Qualitative versus Quantitative Research:

Key Points in a Classic Debate

James NeillLast updated:

Features of qualitative & quantitative research

Main points

Recommended links

Features of Qualitative & Quantitative ResearchQualitative Quantitative

"All research ultimately has a qualitative grounding"

- Donald Campbell

"There's no such thing as qualitative data.

Everything is either 1 or 0"- Fred Kerlinger

The aim is a complete, detailed description.

The aim is to classify features, count them, and construct statistical models in an attempt to explain what is observed.

Researcher may only know roughly in advance what he/she is looking for.

Researcher knows clearly in advance what he/she is looking for.

Recommended during earlier phases of research projects.

Recommended during latter phases of research projects.

The design emerges as the study unfolds.

All aspects of the study are carefully designed before data is collected.

Researcher is the data gathering instrument.

Researcher uses tools, such as questionnaires or equipment to collect numerical data.

Data is in the form of words, pictures or objects.

Data is in the form of numbers and statistics.

Subjective - individuals’ Objective – seeks precise

interpretation of events is important ,e.g., uses participant observation, in-depth interviews etc.

measurement & analysis of target concepts, e.g., uses surveys, questionnaires etc.

Qualitative data is more 'rich', time consuming, and less able to be generalized. 

Quantitative data is more efficient, able to test hypotheses, but may miss contextual detail.

Researcher tends to become subjectively immersed in the subject matter.

Researcher tends to remain objectively separated from the subject matter.

 (the two quotes are from Miles & Huberman (1994, p. 40). Qualitative Data Analysis)

Main Points

Qualitative research involves analysis of data such as words (e.g., from interviews), pictures (e.g., video), or objects (e.g., an artifact).

Quantitative research involves analysis of numerical data.

The strengths and weaknesses of qualitative and quantitative research are a perennial, hot debate, especially in the social sciences.  The issues invoke classic 'paradigm war'.

The personality / thinking style of the researcher and/or the culture of the organization is under-recognized as a key factor in preferred choice of methods.

Overly focusing on the debate of "qualitative versus quantitative" frames the methods in opposition.  It is important to focus also on how the techniques can be integrated, such as in mixed methods research.  More good can come of social science researchers developing skills in both realms than debating which method is superior.

Chapter 2: QUALITATIVE VERSUS QUANTITATIVE RESEARCH

Rob McBride and John Schostak

As mentioned in the previous chapter, qualitative research is often contrasted with quantitative research. The picture is complicated when we consider that within each of these broad categories, which we described in Chapter 1, there are what can be called 'sub schools of thought'. Nevertheless, there are some critical issues in which researchers have

a tendency to jump one way or the other, depending on the set of beliefs they hold. We believe too, that some of these issues are more important to students who are carrying out research for the first time, especially in an age when quantitative measures such as school league tables have such a high profile in the public eye. Let us consider an example:

A researcher found that in a school of twenty teachers, fifteen preferred children with special educational needs to be withdrawn from the classroom and five preferred in-class support. Should the school adopt a policy of withdrawal or should it have a mixed policy depending on which teachers were involved? The evidence, as it stands, could be used to support either. More importantly, we cannot be sure of the strength of feeling of the teachers nor why they hold their views. It might be that the fifteen dislike a particular kind of in-class support and would prefer a different kind to withdrawal. Indeed, until we find out more details of why these teachers hold their opinions, a range of possibilities can be imagined for policy and, of course, for practice.

Qualitative researchers are interested in answering those why? questions and are not prepared to simply accept the quantitative answers. That is not to suggest that the quantitative data is not important for to know that fifteen out of twenty have one view rather than another is useful. It is just not enough on its own. We could go further and say that when placed alongside qualitative evidence, quantitative evidence is both clear and powerful. Unfortunately it sometimes appears so powerful that it overpowers the opinions of the people involved and this is a danger we have to watch. In addition there are still many researchers, especially the less experienced ones, who are not prepared to 'go the extra mile' and add the extra understanding to the figures they have collected. This course is centred upon the qualitative element in research and while it is not without problems qualitative research is the major form of educational research now practised. Let us now consider the major points of contrast and debate between the broad categories of qualitative and quantitative research. The section that follows rests heavily upon a structure used by Hammersley [1991]. The arguments used here, however, are very different from his.

A. DATA

Where a quantitative researcher might seek to know what percentage of people do one thing or another the qualitative researcher pays much greater attention to individual cases and the human understandings that feature in those cases. Nevertheless, one finds the latter using terms such as 'frequently' and 'the majority of people' and so on.

It could be argued that the quantitative researcher is more precise but the response would be that with people it is not possible to be so precise, people change and the social situation is too complex for numerical description. We could also ask if it is any more help to know that 58.6% of teachers in a school take one view than to know that to one

degree or another, most teachers take this view? We, as qualitative researchers, would argue the qualitative perception in this last case is a more precise reflection of the situation than the numerical perception. When you begin to interview, as part of this course, you will often find people who will offer conflicting or unclear views. Some of your respondents will say "I think this is a good idea but ..." Now in those sorts of situations do you place this person in the 58.6% or in the remaining 41.4% or create a third category? Quantitative research has a tendency to 'clarify' where clarification is not appropriate.

At the heart of this discussion is a point about knowing. We might ask, "How sure can we be about what we claim to know?" In education we have to deal with what is sometimes called 'soft' knowledge, as opposed to 'hard' knowledge. We see claims [not always justified] about certainty in the natural sciences and mathematics. Qualitative researchers, and indeed educationists, have to be more circumspect. As we have seen above, quantified evidence can be very powerful but it can also hide a great deal about people, especially their understanding.

B. RESEARCH SETTING

Many qualitative researchers have long criticised laboratory based research as 'artificial' and noted that people react differently in other contexts. There are also criticisms about those researched being influenced by the researchers so that conclusions are not sound, especially when compared to research in 'natural' settings. One response to these arguments are criticisms about the artificiality of structured interviews which qualitative researchers carry out. Of course, interviews need not be structured though the central issue is about the extent to which the research act interferes with what is researched. In other words are the conclusions valid, do they reflect what they believe they are reflect or are people responding, above all, to the researchers?

Hammersley argues: [p231]

"In my view this distinction between natural and artificial settings is spurious. What happens in a school class or in a court of law, for example, is no more natural [or artificial] than what goes on in a social psychological laboratory."

To us this is simply wrong. There is an enormous difference. If Hammersley had argued that there is some form of reaction to all forms of research we could have accepted that. He is, however, going much further. In qualitative research we seek to minimise the impact of our interventions [see triangulation below, for example] but also recognise that there are other ways in which we do intervene. This is not too much of a problem if we remember that we are not trying to create objective knowledge. Our knowledge is much softer. We cannot be certain that practical work will always make learning easier. We cannot prove that a pupil will respond positively to using a word processor. Yet we can have a pretty good idea that these maybe helpful to us in certain situations. More importantly we endeavour to 'build' theory from the ground of experience or practice. For

qualitative researchers the context in which practice takes place has an important bearing upon that practice and research should be rooted accordingly.

There are other implications of our position. One of those is how we might transfer our research findings from one situation to another. This is called generalisation and we will discuss that below. Another is that qualitative research does not avoid the complexity of social life. Instead great efforts are made to illuminate and understand social situations and human feelings through immersion and detailed, in-depth exploration.

C. MEANINGS VERSUS BEHAVIOUR

In an earlier version of this introduction to qualitative research we wrote the following:

"Where quantitative forms of research, employing questionnaires and sampling procedures attempt to eradicate the individual, the particular and the subjective, qualitative research gives special attention to the subjective side of life. Rather than asking how many people in a given locality have an IQ of 90, qualitative researchers are more likely to ask how it feels to be considered having an IQ of 90, what intelligence means to a given community, and, what is or is not considered to be intelligent by that community. That is to say, they focus upon the social construction of such things as 'intelligence', 'special educational needs', 'behaviour problems', and so on. In order to find out what a given phenomenon, like special educational needs, means to people it is necessary to ask them and to observe what they do. That is why both interview and observation are key techniques in qualitative research. Rather than starting with a definition of special needs, the definition 'grows' from the data that is gathered from interview and observation. Thus theory tends to be built from the ground of experience rather than through academic reasoning distant from the scene of everyday experience."

Hammersley [op. cit.] accepts that qualitative researchers seek to articulate the views of people studied but adds that qualitative researchers often analyse the data in ways that are likely to be alien to those studied. He also asserts that much quantitative research concerns itself with the 'attitudes' of those studied and is therefore grounded in the realities of people.

We hold our position in these matters. Quantitative research remains, in our view, more interested in what people do without a very complete understanding of those actions. It tends, therefore, to be concerned with behaviour as an end in itself without paying sufficient attention to understanding that behaviour. This is behaviourism. Even where 'attitudes' are explored it is usually through pre-structured questionnaires which do not allow respondents to provide their own agenda. The researcher decides on the important questions. One observes this sort of practice especially amongst those who are not experienced researchers.

A Professor of Education at UEA once argued, in a discussion about the researcher and objectivity, that by the end of an evaluation the evaluator tends to lose his/her personal

views about the project being evaluated. Instead the evaluator becomes an information broker on behalf of others, adopting an even-handed impartiality. This, he thought, was the best we could expect as somebody has to carry out evaluations. The listening brief and the intentions of quantitative researchers is far more 'people centred' than that of quantitative evaluators. The qualitative researcher seeks to understand and to relate the subjective understandings and the actions of those being studied. Moreover, in some cases, the relationship between the researcher and the researched can be a very close one even to the point of collaboration.

D. QUALITATIVE RESEARCH AND NATURAL SCIENCE

As qualitative researchers we have often found ourselves being criticised by natural scientists for not providing quantified conclusions and, equally, we have defended ourselves and criticised their work. Hammersley [op. cit.] points out that there is more than one research methodology in the natural sciences and a number of interpretations of these. Nevertheless, some writers, such as H-G Gadamer, have criticised the broad approach of the research methodologies of the natural sciences. We, among others, have found patterns in these criticisms which are very similar to those noted by writers such as Gadamer and I turn now to these issues.

What has been most disconcerting is the rigidity of thought that we have experienced when discussing qualitative research. We have found that natural scientists place great store on what they call objective knowledge. This is knowledge which fits into a scheme that they are familiar with and about which they claim to be certain. Unfortunately, many natural scientists are not aware that even within their disciplines there are fads and fashions and different ways of conceptualising the data they gather. Different academics carry out pioneering work and come up with different terms for describing their findings and the new terms then take over from the old ones. At the same time some of the concepts widely used begin to change. In short, the frameworks used by all the forms of knowledge we have change and what was once considered objective becomes less certain and more problematical. A great deal of science is involved with solving the puzzles that are predominant rather than actually critically deconceptualising the basic assumptions.

The view described in the paragraph above might be called the objectivist view and we see it sewn into the heart of the national curriculum which has a clear unquestionable framework which encompasses all of the knowledge that children need. A different set of views are held by someone we might call a subjectivist. Decisions made by a thorough going subjectivist rest upon whim, personal taste or bias. Both objectivism and subjectivism can be contrasted to relativism which is the basic idea that when we consider fundamental notions such as rationality, truth, reality, good and so on, each is as acceptable as any other. The relativist relates ideas to a given conceptual scheme or set of social practices or historical era while the subjectivist relates ideas to him or herself.

We could actually accuse anybody who stuck rigidly to any of these perspectives as intransigent and unable, as a researcher, to represent other people in these terms. Qualitative researchers need to be 'good listeners' and have some flexibility as they carry

out their craft. In this respect dialogue between people is a critical element of and the qualitative researcher needs to always be available and, indeed, willing to engage in discussion with all the participants in a piece of research.

E. THE USE OF HYPOTHESES

Another tendency among quantitative researchers is to see their studies as centrally concerned with testing an initial hypothesis. It is rarer but not unknown among qualitative researchers. One contrast drawn in this respect is that between explanation and understanding. It is argued that the quantitative researcher seeks to explain an initial hypothesis but the qualitative researcher strives to understand the views of the 'actors' in a school or a project, for example.

We have found that initial hypotheses tend to be poorly informed and that after a period of 'immersion' in a situation the researcher is better able to draw hypotheses [mature hypotheses] which emerge from experience in a setting. By way of example we know of a teacher who was told by her headteacher to ensure that all of her class were engaged in silent reading for twenty minutes every Tuesday afternoon. It worked particularly badly in her class, two statemented pupils were especially fractious in this time. Her initial hypothesis was concerned with explaining how her colleagues, who she took to be good practitioners, were making this activity work well. To her surprise she found that most teachers in the school were largely ignoring the headteachers instruction and not really doing silent reading at all. She found then that her hypothesis became more concerned at understanding how the policy was formed and how best a school might embrace and influence the actual practice of teachers. Her first hypothesis was rooted in her rather limited early grasp of the situation rather then the better informed understanding she acquired after her first round of research.

While qualitative researchers are more concerned with understanding then explaining this is not always the case. It is sometimes radically suggested that a thorough understanding requires the researcher to have direct contact with the social reality to the point of actually taking part. This is often stressed in the ethnographic literature and opens up the issue of whether the researcher can be or is an 'insider' or an 'outsider' or some mixture of the two. Qualitative research does aspire to an 'insider' view and this requires the researcher to mix in in some way rather than adopting a detached stance. We will return to this question below. In addition we will suggest some ways, below, in which researcher bias might be minimised in the 'insider' situations.

F. THE PERSONAL/PROFESSIONAL ELEMENT

Consider these quotations:

"There is no standing ground, no place for enquiry, no way to engage in the practices of advancing, evaluating, accepting and rejecting reasoned argument apart from that which is provided by some particular tradition or other." [MacIntyre, 1988, 350]

"There is no way to engage with or to evaluate rationally the theses advanced in contemporary form by some particular tradition except in terms which are framed with an eye to the specific character and history of that tradition on one hand and the specific character and history of the particular individual or individuals on the other." [ibid, 398]

These are the comments of an influential philosopher and in more simple words he is making two points. The first is that our perspectives in areas such as rationality, truth, reality, truth and so on are influenced in part by traditions of thought that have historical importance. We are not always aware of the complexity of these traditions but most of us know that we have some kind of allegiance to notions of democracy, freedom, the wisdom of the practitioner and so on. MacIntyre goes further and asserts that we can only take part in discussion with others insofar as we recognise the views of others and they recognise ours. It is very difficult to hold a discussion with someone who believes that all children with special educational needs should be removed from schools altogether.

The second point raised is that there is a bit of each individual in his or her own views. Ultimately we cannot remove this piece of us. Now when we are speaking of research these two points have major ramifications. One of these is for the validity of the conclusions we draw. Are we actually representing people and situations fairly and accurately if our own views are present? This leads us to two further issues, namely, what do we do about the personal/tradition element and also can we generalise our research findings to other people and places? Let us deal with these in this order.

Let us begin by saying that we take it that in this case the most important tradition that affects most teachers is the educational one. Over the last century and a half, and indeed before, a huge stock of experiences, ideas, a history and a an educational literature have grown. Practitioners and theorists have developed both separately and together to create a culture or social practice of similar minded people. They do not all agree but there are strong ties, often expressed as stories or jokes, which enable us to debate what we do and believe. It is the strength of this tradition which is an important part of what is called a profession. Currently this professional standing is under rigorous and aggressive attack, primarily from central government.

Rather than suppress our traditions and our personal views - steps sometimes taken when quantitative researchers claim objectivity - qualitative researchers tend to draw attention to them. This form of reflection allows the reader to judge the biases and position of the researcher and simultaneously gives the researcher the opportunity to rigorously explore his or her own interpretations of the data s/he has collected.

G. GENERALISATION.

We are told that we have an entitlement curriculum which all children should receive. This goes beyond mere entitlement to become compulsory. Implied therefore is the notion that all children should experience the national curriculum. We are told that all children should experience a course in English grammar and should be able to speak standard English. Again, implied in this expression is the idea that this is the case for all

children. We hear similar statements about what is a good school with respect to the pastoral systems they require, the way they are managed, classroom conduct and so on. These are examples of generalisation.

In research we have to be very careful not to suggest that what we have concluded from one piece of research is then applicable to all other classrooms, teachers, schools etc. It is not necessarily the case that quantitative researchers generalise but there is a tendency for conclusions to be drawn from what appears to be the numerically largest group. Recently when bidding for research funding we were told that our conclusions had to be quantified so that a board could make well informed decisions. By this was meant that a single decision could be made for all to adhere to.

In our experience the tendency to generalise is most marked among those who seek clear and simple conclusions and among policy makers. The reader might imagine that the policy making process would have a different complexion if policy makers were given clear numerical data in contrast to the data from qualitative research. The latter might present a small number of cases which were researched in great depth. Policy makers would not be able to treat the conclusions as hard knowledge but as a soft understanding of what was making their employees tick. The policy would not tend towards being firm and prescriptive but flexible and open to some interpretation. For those policy makers who believe that they need to be in control, the second of these two possibilities can be extremely threatening, even in a democracy.

Up to this point the term generalisation has been used to refer to the move from one case to all others. We can talk about generalisation from one to any number of other cases. In qualitative research we tend to study individual cases and once we have drawn conclusions we have at least two ways forward. The first is to say that this is soft knowledge but the conclusions are probably plausible to most practitioners. The strength of this case is increased if we have tested our conclusions on other practitioners. A second is to say we are not sure if our findings can be applied to other social situations and that the judge of this should be those practitioners who work in other situations. There is plainly an overlap in these two positions. Either way we have to be extremely careful not to jump from one situation to another and thereby make unfounded claims.

H. ACTION

Qualitative research pays considerable attention to the action it seeks to bring about. In this respect there are efforts to ensure that the gap between theory and practice is minimised, or more commonly, that theory and practice interact.

Qualitative researchers are concerned to 'build' theory from the ground of the experience of practitioners; to research face to face levels of interaction; to focus on the everyday or routine. These are allied to a fundamental respect for individual human beings, sometimes exemplified by a concentration upon the 'underdog'.

Qualitative researchers do not seek the 'detached objectivity' of the quantitative researcher. Rather he/she tries to engage practitioners in his/her research and to report findings in terms which are familiar to the subjects of investigation. Ultimately, it is this engagement which gives subjects a stake in, and an understanding of, the research. This is considered the basis for action and change. Indeed we might consider an understanding of action and change as the stock-in-trade of the qualitative researcher.

I. PROFESSIONAL DECISION MAKING, THE REPRESENTATION OF THE CASE AND GENERALISATION FROM CASES

Decisions are made all the time, often with hardly a thought given to how the decision was made. When asked, we might say 'I drew upon my experience'. That implies that something about the present situation raises an echo of previous similar situations. The reasoning may be: 'If I did 'x' on previous similar situations, then I should the same in this situation as described more fully in Chapter 8 on learning theories. In schematic form this can be expressed in simplified format as follows:

The above diagram was inspired by a discussion with a group of health professionals who were reflecting upon how they came to decisions. They were in agreement that no matter how similar one situation may be to a past situation, no case is identical. Thus it is always important not only to recognise the similarities but also to identify the differences. Over the years a professional develops a rich repertoire of cases together with their similarities and differences. This enables sophisticated modifications in courses of action as cases are monitored for their variations. This process is an example of generalisation across cases to formulate general principles, categories of explanation and theories as a basis for organising experiences about immediate cases.

By studying cases, considerable sophistication can be brought to bear upon the analysis of each new instance of previous similar cases. A teacher, for example, may have found behaviourist strategies to control a child's behaviour effective in the past. Thus when confronted by behaviour defined by the teacher as 'undesirable', is likely to adopt the same strategies that worked previously. Through practice, the teacher may become increasingly sophisticated in the application of behaviourist principles. In short, the

teacher becomes a behavioural technician. This is the kind of technicist approach criticised in chapter 8. The child is always only seen as the 'problem to be solved'. However, a wider vision can be developed as chapter 8 makes clear, in which the whole technicist approach can be critiqued. To do this the teacher must move to another level of analysis. The child must be seen in context. There is the immediate context of the child which relates to his or her own biography, local community and family experiences. Then there is the context of the classroom with its particular style of learning management, teacher control and so on. Then there is the school, its ethos, its values, its style of management, its levels of resourcing and so on. And so on. This is represented in the following diagram:

Rather than focusing on the child as the source of 'the problem', the teacher now applies a professional critique to the context within which the child, the teacher, and other staff are operating. This kind of analysis can be applied to any situation: the health service, business management, or even the analysis of domestic life. As increasingly sophisticated analyses of the social situation are made, then the evidence base upon which the professional decision maker draws is going to be expanded to include information not previously considered. This will have an effect on the quality of decision making.

In representing the case, it is insufficient to focus on a single instance such as a particular child without seeing that child within a greater context. Even the notion of a child makes no real sense without setting the concept 'child' within the history of the emergence of 'childhood' as a social category. Many historians have pointed to the historical fact that childhood is a relatively recent concept. Certainly, childhood cannot be understood

without recourse to the cultural beliefs and values that a particular group holds about children - particularly about deference to adults, and sexual taboos. Childhood is also surrounded by a complex of laws relating to sexuality, work, and welfare. All this is part of an adequate representation of a case concerning the behaviour of a child. In short, the case of the child's behaviour is socially constructed and can only be adequately represented within the context of that social construction. the same is true for any particular focus for research - the management style of a particular manager, the decision making of a nurse. Each of these are socially constructed. How did the concept of a 'nurse', 'manager', 'teacher' arise? What were the historical antecedents of each concept? How is the role of each constructed within a given society, organisation, department? To arrive at an adequate representation of a case as the basis for generalisation and decision making, its many contexts have to be identified and analysed.

CONCLUSION

By allowing theories to form through what people say and do, qualitative research cannot be easily accused of imposing its theories upon people. Equally by keeping detailed records of what is said and of what happens qualitative research does not reduce the complexity of social life to easily manipulated equations. Rather than skating on the surface of everyday life, its close contact and detailed recording allows the research to glimpse beneath the polished rhetoric, or the plausible deceits; it is able to take more time to focus upon the smaller yet powerful processes which other methods gloss over or ignore. For example, many so called objective facts of everyday life are revealed as being socially constructed. That is to say, if a child is categorised as having 'special needs', that can either be treated as an objective fact which requires no further investigation; or, it can be treated as a social process requiring explanation. If it is simply an objective fact, then the only practical consequence is to find the right 'treatment' for the child. If it is seen as a social process, then the consequence is to inquire into the social functions served by labelling some people as having special needs. It may be found that some social classes, or ethnic groups are more likely to be categorised as having special needs than others. If this is the case, then the objective fact 'special needs' is being de-constructed to reveal how it was constructed; in this fictitious example, constructed through processes of social discrimination.

wilderdom.com/research/QualitativeVersusQuantitativeResearch.htmlwww.enquirylearning.net/ELU/.../Research/Res1Ch2.html