This page gives a recap of the issues we discussed in the second class of the module. Here we examine the differences between quantitative and qualitative research and explore further the nature of science and its underlying philosophical assumptions. Please note that we are only scratching the surface of these issues here! These are things that philosophers of science have been debating for centuries, and will no doubt continue to do so for centuries to come. It is sufficient for our purpose to just have a feel for them.

Quantitative versus qualitative research

Most sport, health and exercise science research, like research in the human, behavioural and social sciences in general, adopts the methods developed over centuries in the natural sciences. There is, however, much debate about the appropriateness of this approach when applied to the behavioural and social domain. This debate is often cast in terms of a distinction between quantitative and qualitative research.

Quantitative researchers broadly embrace the assumptions and methods of the natural sciences. They quantify the variables of interest (put numbers on them) and examine the relationships between the variables mathematically (using statistics). Quantitative research generally adopts a deductive approach to its reasoning: hypotheses are developed based on theory and then data are collected to test the hypotheses. So it moves from theory to the data. Quantitative research is generally nomothetic: its focus is on generalisation and working towards the development of universal statements or laws. Hence the data collected tend to be aggregated across individuals (e.g. group means are compared).

Qualitative researchers, on the other hand, reject the idea that the assumptions and methods of the natural sciences can be applied to the behavioural and social sciences. They argue that human beings are each unique individuals and that the complexities of human social interactions cannot be reduced to mere numbers. Instead they aim to uncover and report the quality of the social world in which we live. Qualitative research usually adopts an inductive approach to its reasoning: observations are made (data are collected) and then it seeks to work towards a theoretical integration of what it has found. So it moves from the data to a theory. Its approach is generally idiographic: the focus is on individuality and uniqueness. Hence the data collected in qualitative research tends to be oriented toward individuals and case studies.

The debate between advocates of the quantitative approach and those who favour qualitative research has often become heated and acrimonious, almost to the point of each side being reduced to mere name calling. Some qualitative researchers refer to quantitative researchers as 'quantoids', suggesting a mechanical, robotic approach to research. How about 'qualiflowers' as a name for qualitative researchers? Certainly both sides have often engaged in promoting unfair and unrealistic caricatures of the other side.

For a much more reasoned debate on these issues in a sport and exercise science context, it is worth reading the trilogy of papers by Lawrence Locke, Robert Schutz and Daryl Seidentop on qualitative research published in Research Quarterly for Exercise and Sport in 1989 (click here for the references). They are avaliable from the library. This edition of RQES was something of a landmark in the sport and exercise sciences and, along with the Scanlan et al. studies on sources of stress and enjoyment among elite figure skaters which we will meet in the next class, did much to lead to greater acceptance of qualitative research in our area.

Different methods, different conclusions?

Underlying the divergence of opinion across the two sides of the debate are some very real differences in beliefs about how we come to understand the world we live in. Sometimes these different methodological approaches lead to markedly different conclusions. Here is a good example in the context of two papers examining the psychological impact of imprisonment.

James Bonta and Paul Gendreau (1990), in a paper entitled 'Re-examining the cruel and unusual punishment of prison life', reviewed the quantitative research literature on the effects of imprisonment in order to find out whether it is psychologically damaging to individuals. They set out to determine the objective reality of imprisonment by examining studies that used standardised measures of the variables of interest and employed statistical tests of the relationships among the relevant variables. Such measures included psychophysiological assessments of the effects of overcrowding, psychometric inventories and institutional records of behavioural misconduct. Their findings showed that high population density was not always associated with adverse changes in physiological indicators or self-reports of stress; that there were no differences in scores from standardised psychological tests among inmates who had been incarcerated for different lengths of time; and that experimental research on sensory deprivation has demonstrated that solitary confinement is not inherently harmful to prison inmates. Moreover, the authors found that imprisonment may actually have positive effects by isolating inmates from a highly risky lifestyle in the community.

Bonta and Gendreau concluded that, contrary to popular opinion: "The facts are that long-term imprisonment and specific conditions of confinement such as solitary, under limiting and humane conditions, fail to show any sort of profound detrimental effects". They went on to say that: "many prisons may actually be conducive to good health". Bonta and Gendreau criticised much of the previous work in this area, which has reached contrary conclusions: "When it comes to scholarly inquiry in the field of criminal justice, a pernicious tendency has been to invoke rhetoric over reality and affirm ideology over respect for empirical evidence ... if we are to make progress in understanding what our prisons do to inmates, then we must respect the available evidence".

Bonta and Gendreau's article was quickly followed by a critical response from Julian Roberts and Michael Jackson (no, not that Michael Jackson) in a paper called 'Boats against the current: A note on the effects of imprisonment' (1991). Roberts and Jackson criticised Bonta and Gendreau's paper on the grounds that the measures used in the studies they reviewed could not capture the subjective reality of imprisonment: the meaning that the experience has for inmates. Using evidence from interviews, anecdote, life histories and sworn testimonies from ex-prisoners, as well as what they describe as "centuries of human experience" the authors concluded that imprisonment does indeed have serious detrimental effects on the physical, psychological and social well-being of prisoners.

Roberts and Jackson argued that the experience of imprisonment can never be understood in terms of objective, quantifiable measures of physiological and psychological outcomes: "Can the effects of living on death row be captured by psychiatric interviews or the MMPI [a standardised psychometric questionnaire]? We would argue that [Bonta & Gendreau's] approach substitutes a spurious objectivity for the human dimension of punishment as it is experienced by prisoners". They went on to state that the evidence cited by Bonta & Gendreau was: "in effect, quite irrelevant to the real-life experience of prisoners".

So, two different approaches lead to two quite different conclusions. It is interesting that less radical and zealous qualitative researchers often cite one of the advantages of their methods as being that they can be used to triangulate findings obtained from quantitative studies. That is, they can be used to find support for previous findings by looking at the phenomenon from a different perspective. Usually this is the case; qualitative findings often corroborate quantitative ones. Clearly it did not happen here!

You can see that the quantitative approach advocated by Bonta and Gendreau was founded upon a belief that a dispassionate and objective stance with regards to the data was necessary in order to understand the phenomenon of interest. Roberts and Jackson, in contrast, emphasised a subjective approach that aimed at 'getting inside the prisoners' heads to understand the meaning of the experience for those individuals. This is the crux of the distinction between quantitative and qualitative research. Let's now examine the underlying assumptions of the two approaches in a bit more detail.

Fundamental axioms of natural science

The 'traditional' approach to science is based upon a set of fundamental assumptions, or axioms, about the nature of the universe and how we can come to know and understand it.

Positivism

The most fundamental of the axioms is positivism. Positivism is the view that the universe consists of real, independent phenomena that we can come to know through direct observation. Isaac Newton neatly expressed the doctrine of positivism:

Absolute space, in its own nature, without regard to anything external, remains always similar and immovable. Absolute time, true and mathematical time, of itself and by its own nature, flows uniformly, without regard to anything external.

Sir Isaac Newton

 

So positivism holds that there is a reality 'out there' that exists independently of any observer (without regard to anything external, as Newton put it). It is like the old conundrum about a tree falling down in a forest: if there is no-one there to hear it, does it make a noise as it hits the ground. A positivist would certainly answer 'yes' to this question.

Empiricism

Furthermore, in its purest forms, positivism holds that we can only come to know and understand events that are directly observable through our senses. This is the principle of empiricism. If this were really the case, then it would clearly limit the phenomena that we can study 'scientifically'. For example, it would rule out of court most of the subject matter of modern psychology, which is largely built upon the study of constructs that are not directly observable, such as attitudes, beliefs and consciousness. Indeed, at one time the field of psychology was largely dominated by a school of thought called logical positivism that held that the only scientifically justifiable subject matter for psychology was observable behaviour. Anything metaphysical (events that are not directly observable), such as the mind, attitudes and beliefs, was ruled out of bounds for the discipline.

Recall, however, that in the class we saw how even modern physics no longer holds completely to the doctrine of positivism. Remember how Einstein's theory maintains that time does not flow in an orderly fashion but that the sequencing of events in time depends upon the velocity with which the observer is moving with respect to those events. Consider how far removed that idea is from Newton's view of time in the above quote.

Postpositivism

Nowadays, most scientists adopt a postpositivist stance. Postpositivists accept that events that are not directly observable can be a legitimate subject matter for science if we can infer their existence or behaviour from events that are directly observable. No exercise physiologist has ever seen muscle soreness; no psychologist has ever seen an attitude; and no particle physicist has ever seen a quark. But a physiologist can infer that someone has muscle soreness from their response to a soreness rating scale; a psychologist can draw inferences about someone's attitudes from their scores on an attitude questionnaire; and a physicist can make inferences about the behaviour of a quark from the pattern it leaves on a photographic plate.

Determinism

Another axiom of traditional science is determinism. Determinism holds that everything that happens has a specific cause and gives rise to a specific effect. It follows that, in principle, if we can identify causes and their effects we can predict the future of any part of the universe with absolute certainty. Consequently, understanding the causes of events is at the very least one of the principle aims of science. If we can understand what causes an event, we can intervene to ensure that it happens or to prevent it happening. For example, if we know what causes athletes to be able to control their anxiety in high pressure situations, we can implement a programme to ensure that they do control their anxiety. Conversely, if we know what causes muscles to waste in a degenerative disease like rheumatoid arthritis we can intervene to stop it or even reverse it. At the end of the day, isn't this the sort of thing that science is for? Remember Lewin's adage that there is nothing so practical as a good theory. We will return to the issue of causation in greater detail in a later lesson.

Again though, postpositivists are rather less rigid than traditional positivists with respect to determinism. Recall Einstein's words on the certainty of mathematics:

As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality.

Albert Einstein

 

Postpositivists would accept that we may not be able to know things with absolute certainty. Like most things, this is not a new idea. The Greek philosopher Xenophanes, in the 6th century B.C. said:

The gods did not reveal, from the beginning, all things to us; but in the course of time, through seeking, men find that which is better. But as for certain truth, no man has known it, nor will he know it; neither of the gods, nor yet of all things of which I speak. And even if by chance he were to utter the final truth, he would himself not know it; for all is but a woven web of guesses.

Xenophanes, 6th century B.C.

We can do a bit better than just relying on sheer guesswork, though. Postpositivists maintain that knowledge is probabilistic: we can predict with some degree of certainty and we can specify the degree of certainty using the laws of probability and inferential statistics. So we have made some advances in the last few thousand years!

Materialism

Yet another fundamental axiom is materialism. Materialism is the idea that whatever exists must exist in some amount. Therefore, in principle at least, it is possible to measure whatever exists. In other words we can quantify the events that we study scientifically. This, of course, is why the application of the traditional scientific method is often called quantitative research.

 

Operating principles of science

These, then, are the principle axioms of science. These fundamental assumptions lead to a set of operating principles about how science should be done.

Objectivity

The first of these is the principle of objectivity. The scientist must remain dispassionate and detached in making observations and interpreting data. In other words, the scientist should be isolated from the thing they are studying so that they do not influence or contaminate it in any way.

Qualitative researchers generally argue that it is not possible to remain detached in this way. As thinking, feeling human beings, researchers inevitably bring their own biases and perspectives to a research situation and will interpret events according to their existing beliefs and expectations. Furthermore, qualitative researchers argue that it is only by immersing oneself in the phenomenon being studied that one can truly get to understand it. Instead of trying in vain to remain detached, the researcher should become fully involved in the social reality of the thing they are studying. Then they should explicitly describe their background and perspective in any research report so that the reader can take that into account. The onus, then, is on the reader or consumer of research to accept or reject the authors' account.

Quantitative researchers counter that by not remaining objective and detached, the researcher can become selective in what they observe and note, ignoring any data that do not fit with their preexisting ideas. Seidentop (1989), in one of the RQES trilogy papers pointed to this problem: "Neo-Marxist qualitative researchers declare the truth of neo-Marxism, making judgments that, in fact, help to ensure that neo-Marxist truth emerges from their data." The implication is that the researcher might only observe and report those things that fit with their preexisting agenda.

Recall, however, how even in the realm of physics, quantum mechanics has shown that the observer is an integral part of the event being observed. The classic example is that light behaves in different ways depending on how it is observed. Fritjof Capra noted this:

In modern physics, the universe is ... experienced as a dynamic, inseparable whole which always includes the observer in an essential way. In this experience, the traditional concepts of space and time, of isolated objects, and of cause and effect, lose their meaning.

Capra (1983)

Most researchers, both quantitative and qualitative, now take a more pragmatic, postpositivist stance with respect to objectivity. Quantitative researchers accept that it is not always possible to be entirely objective but seek to be as objective as they can. Similarly, most qualitative researchers accept the dangers of over-identifying with their participants and seek to minimise bias and data distortion. For example, as we shall see in the next lesson, it is common practice in analysing qualitative data to have different researchers independently check the interpretation of the data.

Replicability

A second operating principle is that of replicability. If the universe is organised in an orderly and consistent fashion, as positivism holds, then it follows that scientific observations should be repeatable. Replication is generally held to be vital to science. We should not accept that the results of a single study are valid. Instead, we need to see that the results can be replicated by different researchers in different places and different conditions. The more that a finding is replicated, the more faith we can have in it. You may recall the great fuss that was made in the media a few years ago when a research team announced to the world that they had discovered 'cold fusion': a way to generate nuclear energy without the high temperatures normally needed. This offered the prospect of cheap, easily available and safe energy for the world. Immediately researchers all over the world attempted to replicate the study, with no success. It was concluded that the original study was contaminated in some way. Imagine the time, resources and human energy that could have been wasted in building cold fusion plants if everyone had accepted the initial report as valid before attempts were made to replicate it.

Parsimony

A third operating principle is that of parsimony. This is that all data should be interpreted in the simplest, most succinct form. We should avoid unnecessarily complicated explanations. Parsimony is often referred to as Occam's Razor, after the 14th century philosopher and Franciscan friar William of Occam, who is regarded as being the first person to formally express the principle. He was using it to justify his theological arguments, which did not go down well with the Pope, but that's another story. Later it was adopted by scientists to mean that if we have two or more competing theories that explain the data equally well, the simplest one is the best. Isaac Newton expressed it like this: "We are to admit no more causes of natural things than such as are both true and sufficient to explain their appearances."

To understand the principle of parsimony, recall our discussion of falsifiability in the first lesson. There I showed how two competing theories could fit a data set equally well:

 

As I said then, the 'squiggly line' theory is much more complicated and less elegant than the 'straight line' theory. The principle of parsimony would dictate that I choose the simpler explanation. The more complex one just carries unnecessary baggage.

 

Fundamental principles of qualitative research

If the quantitative approach can be summed up by the proposition that reality is 'out there' waiting to be captured by objective observation, qualitative research can be described by the idea that reality is 'in my head'. Others only have access to my interpretation of reality to the extent that they share my view of the world. A qualitative researcher's findings are a personal construction of how they view events and their job as a researcher is to persuade others that their interpretation is valid.

 

This view stems from two fundamental principles: phenomenologism and constructionism.

Phenomenologism

Phenomenologism holds that a person's perceptions define reality. Since different people will perceive events in different ways, it follows that in order to understand a personís reality we have to look at the world through their eyes. Qualitative researchers seek what the sociologist Max Weber called verstehen (the German for to understand): a deep understanding of a person's behaviour or situation in terms of the personal (phenomenological) meaning he or she attaches to it.

Constructionism

Constructionism (also called constructivism) is the view that there is no objective reality 'out there', at least in the social and psychological world. Instead, people actively construct their reality on the basis of their beliefs and expectations. To a large degree, different people share the same reality, particularly within the same culture, because our beliefs and expectations are largely culturally transmitted. Clearly if we all had totally separate realities, society and all interpersonal interaction would completely break down as we would be unable to communicate with one another effectively (recall from the class that we all agreed that the whiteboard was a whiteboard and not an elephant).

Nevertheless, at some points our views of reality diverge. Think of those two different interpretations of the effects of imprisonment. There are actually two versions of constructionism. The strong version is that there simply is no 'ultimate truth'. Instead the truth is however one constructs it. The weaker version is that although there could be an ultimate truth out there somewhere, we might never come to know it. So in the meantime the truth is how we construct it. Moreover, our understanding of the truth will change over time as we learn to look at the world in different ways. Here is George Kelly, whom we met in the last lesson, on constructionism, clearly showing that he embraced the weaker version:

Whatsoever nature may be, or howsoever the quest for truth will turn out in the end, the events we face today are subject to as great a variety of constructions as our wits will enable us to contrive. All our present perceptions are open to question and reconsideration ... even the most obvious occurrences of everyday life might appear utterly transformed if we were to construe them differently.

Kelly (1970)

It should be noted that Kelly went on to say that just because reality can be construed in different ways, it does not mean that one person's reality is as good as any other's. Carried to extremes the strong version of constructionism leaves us with a real difficulty. If each person's construction of reality was equally valid, we'd be left floundering in a world in which there were no absolutes. Faced with a real life problem, then, what would we do? Whose version of reality would help us to solve the problem? Locke, again in the RQES trilogy, put it like this:

If there is no way to describe the nature of the social world in its own terms .... then are we not adrift in a world where no compass points to right or wrong, and all is reduced to a debate of taste and personal opinion?

Locke (1989)

 

Who is right and who is wrong?

We have now examined the fundamental principles underlying the traditional, quantitative approach to science and the radically different qualitative approach. You are probably left wondering what this all means for you and your studies of the science of sport, health and exercise. Who is right and who is wrong? I can't answer that question for you. You'll have to make up your own mind. Clearly, for some aspects of our area, the natural science model is entirely appropriate. For example, it is hard to imagine how you would use qualitative methods to study the effects of exercise on hormonal imbalances. You can't interview hormones!

When it comes to the social and behavioural sciences, though, we each have to decide where we stand. Is the natural science model appropriate in seeking to understand the psychology of sport, for example? Can a quantitative approach give us an understanding of a top athletes' experience of competitive anxiety in an Olympic final or the difficulties felt by an obese person in trying to maintain a programme of regular exercise? Equally, can qualitative approaches give us trustworthy, valid and reliable findings? Most researchers these days adopt a pragmatic stance: both approaches have their merits and drawbacks and the choice of method depends on the nature of the research question.

In the remainder of the module we will go on to examine how we actually do science. First, in next week's class, we will look at a qualitative approach, examining how we conduct interviews and analyse the resulting data. The coursework assignment, which will involve a qualitative data analysis, will prompt you to think about the scientific merit of qualitative research. We will then go on to examine quantitative methods in great detail. By the end of all this I hope you will have come to a position of your own on the quantitative-qualitative debate.

Evaluating research

In the meantime, I would encourage you to consider three things when evaluating the scientific merit of an approach to research in general or the trustworthiness of a particular study.

First, how objective is it? We might accept that it is not possible to remain entirely detached from the thing we are studying, but to what extent might the researcher's personal vision influence the data collected or the way that it is interpreted?

Second, does the method used allow us to generalise the results to other people, or to similar people in different contexts, or do the findings only apply to a very narrow section of the population of interest and therefore have little practical use?

Finally, to what extent do the findings allow us to determine causal effects? For example, if an athlete tells us in an interview-based study that they failed in a major competition that they were expected to win because of the stresses imposed by the media, can we accept that that really was the cause of their failure and not something else?

 

Now do Fundamental Principles of Science MCQ self-test in Blackboard.