2) If the results are not similar, to work to derive an explanation

Research is usually for gathering data (facts) and testing hypothesis (to expand and develop your theory of the Universe) - which is just a more specific sort of fact. Qualitative research, by its very nature, is much more in the initial fact-gathering stage. You don't necessarily even know what you're looking for. Pumpkin made a comment about how Darwin's original qualitative research is what led to the Theory of Evolution - 'there is something going on here' - he studied it, launched it, but the overwhelmingly vast majority of quantitative study on evolution is of course after he died. You use what you find as building material, and ideally, you can start making quantitative statements with enough qualitative data - economics, linguistics, and so on have a lot of qualitative foundations underneath the quantitative data that now underlie them.

A lot of the stigma about social sciences is because socioeconomic factors are so much more vastly complex. Hard sciences move from qualitative to quantitative fairly quickly - but there are a lot of difficulties in doing the same thing with humans.

[HockeyGod]

Well in terms of research, I mean predictable as in - were you to go out into the field and collect the same data again, I would certainly hope you'd expect
1) To find similar results, or
2) If the results are not similar, to work to derive an explanation

Research is usually for gathering data (facts) and testing hypothesis (to expand and develop your theory of the Universe) - which is just a more specific sort of fact. Qualitative research, by its very nature, is much more in the initial fact-gathering stage. You don't necessarily even know what you're looking for. Pumpkin made a comment about how Darwin's original qualitative research is what led to the Theory of Evolution - 'there is something going on here' - he studied it, launched it, but the overwhelmingly vast majority of quantitative study on evolution is of course after he died. You use what you find as building material, and ideally, you can start making quantitative statements with enough qualitative data - economics, linguistics, and so on have a lot of qualitative foundations underneath the quantitative data that now underlie them.

A lot of the stigma about social sciences is because socioeconomic factors are so much more vastly complex. Hard sciences move from qualitative to quantitative fairly quickly - but there are a lot of difficulties in doing the same thing with humans.

Makes sense when you use it in that definition. I think in research methods the term, rather than predictability, is reliability. More specifically test-retest reliability. Test-retest reliability is the phenomenon that if you do something and get an effect, doing the same thing will likely get the same effect. Another form of reliability, inter-rater reliability, is the phenomenon that if you do something and get an effect, I (a different rater) doing that same thing will likely get the same effect.

Related is validity. External validity is the ability to apply the results from the lab to the real world.

I guess that's my confusion. Predictability in the quantitative paradigm is the determination of how likely C will happen when you put A and B together.

Also, I agree that qualitative can be used as exploratory research - to tease out relationships and then test them...but it's not that limiting. I can give you an example if you'd like...but I'd actually like to posit the best of both worlds - mixed-methods studies. My dissertation actually was mixed-methods. I used a combination of quantitative data (population density, financial variables, time, etc.) as well as qualitative data (interviews with experts).

Wow...I love talking about this stuff! Perhaps I should do some bidding to continue this 1:1 though I guess it wasn't designed for that purpose 

[Snozz]

The truly frustrating part about science, and what frustrates scientists is the fact that "truth" is unattainable. To know everything about everything in existence is, to me, an unreachable goal. The planet changes, its inhabitants change, and those changes can be unpredictable. We can get as close as we can to truth, but the word "truth" in itself can be subjective. We can get as close as we can to understanding the world and the people in it, but change is what prevents us from getting there.
That, and ethics which is an entirely different argument.

Scientific method is both helpful and at times, hindering in my opinion. We can try to account for every variable in a study and isolate the situation as much as we can, but isolate and control too much and you run the risk of removing the situation too far from how it may occur naturally in the world. The lab only tells us so much.

On the other hand, the ability to replicate certain situations and circumstances and our advancements in the study of people in general have come a long way. Though we still have a way to go.

I'm also excited about the field of genetics research.

This is random and probably not that related to this topic, but I wanted to put in my two cents. :)

[Oniya]

Scientific method is both helpful and at times, hindering in my opinion. We can try to account for every variable in a study and isolate the situation as much as we can, but isolate and control too much and you run the risk of removing the situation too far from how it may occur naturally in the world. The lab only tells us so much.

That's when you get into things like 'statistical physics', countering the absence of fine measurement with large sample size. 

[doodasaurus]

I come in here and find this?!  My education is in philosophy of science and my partner is a working scientist!  So, newbie though I am, I am amused to find a sci-phi discussion pinned to the top of the board.  :)

[MHaji]

A scientific law is, to put a banal description, a pattern observed in all available data that is apparently never violated, anywhere in the Universe. Whereas a legal law is declared to get people to stop doing something, a scientific law is pointed out when nothing appears to break it.

I'd like a citation on this, since I can name specific counterexamples. Here's a classic, which I'll try to explain in lay terms so that people who never took (or don't remember) physics class don't get too lost:

Ohm's Law states that if you take a conducting material and apply a voltage difference to it, the current you get is directly proportional to the voltage. In other words, suppose you have a wire, and try hooking it up in a circuit with different batteries. If you plot the current through the wire against the battery voltage on a graph, the graph will be a straight line. The slope of this line is called the resistance of the wire; the steeper it is, the "more resistant" the wire is to transferring electrical charges.

Here's a picture from "GCSE Bitesize" at bbc.co.uk:



In reality, this generalization is extremely useful, but doesn't hold across all materials, or under all conditions. An "ideal resistor" is one where the line is straight no matter what, but anyone whose played around with actual resistors knows that they don't behave perfectly. For one thing, running a lot of current through a material will change its temperature, which can play havoc with oversimplified predictions.

But Ohm's Law is one of the classic, canonical scientific laws; it's one of the first ones that gets taught! So, if we use the definition above, we have a choice:

a) Say "Any cases where Ohm's Law gets broken don't actually count"

or

b) Say "Ohm's Law isn't really a scientific law."

Both of these have serious problems. The first turns it into a circular statement: "A scientific law is an unbroken pattern in nature, except when it gets broken." The second takes a perfectly useful, commonly understood term used by scientists and invalidates it.

And if Ohm's Law seems like a bad case, what about the Ideal Gas Law? It describes the behavior of an ideal gas; any difference between the behavior of real gases and an ideal gas arise because some gases are less ideal than others.

These issues aren't nearly as problematic if we use the following description of a scientific law instead:

"A scientific law is a descriptive (non-explanatory) rule with the following characteristics:

1) It is quantitative.
2) It can be formulated simply.
3) It's not a trivial statement of mathematical truth. ("If you have x dogs and y cats, then the total number of cats and dogs is x + y" isn't the Cat-Plus-Dog Law.)

AND

4) It is generalizable, having been tested across a wide breadth of circumstances. However, its scope does NOT have to be absolutely universal."

Now Ohm's Law does okay! Even in cases where it starts to break down, we can still describe a material in terms of how non-Ohmic it is, and have a useful description. If we are even more careful, and take into account the affects of heating and so on. The Ideal Gas Law also works out now, especially when you add in a few more terms to account for differences in the behavior of real gases.

To say that scientific laws are absolute statements, whereas theories are simply broad models, is a dangerous oversimplification. It misleads people who study science into thinking that textbook laws must be universal, when that only applies to some (such as the Law of Conservation of Energy), but not all. Finally, it leads people to state stuff that's equivalent to the Cat-Plus-Dog Law above; this may be why some introductory textbooks state that the Hardy-Weinberg Law is "p^2 + 2pq + q^2 = 1," if we assume that "p + q = 1".

(Yes, the Law of Conservation of Energy is never, ever broken, once one takes into account that mass is a particular way of describing energy. But I don't know of many other laws in science that hold this level of absoluteness. Even the Second Law of Thermodynamics can in theory, with incredibly tiny probability or over insanely long timescales (longer than the universe's lifetime), be broken.)

EDIT: Edited to add one more caveat to the definition of a scientific law.

[Vekseid]

You are almost completely right and will edit my post accordingly, thank you. Was mostly only thinking of Einstein's postulates and conservation of energy, which are pretty absolute.

[Lrrr]

Just a couple of inflammatory but somewhat informed comments on the scientific method and terminology.

1.  The scientific method does not recognize "facts" as the word is used in common parlance.  It only recognizes observations (data) along with any associated errors.
2.  The scientific method has no mechanism to "prove" anything.  The word proof or any of its derivatives should never be used in a scientific discussion.  e. g. "The length of the bone proves it is from a fish."  Huge no-no.  "The length of the bone is strong evidence that it is from a fish."  Much better.
3.  The scientific method never finds truth - it only finds ever closer approximate explanations (theories) to explain how the cosmos actually works.  Newton's classical physics was superseded by Einstein's general relativity.  After 100 years, Einstein's approximation is showing cracks due to it's inability to play nice with accepted theories of gravity quantum mechanics.  I have no doubt that a new theory incorporating both Einstein's work and the theory of gravity quantum mechanics is forthcoming.  It, too, will be only a better approximation to reality.

Note edits above in red - 2015-01-11

Scientific method:
1. Observe
2. Formulate a new hypothesis that fits the observations
3. Observe some more
4. Test hypothesis with new observations
5. Debate the hypothesis' ability to fit the observations and related accepted theories.  Does the new hypothesis work better than the established theory (if there is one)?
6. If the fit to the data is better adopt the new theory
7. Go back to 1 or 3 depending on the result of 5 and 6.

Most of this has already been mentioned but my brain was confusled so I tried to gather it in one place and state it as briefly as possible.  And now, back to our regularly scheduled program ...

[LostInTheMist]

DISCLAIMER: THE ONLY POST I AM RESPONDING TO IS THE ONE BELOW. I AM MAKING NO STATEMENTS ABOUT ANYONE ELSE'S POSTS OR POINTS OF VIEW. THIS POST ONLY.

But faith isn't science and often is used to dismiss current scientific and provable fact. Science is advanced by observation of things that happen and are proven and duplicated in labratories. Faith doesn't require any proof, it's mainly just an overriding belief that one's view is true even when confronted by facts that can disprove it. Such people often dismiss science in favor of their faith.   

I sometimes disagree with Veksied on things (politics and stuff), but his post was accurate and definitive according to scientific forms. It is accurate and cannot really be disputed unless you like to warp the meanng of the language.

I have a problem with your definition of "faith" here. Faith cannot and should not be used to try and dispute or disprove science, and those who try to use it (many of whom are brilliant doctors of philosophy or theologians) as such have an inherent misunderstanding of science and faith.

But by the same token, those who think that science can "disprove" faith (many of them doctors or scientists) have the same inherent misunderstanding of science and faith.

Faith is not stating that "God created the world". That's a statement of a belief that happens to be wrong. Faith is the belief in an idea that cannot be proven or disproven by science.  If your belief can be proven right or wrong, it's not faith, it's a belief. If your belief CANNOT be proven right or wrong (i.e. God exists; Angels exist; If you're good, you get into heaven; etc.) then it's an aspect of faith.

Note that I'm talking about religious faith above, but there are plenty of things in which we have faith that are not religious. When we're down, we have faith that things WILL get better. Not proven or disproven until it's already happened. When we're jumping from a plane, we have faith that our parachute will open... again, not proven or disproven until it's already happened. You have faith as you are reading this that you will not die between the bathroom trip you're about to make and the seat you're going to take to write a long article to object to my statements. :)

I have faith that someone out there is not going to read my whole post (and/or not bother to consider it before responding), and by the time I wake up, someone will have vehemently disagreed with something I never said, never meant, or never hinted at.

[Lrrr]

You make a statement that all research is predictable. So an example of unpredictable research? In 1996 the Federal Government passed sweeping controversial welfare reform. Aid to Families of Dependent Children (AFDC) was radically transformed into Temporary Aid for Needy Families (TANF). The research that is conducted on TANF for the most part is not predictive. It is evaluative and descriptive. Research has evaluated the effectiveness of the policy, cost-benefit analyses, trend data on enrollments, demographic data on participants, etc. However, it does not make predictive statements. Don't get me wrong, there are plenty of quasi-predictive statements. For example, a child born on TANF has a higher propensity of being on TANF as an adult. That's fairly predictive...sad, but predictive.

The original statement that all research is predictable confused me too at first.   Did you mean to say that all research is done to test predictions of hypotheses Vekseid?  In either case, I, too, take exception and would suggest that research comes in countless forms and is used for countless purposes, only some of which are targeted at testing predictions of scientific hypotheses and theories.

I will say this - if a hypothesis DOES NOT make testable predictions, it is NOT a scientific hypothesis.  The scientific method requires hypotheses to be testable to work properly and this is an area where there is much confusion.  In fact I think this one, single misunderstanding produces most of the heat in the great debate between science and faith (and other related debates).   I'm not at all saying there aren't untestable but still useful hypotheses - just that they can never be subjected to the scientific method since they make no predictions against which observations can be compared.

With respect to the statement "The research that is conducted on TANF for the most part is not predictive.", it's a bitch but there it is.  IF we define "scientific" as an activity conforming to the scientific method, then the vast majority of the research is not scientific regardless of how fancy the statistics or other manipulations are.  That doesn't mean it isn't useful at all, just that it's not scientific.  As previously mentioned, there are many methods of advancing overall human knowledge that don't fit the scientific method.

On the other hand, it would take relatively little effort (compared to the huge overall TANF research effort) to make some predictions using the observations that were gathered.  The scientific method does not place any requirements on the accuracy or reliability of the observations used for formulating or testing the hypothesis.  Therefore, stating that a hypothesis or even a theory is "scientific" makes no statement about its value other than we have nothing better at the moment.  Branding a point of view as scientific lends a certain cachet of legitimacy despite the fact that there is plenty of room for marginally useful hypothesis and theories under the "scientific" label.

One other thing I'd like to mention.  Math does not conform to the scientific method and is therefore NOT a science.  It sure is useful when evaluating things that do conform to the scientific method, but math is axiomatic in nature and built on absolute proofs of theorems through a logical process - not a scientific one.  I suspect that will raise some hackles but I'm willing to let it all hang out if someone wants to discuss the details of the issue.

Research and the Scientific Method aren't always about prediction, but they are about knowledge generation.

I'll make a gentle poke here and say that, although I agree with you about the Research part, I don't think it's possible to use the scientific method on a theory that doesn't make predictions (as discussed above).  Therefore the Scientific Method really is ALWAYS about predictions.

[Robert dElla]

In science, facts are not falsifiable.   An interpretation of an observed or inferred fact might be right or wrong, but that doesn't change the nature of the fact itself.

"Fact" is one of those words that might be described as a term of art.  It depends on the context, and the OP might be conflating the scientific term with the way "fact" is used in the legal sense.

At law, a fact is whatever the finder of fact (jury, judge, arbitrator, etc.) says it is.  It could be anything from "the light was red" to "guilty as charged."  While these facts are not likely to change, even in the event of a witness recanting, it can happen.  DNA evidence might exonerate, in which case the fact of guilt changes to the fact of innocence.

Karl Popper and Thomas Kuhn have a lot to say about this.