This work is available here free, so that those who cannot afford it can still have access to it, and so that no one has to pay before they read something that might not be what they really are seeking.  But if you find it meaningful and helpful and would like to contribute whatever easily affordable amount you feel it is worth, please do do.  I will appreciate it. The button to the right will take you to PayPal where you can make any size donation (of 25 cents or more) you wish, using either your PayPal account or a credit card without a PayPal account.
The Qualitative Quantitative Debate in the Social Sciences
Rick Garlikov

Ray C. Rist's article "On the Relations Among Educational Research Paradigms: From Disdain to Dentente" and John K. Smith's article "Quantitative Versus Qualitative Research: An Attempt to Clarify the Issue" show a misunderstanding of (physical) science, philosophy, and philosophy of science. Rather than clarifying issues or topics they deal with, both confuse them more. This paper is not meant to be a complete attempt to write Rist's and Smith's papers for them or answer the questions that seem to be at issue in the social sciences, if Smith and Rist are right about the problems; it is merely a paper to show a number of important places where Smith and Rist have gone wrong.

Smith does not seem to understand the relationship between ontology (or metaphysics) --what exists or is true-- and epistemology --our knowing about it-- as it pertains to issues in science or in philosophy. Rather than try to explain what these broad categories deal with in philosophy, let me just discuss the relevant aspects for the sciences. 

First of all, the debate of realism versus idealism NEVER is at issue (not legitimately anyway) in the sciences. Science deals with observable phenomena, and observably testable theories about those phenomena. Logic (including mathematics), when appropriate, is also used in devising, arguing, and testing theories. But the subject matter is that which is observable, and the theories "explaining" observable (i.e., empirical) phenomena must themselves be logically possible to test in some empirical (observable) way. (The test may not be physically possible at some particular time -- e.g., to test Einstein's hypothesis that light rays would be affected by gravity required waiting for the right kind of eclipse to occur and being at the right place with the right equipment to record and measure the direction of light rays -- but there must be some logically possible, or logically conceivable, test that will yield empirically confirming or disconfirming evidence for the theory.) Idealists and realists have exactly the same empirical phenomena before them to explain in the way a natural scientist does, and they have exactly the same empirical phenomena in front of them to use and/or evaluate as evidence for, or against, any given theory. The realist/idealist issue is not about scientific explanation.

The realist/idealist debate in philosophy is about something else. It is about what is the ultimate "cause" or basis of our sense perceptions: whether observable phenomena are simply (the result of) mental occurrences (i.e., idealism) or whether there is something non-mental (physical) out there "behind" our perceptions that we are "looking at". Scientists as such do not enter this debate; science is interested in what can be known about aspects of the physical world and the various physical causes of our various perceptions.

Of course physical scientists have often debated the existence of a particular thing that has been postulated to explain particular, observable phenomena. Caloric was a substance some scientists once theorized explained heat and heat transfer. The aether was at one time proposed as the medium through which light waves moved. These things were themselves not observable phenomena, but their postulation and logical nature allowed for experiments that could give observable (empirical) evidence that could confirm or disconfirm them. (Confirming is not the same as proving. That is explained in the accompanying paper, "Scientific Theory -- An Explanation".) The existence of both caloric and aether were disconfirmed by empirical tests along with certain logical arguments. But the debate had nothing to do with whether things in general exist as physical reality or only as ideas in one's mind. There are various questions and theories now about the nature of different forces (atomic, gravity, electro-magnetism, etc.) in physics but those are questions and theories involving only physical phenomena (both as the things to be explained and as the things to count as confirming or disconfirming a particular explanation given); there is nothing being dealt with in physics about the possibly non-physical origin or nature of observable phenomena in the way that metaphysicists or philosophers of mind think about it. What physicists do is consistent with both realists and idealists and with both a correspondence theory of truth and a consistency theory of truth. So to raise these issues in social science philosophy is a red herring, and it is an inaccurate perception of what natural scientists or philosophers of science are doing.

Now there is a separate question in physics about the logical nature of things like Newton's laws -- e.g., whether f=ma is just a definition of force or mass -- or exactly what sort of things his laws say about the world and what evidence, or kind of evidence, counts toward them, etc. Questions of meaning are always important. For example, physicists today claim the universe is expanding, but they do not explain what it is expanding into; and there is reason to believe from the way they write about it that what they mean is that the observable universe is expanding -- that all the matter and space that could ever theoretically technically be observed, by anyone within the known limits of the observable universe (i.e., on our side of the boundaries), is expanding.

There are logical and conceptual issues in social science also, such as the "nature" and/or various definition of intelligence, and whether some concepts of intelligence might be more fruitful than others, etc. Some notions of intelligence may be testable (as caloric or aether are) but some concepts may be more logical in nature, in the way the concept of mass or force is. But both quantitative research and qualitative research have to grapple with such issues; and neither orientation is logically connected with or committed to any kind of, or particular, definitions in given cases. One is simply looking for definitions or concepts that make sense, fit the known phenomena, are empirically testable, and that give a theoretical framework that will allow much further empirical progress in the field. If some concept of intelligence led to a theory that allowed all kinds of terrific progress to be made in education, that would be a very persuasive notion of intelligence to adopt until and unless observable phenomena were found or pointed out were not logically compatible with it.

It has been a long time since I read Kuhn, but I doubt that his notion of paradigms in science applies in the way Rist uses it -- to methods of research or discovery, such as the qualitative/quantitative debate. That book is about how models or theories that are well-supported and accepted are hard to get widely seen to be no longer tenable. E.g., the mechanical particle theory of light; where it was only after a vast amount of evidence accumulated, not just the various initial amounts of evidence, that the theory or model of light's being simply particles had to be modified or discarded. The problem that Kuhn points out to us is in part psychological -- scientists, being people, and people hating to give up well-entrenched beliefs, especially those that have seemed to explain much and that have predicted some astonishingly otherwise unsuspected things. But the problem also has a basis in logic, for false, improbable, or incredible conclusions are as much a sign of untrue premises or invalid logic as they are a sign that their contradictions are false. Any argument of the form:

P implies Q
P______
Q

where on other grounds you strongly believe Q to be false, is an argument that casts as much doubt on "P" or "P implies Q" as it does on "not-Q". I do not see that it is helpful or relevant to methodological debates at all to bring in Kuhn's language, concepts, or arguments regarding the resistance to change of entrenched paradigms. There may be some things Kuhn says about science that apply in some ways to what is supposed to be a quantitative vs qualitative debate of method of research, but I doubt the book's main concepts and arguments have anything to do with it.

In fact, I do not really see that there is a legitimate debate between qualitative research and quantitative research. Both kinds of research have uses and legitimacy. And I do not see how the idea of "understanding" or verstehen has more to do with one "side" or the other. Nor, similarly, does this subject vs. object notion much apply. Newton wanted to "understand" (what we now call) physics; but people, as bodies with mass, are subject to physical laws and principles just as inanimate objects or inanimate bodies are. Further, much medicine, as physical science, uses physical principles (pressures of moving fluids, solubilities of gases in liquids, osmosis, electromagnetic forces, and nuclear resonance etc., etc.) in understanding and dealing with people's health. Sure, social scientists are people who study people, but physicists and medical researchers using physical (and chemical, etc.) principles are also people who study things, some of which are people. Even apart from ethical concerns about manipulating people and exposing them to harmful treatment, people may be harder to study or understand than inanimate objects are for a number of reasons I will go into shortly, but that is not because they are somehow different kinds of things or require different kinds of research or a different logic. Social science is not a different kind of science from physics; it is science with a different subject matter (in the main) from physics, just as chemistry is not a different kind of science from physics but is simply science with a different subject matter (in the main). There are areas of overlap between chemistry and physics just as there are areas of overlap between people and chemistry or people and physics (where people or parts of their bodies are objects following chemical and physical laws).

One argument that is specious is the one that tries to show there is a difference between social science and physical science in that in somehow studying people or making predictions about them, we change their behavior or give them the opportunity to change their behavior in ways we do not about inanimate objects. In physics on the subatomic level, there is at present serious and accepted argument that the means of observation necessarily and always disturb and change the phenomena being observed. Now, of course, it is true that there are some ways of studying people that cause them to act differently from how they would if they did not think they were being studied or if there environment had not been relevantly disturbed somehow by the experimenter. And it is true that people might change their behavior if they are informed about how they behave. But that is only to say that different stimuli or different forces acting on a person or an object may give different effects from conditions without those forces being present. Surely there is research on people that does not disturb their environment or let them know they are being studied. And surely we can make accurate scientific predictions about people without having to tell them our predictions. We can write them down in a sealed envelope or tell them to others or put them on a recording, any of which can be shown later to verify they were made previous to the act predicted. If I predict to other parents that a child will fall asleep around 8:00 at their house without telling the child, who may be perverse, that prediction can be accurate. If telling the perverse child the prediction makes it successfully will to stay up later, or fall asleep earlier, and thus alters the accuracy of that statement, that in no way undermines the accuracy of the prediction that would be made without telling the child. 

There is nothing about the nature of prediction that requires us to share our prediction with its subject ahead of time. And there is nothing in social science that makes only such shared predictions real predictions. "Secret" or unshared predictions are still scientifically valid predictions. Changing the "prior" conditions or stimuli or forces acting on someone by predicting his behavior to him is a different thing from making predictions without telling the person (subject) involved in your predictions. The latter is as much social science as, if not more than, the former. Doing the former means changing the conditions so that now you have to experiment all over again, unless your original experiment was about studying how people reacted to such predictions. Possibly in some cases we can even predict how telling our predictions will affect people's behavior. Information can be a stimulus to people's actions; it is not a stimulus to the actions of planets. The fact that different stimuli affect different objects in different ways, if at all, does not make a relevant difference in the logic of studying the different objects, just in the kinds of variables (or forces) one has to isolate against. In physics, temperature and pressure affect volume of a given gas, but the amount of light in the room or the kind of container or the time of day, etc. do not. But that does not mean that a researcher in physics who first tests for volume changes in a gas under different light, time, or container conditions is not doing science in the same way that Boyle or Charles did. It just turns out that one does not need to worry about amount of light or type of container -- or what one says when near the gas -- when studying gases. If telling someone something influences their actions, especially in a contradictory way, it would be absurd to insist that it is somehow not social science, or not really testing predictions in social science, to make predictions if you do not tell the person/subject what those predictions are before he does or does not do the action in question. The question is whether accurate predictions can be made under a certain set of relevant, specified conditions, not under those conditions plus the added condition of "telling" the subject. That changes the conditions of the prediction, just as it would to tell a chemist that he must predict the volume of a certain gas at a certain pressure, but that each time he makes the prediction, the pressure (or some other relevant variable) will be changed before the test is run.

I suspect one of the two main present differences between social science and much physical science (i.e., primarily chemistry and physics, as opposed to things like biology) is that in physical science in many areas we understand what many of the relevant forces or influences are (and how they work) that influence or control objects, but in the social sciences, as in much medical research, we still have not always been able to isolate or determine the relevant influencing factors or determine how they operate. It seems to be easier to control relevant variables in experimenting on the physics of pool balls than it is to control the relevant variables in dealing with people learning things, because more factors seem to influence people's learning (e.g., what they already know, what "triggers" them to apply that knowledge in a particular situation, how awake they are, how interested they are, how they understand what you are asking them, whether they are paying attention to you, how much, etc.) than influence the motion of pool balls. Mental behavior seems to be more complex and/or have more relevant or unknown variables than does the physical behavior of simple objects, but that does not necessarily make the study of the former different in logic or in kind from the study of the latter, it just makes it more complicated. Physics at present perhaps faces the same difficulty in regard to cold fusion, if there is such a thing, and to superconductivity.

[I deal with the second main current difference between physical science and social science in the accompanying paper, "Scientific Theory -- An Explanation". That aspect is essentially that it seems difficult to imagine predicting some kind of astonishingly new, previously unknown or unsuspected kind of behavior, under some set of circumstances, that would confirm a theory of behavior to any convincing degree. What would be needed would be something in the social science arena as astonishing and previously unsuspected as the atom bomb, holograms, computer chips, fiber optic information transmission, etc. When all behavior is known, science cannot make predictions; theories become "explanatory" only, not predictive, and not then confirmable.]

That finally leads me into the supposed qualitative-quantitative research debate. First, some observations. 1) I take medical research (apart from psychology) to be a physical science, and much of medical research is statistical in nature. E.g., people who smoke are at some percentage of higher risk to get certain diseases than those who do not smoke; people who eat certain foods are more likely to get certain conditions than.... But medicine is yet to pin down the exact factors, conditions, etc. that cause or preclude various ailments. 2) Two of Newton's laws are not quantitative at all, and there is serious question about the nature of the one that seems to be quantitative. 3) Einstein's conceptual, but non-mathematical, realization about the nature of "simultaneity" and how we determine whether two events are simultaneous (and/or how much time elapses between them) is central to his ideas and discovery of relativity. 4) Galileo's explanation and laws about falling bodies are not based on observations and measurements of actual falling bodies at all, and they would not even come close to describing how bodies actually fall on earth (in our atmosphere). 5) Archimedes' discoveries are conceptual and logical (and partly mathematical) in nature, more than just empirically arrived at through observation. Many, but not all, scientific laws are mathematical in nature. For even the mathematical ones, understanding their significance or "why" they work in some intuitive way sometimes (perhaps even always) takes one outside of mathematics.

The point is that scientific discovery and understanding is where you find it and can make a logical and conceptual case for it in the light of known or discoverable empirical phenomena, whether that case includes mathematics or not. There is no known particular tool or technique or line of reasoning or train of thought or algorithm or formula for automatically making scientific discoveries or advances. And trying to narrow logic and creative understanding to some particular, and therefore necessarily narrow, methodology would seem to hamper, rather than promote discovery and understanding. And further, if some particular methodology were established for validity or justifiability, it would be ridiculously impossible then to disprove a theory that the methodology confirmed, but which some other perfectly logical considerations showed to be misguided. Trying to define or limit scientific understanding to a particular methodology limits the ability to discover, to confirm, and to disconfirm. Scientific discoveries have come from luck, from observations under the right conditions, from logic, from conceptual insight, from mathematical deductions, etc. by individuals or by groups of people working together or one after the other through time. Even the search for causally relevant influencing factors, and how they operate, can be difficult. I think there was something like a hundred years between the discovery of Charles' Law and Boyle's Law, one showing that pressure inversely effected the volume of gas and the other showing that temperature directly effected it, laws that seem to us today like they should have been easy to figure out. The notion that matter could be converted to energy was one thing; the notion, and its acceptance, that energy could be converted into matter was something else. If dealing with statistics helps one identify interesting phenomena or find causal factors, fine. If qualitative methods help that, fine. But if Galileo had dropped a whole bunch of different objects -- feathers, birds, paper, cannonballs, coins, etc. -- from various cliffs and been able to statistically analyze the times of their falling, he may never have discovered or conceptualized his laws of motion. That does not mean statistical methods are never fruitful in any kind of science.

Now, I suspect that often when one gets a mathematical explanation, someone tries to also get a more intuitive or non-mathematical understanding of that formula. Sometimes that is possible, but, sometimes perhaps it is not. This is true in physics and I think it is true in social science as well. Richard Feynman worked on the problem of why the wobble on a Cornell University cafteria plate, spun up in the air, went around at a different speed from the Cornell emblem on the plate. He said that after he worked out the mathematics of it (which he later applied to the spin of electrons and won a Nobel prize for his discovery), he tried to understand in more intuitive terms how those mathematics worked to give the spinning plate the appearance it had. But in some high level physics, all that people have (so far) are equations they have worked out. There is no physical model or intuitive example that the mathematics lends itself to (yet). And many physicists don't even seek physical or intuitive models. They are content with mathematical ones. Math and intuition sometimes go together, sometimes not. There is the trick problem of how fast a race car must travel the second mile of a two-mile qualifier in order to average 60 miles per hour if it has done the first mile at 30 miles per hour. (Although it is difficult to believe, no speed will work, since the first mile took the whole two minutes that you need to do both miles together in order to average 60 miles per hour for each.) Working in math alone can be deceiving. One trick question I misinterpreted one time cost me a great deal wasted work. It asked which train will be closer to New York when a train leaving New York for Boston (at one time and speed) passes a train leaving Boston for New York (at another time and speed). (I worked that out for two hours mathematically only to find out that they were both the same distance from New York when they passed each other -- only to realize that of course they would have to be!) Watch Mr. Wizard or go to any instructional science museum and you will see things you never realized and couldn't guess, that you have to look at in a different way to understand. Some of these things are simple once you see them, but seeing them is often difficult. And some things are not simple to "see" even when you "know" the principles behind them. 

Of course logic is not infallible; paradoxes arise that show us that. Zeno was a Greek philosopher of long ago who gave the following ones:

1) If Achilles and a turtle run a race and Achilles gives the turtle any kind of headstart at all, the turtle will win the race because Achilles -- no matter how fast he is, or how slow the turtle is -- cannot catch the turtle. Because when they both start running it will take Achilles some amount of time to catch up to where the turtle started from. When he gets there, the turtle will have moved ahead. It will take Achilles some time then to get to where the turtle is when Achilles has gotten to his starting place, but the turtle will have moved ahead from that spot by then. In fact, each time Achilles catches up to where the turtle was at the previous moment, the turtle will always have moved ahead from that spot. Hence, Achilles can never catch the turtle, let alone pass it to win the race.

2) You can never leave a room because in order to do so you must first walk half way to the exit from where you are, but to get to that place you must first go half way, etc. There are an infinite number of such halfway places and you cannot go to an infinite number of places in your lifetime, so you cannot exit the room.

Obviously such reasoning(s) is (are) spurious; but it is not easy to see where they go wrong. Logic involved in science can be like this too except that you do not know what the answer is supposed to be so you do not know whether you have even gone wrong or not, let alone how. In Aristotle's day, for example, one question seemed to be what force kept a spear in flight once it left the force of the arm throwing it; since nothing was still pushing it, what made it keep going. Rest was assumed to be the natural state without a force's acting. Newton (or someone) asked instead what force made the arrow stop. It has turned out, it seems, to be more productive to assume that bodies in motion or at rest have an inertia to remain that way unless acted upon by another force, than it is to assume that things only (come to) rest unless acted upon by a force. But neither assumption is exactly experimental and each has attendant theoretical problems and solutions. Newtonian mechanics makes us postulate all kinds of "invisible" forces such as gravity, and makes us try to understand then what exactly gravity could be and how it could work.

Now there is a real question, once you have worked out the mathematics, or recognized the correlation, of phenomena, whether there is "something" (again, something physical and "real", not ideal or mental) behind that correlation which causes it -- particularly if it turns out that "something" is not itself directly observable. For example, first of a discoverable, observable cause: one of the planets, Uranus I believe, was sighted because astronomers looked where mathematical equations showed some planet would have to be to cause the discrepancies in another planet's observed orbit. But, electrons, for example, are not something that can be observed; their existence is predicated on other observable phenomena, from seeming "clues" their interacting behavior "leaves". It is a philosophical question whether there is a "thing" called an electron or whether it is just a purely heuristic conceptual model that helps us tie together and/or somehow discover other observable phenomena even though the electron itself may just be a figment of our imagination. Positivists would argue that there is no reason to posit the thing, since we can never observe it. We only have observable phenomena itself -- that phenomena may not be evidence for anything. If we know something is a footprint, we know it is evidence of an animal; but sometimes the question is whether something is a footprint or just an accidental marking in the sand. (Or something else altogether. I am told some pranksters one time cut out cardboard in the shape of huge feet and then used them to make giant green "footprints" between two Detroit well-known statues -- a large green male statue, and a large nude female statue. The footprints appeared one morning, signaling a tryst, or perhaps statutory rape.) If you are not sure something is accurately to be described as a footprint, it is not proof there was an animal. Positivists would argue that the evidence we have of theoretically unobservable phenomena (such as forces or electrons) is not "evidence" for anything at all (not a footprint at all). It is just phenomena. Whether positivists are right or not about "in principle" unobservable phenomena, however, is a philosophical issue irrelevant to the kinds of pursuits of scientists, as scientists. Some scientists find it useful to look for, or think of, unobservable but physical things (hence, we are not still in the realist/idealist debate) causing the phenomena or (mathematical) correlations they observe. Others do not; they are just content with working out logical or mathematical predictions for further observable phenomena, regardless of whether any "thing" causes that phenomena. But neither viewpoint is germane to what kinds of research one might do. Two scientists, both qualitative or both quantitative researchers, might work side by side, do the same kinds of mathematical or logical work and come to the same conclusions with only one of them having some sort of "picture" in his head of an invisible physical thing that causes the phenomena he derives. The other may not need or get much use out of such a model or "picture" or theoretical "explanation".

Smith also seems to think there is some difference or dichotomy about the notion of values relevant to whether one is a qualitative or quantitative researcher (or whether one is doing qualitative or quantitative research at the time -- assuming one person could at different times do different kinds of research). He summarizes some interesting distinctions and viewpoints but they hardly seem to me to be ones that fall along the lines of qualitative or quantitative research. No intellectually honest person looks merely for social agreement from his peers; one wants social agreement that results from proving his case to his peers, not social agreement because he deceives them all. Agreement is supposed to result from everyone recognizing the truth; truth does not result from everyone having agreement. And all scientists as such want to be objective and want their evidence and logic to cause the agreement because others see it is objective. Bias is irrelevant, even when it exists, to objectivity. One can be both biased and objective. One can wish to prove something will come out a certain way, and actually prove it -- or be unhappy one has disproved it.

Interpretation versus correlation. The point is to arrive at an understanding of phenomena -- whether that phenomena is statistical in nature or whether it is non-statistical in nature. The explanation for why a coin lands about 50% of the time as heads may be just mathematical in nature -- it has two sides, each side is equally probable, etc. If it always lands heads, or lands heads 75% of the time over various trials one may look for a different explanation (each side is heads, it is weighted a certain way, the toss is fixed, there is a rigged magnet, or some such). Whether Newtonian explanations are conceived of mathematically or interpretively, they must still explain why feathers fall at different rates from cannonballs in the air and why they fall at the same rate in a vacuum. Regardless of who finds out phenomena, whether it is qualitative or quantitative, probable or exact, a theory meant to explain a subject must explain or give a logical account of relevant known phenomena, and predict phenomena not known at the time. Had Galileo done statistical studies of the "hang time" of different objects dropped off cliffs, any theory of falling bodies would have to explain why those hang times were different. Any theory meant in some way to completely explain the phenomena on a topic will have to explain both the accurate quantitative and qualitative discoveries in the field.

Statistical studies can be a help in finding out what factors to try to study in isolated ways, or what factors to try to think about logically as to how they might contribute to understanding the problem. But statistical studies are not the only method for discovering key factors. Galileo thought about what factors were important in studying free falling bodies and did experiments to get his laws without ever experimenting with actual free-falling bodies. (He used bodies rolling down an inclined plane.) He arrived at what the important factors to think about were without using statistics. But some factors in science would never have been discovered without statistical analysis. In the social sciences, it is hard to get pure experiments or totally controlled conditions for many important questions. Statistical methods may contribute by showing that though all factors cannot be controlled, they still can be accounted for. 

But the point of science, physical or social, is to find and demonstrate what the causally relevant factors (conditions) are for given phenomena or range of phenomena; and it does not matter whether one uses qualitative means or quantitative means for either the discovery or the demonstration. It is the relevant logic of the evidence, mathematical or not, statistical or not, that is important.

Concerning Rist's "Reliability vs. Validity". All reliable data (whether quantitative or qualitative) are to be explained by a theory that is meant to explain a topic in full. And all reliable data, it seems to me, whether qualitative or quantitative, is valid. It is just that not all interpretations or data, whether qualitative or quantitative, are important or show main (significant or primary) causal factors. Aristotle and Newton both looked at the spear in the air qualitatively, but Newton's particular approach was more fruitful (till the present at least) and therefore (till the present at least) more important. One particular statistic correlation may be more fruitful to pursue than another. There is nothing about either method that guarantees someone will discover factors important or central to understanding the particular issue; and nothing in either method prevents anyone from doing so. Further, both kinds of researchers understand the hunt is for the important factors that explain the interesting phenomena. Quantitative researchers often tend to look for correlations in things they think will bear fruit because of some qualitative idea they have; they do not just go in and measure everything or any old thing that could be measured. And qualitative "explanations" cannot just ignore statistically significant correlations; and usually do try to account for them. One cannot just slough off an observed phenomenon, statistical or interpretive, as an anomaly unless one can explain (how) the anomaly (arose). It would do a physicist no good to say a feather falls as fast as a cannonball, unless he can show why under normal conditions that is not true. He has to demonstrate that air pressure is not a relative factor under "ideal" conditions and explain how it impinges on, and works against, those ideal conditions.

Rist's "Subjective vs. Objective". I don't understand what he is trying to get at here at all. He seems to think Scriven makes an important point distinguishing subjective from objective in two different senses, but then thinks those distinctions don't help because qualitative and quantitative research still disagree about what provides subjective confirmation and what provides objective confirmation. Rather than try to decipher what he is getting at, let me try a different approach -- one similar to Scriven's, but not making Scriven's mistake of equating one kind of objectivity with inter-subjectivity or of equating the other kind of objectivity with confirmability.

Inter-subjectivity is not objectivity. If everyone were overnight to develop the same favorite color or favorite flavor, we might say flavor is inter-subjective, but we would still not say it is objective. Subjective-objective dichotomy in one sense simply means a matter of taste versus something not depending on a subject's perceptions or phenomenology. That an object appears blue is subjective (inter-subjective as long as you do not have color blind people), but that it absorbs and reflects certain wave-lengths of light is objective, regardless of how it might look to different people. The normal human eye, for example, can not see light "temperature" differences in certain ranges, but instruments and different kinds of film can. On "outdoor" film, flourescent light might look purple, candle light or incandescent light might look very yellow. The color is "subject" to the kind of film used. But the wavelength of the light and the emulsion tendencies of the film are objective. And one can be objective in knowing how the pictures will turn out, knowing what film is being used in what kind of light. Arguments often have to be made about what things are subjective and what things are objective in this sense (e.g., I would argue that ethics is objective even though people disagree about ethical principles and values -- the argument has a number of points so I will not give them here; suffice it to that what things are a matter of taste or perspective and what things are not is often not obvious and is sometimes very difficult to know, and takes study and analysis and argument itself).

Subjective-objective in another sense means rational versus irrational. Or it means a guess versus an accurate measurement or assessment. For example, if someone just guesses how many beans are in a jar, he gives a subjective answer (even if he is right); but if someone weighs the contents and divides by the average weight of a bean, he is being objective (even if he is wrong).

Subjective-objective can also mean uses the proper methodology for solving a problem versus using improper methodology, some of which may be irrational or irrelevant. One might have an objective (in the first sense above -- taste independent) method for estimating the number of beans in a jar -- say, one pulls successive books off a shelf and multiplies the number of pages by eight to arrive at the number of beans in the jar. This is obviously an improper method though it does not depend on the estimaters personal tastes nor on his guessing. What is subjective or improper here is the estimator's choice of methodology -- and believing this method reliable or relevant.

Regarding research, my argument would be that one wants methods used to be objective in at least all three of the above senses: taste independent, rational and not just a guess, methodologically reasonable or relevant. This does not mean you cannot study tastes; obviously you can study tastes objectively, e.g., by seeing what flavor's people actually choose in Baskin-Robbins, etc. I think both qualitative and quantitative researchers would agree about my goals (or some similar ones, in case I am wrong about the particulars) for objectivity; what they often disagree about are what methods are in fact proper ones and what methods are in fact devoid of taste or mere perspective. But that is where rational argument and/or experiments testing the method to study known cases to see if you get the right results in those cases, etc. comes in. There may be some qualitative methods that are very accurate and rational and taste- or perspective- independent; there may be some quantitative methods that are not rational, even though quantitative, or that are very inaccurate. (Using a Geiger counter to measure amounts of radioactive material may be very taste independent, but it is not accurate nor rational to use it to measure the radioactivity of materials in lead vials.) Quantitative research is not necessarily objective and qualitative research is not necessarily subjective, though each can be, in the senses I have described. I believe these senses are normal senses of such terms.

Concerning component vs. holistic analysis. Either sort of analysis can be done either qualitatively or quantitatively. Neither is the particular province of either method of research. Generally one tries to do both as far as is possible. One might try to do experiments under ideal (perfectly controlled) conditions and then try to account for imperfect conditions (in physics: friction, air pressure, wind conditions, etc.) and do the experiments under those conditions and show that one can still accurately predict the results.

Finally, I don't think that the social sciences, from what I have read, give enough heed to conceptual matters. Many of the greatest advances in physics have been primarily conceptual discoveries rather than qualitative or quantitative experimental ones, though experiments and observations were prior and critical to the conceptual discoveries. Further, many social science concepts (such as intelligence) are conceptually confused, and so no matter how "objective" or unbiased or quantitatively or qualitatively accurate the research, its meaning and/or its correlations will be confused and unclear.

Rick Garlikov

This work is available here free, so that those who cannot afford it can still have access to it, and so that no one has to pay before they read something that might not be what they really are seeking.  But if you find it meaningful and helpful and would like to contribute whatever easily affordable amount you feel it is worth, please do do.  I will appreciate it. The button to the right will take you to PayPal where you can make any size donation (of 25 cents or more) you wish, using either your PayPal account or a credit card without a PayPal account.