Understanding Research (4 page)
Read Understanding Research Online
Authors: Marianne Franklin
Running through these entry-and-exit points is a line that traces divergent positions on the form and substance of
academic
research; a particular way of producing scholarly –
scientific
– knowledge about our natural and social worlds. Research communities – and larger schools of thought – comprising contending worldviews and research practices crystallize around points along this line; in shorthand, the
quantitative–qualitative divide
.
There are as many ways to define and then slice this divide as there are terminologies to describe and contest its influence on research practices. A sense that there is an underlying division between research findings relying on counting and numerical measurement and those that do not prioritize these ways to gain knowledge pervades modern higher education and research institutions nonetheless. It has come to govern how researchers, instructors, and students not only go about doing any particular piece of research but also how they think and talk about their projects, how they present and defend the validity of any knowledge produced. It inflects the conceptual and technical idioms we use when reflecting on the merits of a completed piece of research; its ‘impact’ or ‘contribution to knowledge’ for instance.
At times polarized and at other times evident only as a quiet but steady undertow, this dividing line also influences how researchers assess the prospective and ongoing work of others in our varying roles; as teachers, external reviewers, examiners, team-members, and students. We may well be making these assessments in more than one capacity. For instance, dissertation supervisors can also be in the process of submitting proposals to funding bodies themselves or working on their own or within collaborative research projects. Students, even as they embark on their first major piece of academic research, may already have data-gathering and analysis skills and so conceptions of what it takes to do good research – from the workplace, in journalism, PR and marketing, or NGO work for instance.
This fact of academic life, an occupational hazard in many ways, is backed up by a formidable corpus of literature across the disciplines. That said, talk of a divide is not to propose either an absolute or intractable rift. This means being aware of where and how it operates as a hard-and-fast rule, or as an implicit imperative that sees students, supervisors, and colleagues getting irate, talking across one another in a research seminar Q&A. It means being able to tell when these tensions divert us from the task at hand as we grapple with conceptual confusions or practical obstacles during the course of a research project. Expressions of seemingly intractable differences often disguise a number of actual and potential points in common. This requires we
develop a sixth sense so we can take note of how differences
within
various approaches can be as intense as those between them.
That said, even carefully talking of a
divide
assumes that everyone agrees on what, and where to draw this particular line. It also assumes there is some sort of consensus about what the terms
quantitative
and
qualitative
mean, along with corollary key concepts. The one thing that is clear is that there is no clear agreement; more on this in due course. Nonetheless, many working researchers – and students – are quick to state which ‘camp’ they, and others belong to if asked; often before being asked. For those undertaking research in settings where one approach predominates – in a natural science department, within a larger research project, or as a prerequisite for funded research, these meta-level distinctions are often a non-issue.
In other settings – in faculties professing an interdisciplinary approach, those with diverse or international teaching or student bodies, or where sharply diverging research approaches co-habit in the one department – students are expected to proceed as if these distinctions and what to do about them during a research project are self-explanatory.
When embarking on your first or even successive piece of substantial research, in even the most inclusive setting, all topics and methods for getting information (‘data-gathering’) appear to be equal. The stated objective for you is to design, undertake, and write up an ‘original’ piece of research under your own steam; as an ‘independent’ researcher. Quite early on though, even the most inclusive environment can become less accommodating; what you consider to be original, innovative, or a ‘hot’ topic is met with comments such as ‘that’s been done to death’, disinterest, or disapproval. As different staff members give conflicting advice or no advice at all, the formal requirements for research dissertation work in your institution and informal expectations appear increasingly opaque if not at odds with one another. In the meantime you are struggling to get off the starting block.
These setbacks are commonplace, for students and full-time scholars working within the various disciplines and institutional geographies that now make up the social sciences and humanities. We are all regularly confronted with thorny practical and philosophical issues regarding the nature and purpose of the research we are undertaking. The ante is being constantly upped by calls from society at large, usually voiced by state or private funding bodies for scholars to undertake research that is applicable –
relevant
– to public policy priorities, a greater good, or towards some sort of national or regional competitive edge in research and development.
Where the line gets (re)drawn and its impact on research practice differs from place to place, and over time; here readers can, and will draw on their own experiences. Apart from its role as an organizational device, in this book I use the term less categorically than
heuristically
; as a framing device for unpacking both opposing and intersecting approaches to designing, carrying out, and then communicating the outcome of a research project. Knowing where the pitfalls lie helps anyone starting out to achieve this goal with some modicum of success and sense of a job well done, whatever your level or ambitions.
Let’s turn now to some more specific terms and their various usages; these often trip up first-time researchers undertaking projects in consciously or organizationally mixed research contexts, as well as those interacting with others working on similar or overlapping research topics yet doing so from the ‘other side’.
All along the spectrum spanning qualitative from qualitative to quantitative in theory and how these translate into practice, seemingly common terms of reference can belie complex and longstanding debates. These terms create an undertow in that any researcher is both the
beneficiary and the victim of the linguistic tradition into which he [sic] has been born – the beneficiary inasmuch as language gives access to the accumulated records of other people’s experience, the victim in so far as it confirms him in the belief that reduced awareness is the only awareness and as it bedevils his sense of reality, so that he is all too apt to take his concepts for data, his words for actual things.
(Huxley 1954: 23)
The following section looks at several of these terms in so far as their various usages relate to this book’s approach to understanding research in interdisciplinary and intercultural settings.
When people refer to
qualitative
or
quantitative
research modes, what sort of distinction is being made here for practical purposes, and what do these signal about deeper philosophical differences? Basically, leaving aside the way polemics can confuse things, the terms designate how proponents understand
- the role of observation in getting an outcome; and
- the types of data gathered during the research;
- differences in what counts as
empirical
evidence; quantitative (numerical) values and measurements, or qualitative ones (field notes, interview transcripts, ideas); how these are processed, analysed, and then presented.
Whether exploring post-conflict scenarios, how different ethnic groups find expression in, or are excluded from mainstream political processes in liberal democracies, or the relationship between globalization, neoliberalism, and the internet, you may notice that these first conversations revolve around certain assumptions, sometimes explicit sometimes not, about which facts are based on which observations, whether your observations ‘fit’ your theory, or conversely whether your ‘theory’ – or hypothesis – has sufficient evidence to back it up. Indeed you may sense that some topics are synonymous with approaches that concentrate on gathering evidence or, conversely, approaches that treat evidence, and the way such evidence is gathered, more as a theoretical endeavour. As Hans Radder notes, that which
is observable and what it is that we observe at a particular moment depend, among other things, on the available conceptual interpretations. In this sense these
conceptual interpretations can be said to structure the world. . . . [H]owever . . . concepts also abstract from the world . . . In conceiving the world, we aim to go beyond the set of observational processes we happen to have realized thus far.
(Radder 2006: 179)
Not only have these different positions ‘been the subject of fierce philosophical debates’ (Radder 2006: 1) over the ages, they continue to splice the quantitative– qualitative divide in everyday research settings.
The usual image is that quantitative modes of research support regard observation (getting the ‘facts’) and conceptual interpretation (‘theorizing’) as separate issues whilst qualitative researchers regard them as connected; observations make no sense without some conceptual apparatus by which to interpret them. This stereotype is not entirely correct, as we will see in subsequent chapters. That there is a connection between these two activities has been accepted in quantitative domains for quite some time (see
Figure 1.1
).
The actual dividing line is a practical one; which sorts of questions and the means by which we get there best provide us ‘with justified knowledge about an independent world’ (Radder 2006: 2; see also Adorno 1976).
For both modes, separately and together, the relationship between, first, observation and then data, however, is not a clear correlation; what you see is not always what you (think you) get; more on these matters in
Chapters 1
,
3
and
7
. Contrary to stereotypes, even
quantitative
modes of social research use relatively blunt instruments for measuring when compared to those employed in the natural or physical sciences; the relationship between the act of observation and the data observed –
facts
gathered – is always moot.
No matter how they regard issues around fact and observation, the role of objectivity vis-à-vis standpoint, researchers across the board are engaged in some sort of observation in varying degrees; asking questions, taking field notes, reading historical documents, setting up experiments, or running focus groups. However, in quantitative work the researcher’s concern is to
minimize
two elements: first
errors
in observation and, second, to understand the source of the error – whether it is the result of ill-defined concepts, poor instrumentation,
bias
, or poor preparation on the researcher’s part. For qualitative researchers these concerns are considered more a point of departure than a problem to be resolved; in varying degrees that are also informed by underlying
worldviews
and research identities. At the end of the day, most researchers would concede that ‘seeing’ is not always ‘believing’.
The key point here is that there is no immediate consensus about what is meant by
data
, let alone the adjective
empirical
, in either case, even when there are clear distinctions between how the ‘data’ are collected, processed, and then used as substantiating ‘evidence’, or
proof
in some quarters. For practical purposes it is still possible to make two clear distinctions.
First,
quantitative data
refers to the types of information that can be counted or expressed numerically; expressing a certain quantity, amount or range. For example, weight expressed in stones or kilos is being referred to as a quantitative indicator. Usually, there are smaller measurement units associated with the data such as pounds or grams; inches or centimetres with respect to feet and metres as in the case of height or length. Quantitative approaches imply that arithmetic operations will be applied to the
data
. Therefore, simple and more advanced statistics are used as well as presenting the outcomes as values in graphs and tables.
Second,
qualitative data
refers to types of information that are non-countable or not expressed numerically. This information includes elements that are termed ‘intangible’, or ‘immeasurable’ because they express qualities, values, states of mind, and ideas; in themselves open to any number of qualifications. To use the same example as above, weight can be expressed in terms like ‘heavy’ or ‘fat’, ‘chubby’ versus ‘light’, ‘thin’ or ‘skinny’. These terms are qualitative indicators and whilst they correlate in some way to numerical measurements, what these terms mean alone and relative to one another is not simply about their respective numerical value.
Often these differences are understood as objective versus subjective measurements of weight, but I would argue this is too simplistic. As in the case of weight perceptions for those people who are anorexic or bulimic, in light of what friends and family think or recurring public debates about the rights and wrongs of ‘size 0’ supermodels or concerns about obesity, what someone quantitatively weighs as measured in stones or kilograms, may have little bearing on how they see themselves, or others see them in qualitative terms like ‘fat’ or ‘thin’; ‘Rubenesque’ or ‘Twiggy-like’. When couched in such value-laden terms the numerical values assigned to fatness or thinness then take on a double-life; as literal and figurative notional values.
Within the quantitative tradition, data can also be treated in qualitative ways (see
Chapters 1
and
7
). In other words, these two notions of data-types are not necessarily mutually exclusive in practice, everyday life or academic. Moreover, the significance attributed to numerical values also differs over time, cultures, and between generations. An
a priori
strict division between quantitative and qualitative data and therefore how to treat them is more polemical than practicable. For instance, the standard numerical indicator of an optimal weight-to-height ratio in western societies, the body mass index (BMI), is as much a qualitative measurement as it is a quantitative indicator, the optimum position to which increasingly over-nourished and under-exercised populations are supposed to aspire.
However, in other parts of the world the converse is the case; a position towards a BMI that indicates too much body ‘fat’ is indicative of well-being, beauty and relative wealth. The ‘measurement units’ at stake in this case are numerical yet overlaid with social and cultural judgments. So, here not only are aesthetic, social, and cultural criteria implicit in the terminology used even when statistics may well be employed to underpin these qualitative indicators, but so also are the points on the scale itself. The difference between these numerical and non-numerical indicators of body mass, size, or body-image is a question that incorporates the uses of qualitative and quantitative modes of conducting research about these matters, and the conclusions we draw from them.
How do these distinctions work in practice?
Quantitative researchers
may think of data as discrete observations presented by numeric values or points on a graph. For example, when conducting surveys of voters after an election, a researcher would ask them for which party they voted, then record the response they give as a numeric value that is intended to represent a particular party. These bits of information about all respondents in the survey are then recorded and compiled into one file. This is a
data set –
a set of data points or observations (see
Figure 1.2
). This quantitative treatment of data is based on the root meaning of the word; ‘data’ comes from the Latin,
dare
, to give, and means ‘that which is given’. In the quantitative tradition, data-collection often involves technologies of observation and of record that are themselves anchored in previous, often hard-won results, or data sets.
This interplay between previous knowledge and subsequent accumulation is modelled after research conventions in the physical sciences; for instance where you might envision scientists extracting certain data from observing and recording the results of a (controlled) collision in a particle accelerator, collecting data on ice flows, or on planetary motion with increasingly powerful and refined telescopes.
Qualitative modes
tend to eschew using the term. For a variety of reasons to be sure, but the implicit discomfort for approaches that do not work at all with numerical indicators, let alone statistics of any kind, is that ‘data’ has become identifiable with quantitative – numerical – expressions of scientific knowledge, legitimated by statistical methods of analysis.
**TIP: When starting out, be aware of fundamental differences in perception and conceptualization of research itself; it can be counter-productive to set out on a research project convinced that the one or the other type of data represents the sort of research you are
not
doing. Most research today, those engaged in pure theoretical projects aside for the moment, involves the gathering and analysis of various sorts of empirical evidence, or data.
For the sake of argument, and in order to establish some common moment of departure, there is no real reason why non-numerical sorts of data cannot fall under the rubric of the original Latin,
dare
; philosophical and practical debates about what can, or cannot be taken as ‘given’ in social relations, communications, or behaviour notwithstanding. That which is ‘given’ can also be taken to be a non-quantifiable, i.e. non-numerical, entity that need not involve arithmetical calculations: ideas, attitudes, written texts, images, various categories of ‘meaning-making’. In addition the conditions in which these ‘givens’ occur can also be interrogated. Turning the latter into numerical entities or not and their re-presentation is where the quantitative– qualitative divide boils down to graphs versus pages of text in the final product. The upshot is that these idioms become synonymous with oppositional views about the whole enterprise (
Figure 1.3
). As Creswell and others note, much research lies somewhere between these two extremes; the various elements indicate a tendency, even a preference for nominally qualitative, quantitative, or – a combination of the two – ‘mixed methods’ (Creswell 2009: 16; see also Berg 2007, M. Davies 2007).
Figure 1.3
Differences between the humanities and social sciences
Source
: Jorge Cham:
http://www.phdcomics.com
As we see above, different usages for the same terms of reference, or misconceptions of terms specific to a particular way of conducting research, can be off-putting enough, also compounded by the implications continual organizational (re)mergers have for how far different approaches go in order to accommodate each other in the conduct of their research affairs. The virtues of multi/interdisciplinary work are often asserted more often than they are put into practice, a function of interdepartmental reshuffles rather than concerted attempts at methodological synergies or definitional rigour. In addition,
intra
disciplinary differences, whereby the terms of reference sound the same but operate very differently can also be hard to get to grips with in everyday research work. The same word, sometimes a deceptively simple term, can have a meaning and use that is peculiar to a particular method or even sub-discipline. It can also be the bone of contention within a community.
So, where does that leave us in this maze? Can all researchers take the term
empirical
to be an applicable adjective for evidential material as data broadly defined? How direct does any observation have to be before it can be regarded as empirical? Conversely, how much conceptualization is needed in social research scenarios before putting this to ‘the test’ in some way or another? Rather than attempting to respond to these questions about the production of knowledge and how the latter can be expressed and then valued – a job for philosophers and methodologists – let’s consider things more pragmatically. For instance, it might well be that numerical indicators can throw light on a textual analysis or an understanding of how values like weight are what many qualitative researchers, feminists particularly, call ‘socially constructed’ or ‘co-constituted’ (see Harding 1998a, 1998b).
**TIP: Before getting too hot under the collar, and this happens sooner than you might think, it is time well-spent to think about the key terms of reference that inspire you or relate to the topic you want to investigate: (1) Take a blank page/screen and make some initial commitments about what
you
mean before worrying about the ‘right’ reference or buzzword; (2) then go search the literature; take note how ideas are conveyed as a variety of concepts and arguments (see
Chapter 4
). Settling on your own point of departure and selectively reading
relevant literature even at this early stage helps you in navigating competing terms of reference in your area of interest.
20