2

THE APPLICATIONS OF QUALITATIVE METHODS

TO SOCIAL RESEARCH
Jane Ritchie and Rachel Ormston

Chapter outline

• Theore cal and applied research

• The func ons of qualita ve research

• Qualita ve research as an independent research strategy

• Combining qualita ve and quan ta ve methods

The previous chapter has described the different tradi ons of qualita ve
research and its main defining features. In this second chapter, we consider
what these features bring to addressing the types of research ques ons that
arise in social research. We also discuss the contexts in which a qualita ve
approach is appropriate as a sole research strategy, before describing ways in
which qualita ve and quan ta ve methods can be most effec vely combined.

Theore cal and applied research

When describing the uses and roles of different research methods, a dis nc on
is some mes made between theore cal, pure or basic research, and applied
research. Theore cal research is concerned with the aim of tes ng, genera ng
or enhancing thinking within a par cular discipline. ‘Basic researchers work to
generate new theories or test exis ng theories’ (Pa on, 2002: 215). Applied
research is concerned with using the knowledge acquired through research to
contribute directly to the understanding or resolu on of a contemporary issue.
As such, its objec ves are usually set or shaped by specific informa on
requirements or by the need to gain insight into an exis ng problem.

In the social sciences generally, and in social research in par cular, there is
some debate about whether it is useful or even valid to dis nguish between
applied and theore cal research. The arguments that underpin that debate
centre around the necessary and inherent interac on between social theory and
social research (Denzin and Lincoln, 2011; Hakim, 2000; May, 2001; Rossi and
Lyall, 1978; Silverman, 2010). Consequently, it is suggested that all research is
based on certain theore cal assump ons, concepts or hypotheses, even if these
are implicit, unacknowledged or ill-formed. As Silverman puts it, ‘Even down-to-
earth policy-oriented research designed to evaluate some social service will …
embed itself in theore cal issues as soon as it selects a par cular evalua on
method’ (2010: 103). Similarly, it is argued that all forms of social research can
contribute to ‘theory’ by providing greater understanding of, and knowledge
about, the social world. Meanwhile, Denzin and Lincoln (2011) have argued that
good theore cal research should also have applied relevance and implica ons.
Either way, there is a view that social research is at its most useful when
theore cal insights and social inves ga on are mutually enhancing, such that
the collec on of evidence ‘is informed by theory and interpreted in the light of
it’ (Bulmer, 1982: 152). Indeed, as Silverman has said, ‘Without theory, research
is impossibly narrow. Without research, theory is mere armchair contempla on’
(2010: 115).

Although the boundaries between ‘applied’ and ‘theore cal’ research are
somewhat blurred in social enquiry, the terms ‘applied social research’ or ‘social
policy research’ are o en used in rela on to studies which aim to contribute to
developing, monitoring or evalua ng policy and its related prac ce (Hakim,
2000; Pa on, 2002; Robson, 2011; Walker, 1985). The policy or programme
under review may be relevant to na onal, regional, local or ins tu onal
concerns and may take place in any of numerous policy fields spanning
educa on, employment, housing, environment, health, social care, poverty,
equality and human rights, criminal jus ce and so on. The objec ves of the
research may be wide reaching in terms of understanding underlying social
problems or cultures, or they may be highly focused on specific services,
interven ons or legisla on. As such, the remit of social policy research, like the
policy process itself, is mul faceted and extensive. It also requires an
understanding of social theory to provide context to, and more fully interpret,
the evidence generated.

The growth in applied qualita ve research

Un l the la er part of the twen eth century the use of qualita ve methods was
much more evident in research that was concerned with developing social
theory than in more applied se ngs. As Chapter 1 described, qualita ve
research has a long-standing history of contribu ng to an understanding of
social structures, behaviours and cultures, but the wide scale use of qualita ve
methods to aid directly in the development and appraisal of social policy
occurred much later. This was certainly so in the UK but also occurred in other
countries, like the US, which had equally strong tradi ons of using qualita ve
methods (Filstead, 1979). It has been argued that one of the main reasons for
this was that policy-makers saw ‘informa on’ or ‘evidence’ as synonymous with
numbers (Bulmer, 1982; Weiss, 1977). Even in the 1960s and 1970s, when there
was significant growth in the conduct of policy-related research, the main
methods used were sta s cally based, o en involving surveys. As it was then
observed, the demand for ‘hard facts’ about social condi ons established a
norma ve sta s cal methodology for social policy research (Payne etal., 1981).

But in social policy as in other fields, there were increasing calls for much
greater u lisa on of qualita ve methods in order to understand more fully the
nature of the problems that policies had to address, and to appraise those
policies once implemented (Rich, 1977; Weiss, 1977). In other words, qualita ve
research was seen as having a crucial role in providing the ‘enlightenment’ or
‘knowledge for understanding’ that was needed for social policy concerns
(Janowitz, 1971; Sco and Shore, 1979). By the early 1980s this recogni on was
reflected in a slow but steady growth in commissions from government and
other public sector agencies for qualita ve research.

From the late 1980s onwards, the use of qualita ve methods in social policy
research began to increase significantly in the UK. Research commissioners and
funders became more aware of the value of the kinds of informa on that
qualita ve research could provide and the understanding it brought. Gradually
its commission and use to inform different stages of the policy process spread
from the few government departments that had pioneered its use to a broad
range of ministries and policy sectors. The expansion of policy and programme
evalua on in the UK, par cularly in the years following the 1997 general
elec on, were accompanied by an increasing use of qualita ve research as a
way of understanding how programmes and policies work in prac ce and the
processes and factors that lead to success or failure. A NatCen review of the
methods used in research studies for six UK Government Departments (Jago,
forthcoming) found that of 86 studies published between January and October
2012, 18 used only or mainly qualita ve methods, while a further 35 used a
combina on of qualita ve and quan ta ve approaches. Their now almost
rou ne use alongside more ‘tradi onal’ evalua ve methods such as surveys and
randomised controlled trials indicates that qualita ve methods are firmly
established within the range of approaches used in government-funded social
research in the UK.

A related strand of ac vity that also gained pace in the 1990s and 2000s
surrounded the assessment of quality in qualita ve research. Although there
had been a long-standing debate about whether – and if so how – quality
criteria for qualita ve research could be set, calls for systema c appraisal
intensified as the use of such methods for public sector research increased.
Alongside this, there was growing use of systema c reviews and meta analyses,
par cularly in the health field, which again called for judgements about quality
to be made. As a result numerous ‘quality’ checklists, frameworks and
guidelines were developed, all of which had a primary concern with iden fying
features of ‘good prac ce’ in the conduct of or outputs from qualita ve
research. The UK Government also commissioned its own study, both to review
the literature on quality assessment and to produce a set of criteria that could
be used to appraise studies using qualita ve methods (Spencer etal., 2003). But
despite all this ac vity – or perhaps because of it – the debate about standards
in qualita ve research s ll con nues, underpinned by related debates about
both philosophical orienta ons and methodological principles (see Spencer and
Ritchie, 2012).

Although qualita ve methods are now embedded within applied policy

research in the UK, some researchers con nue to argue that many policy-makers
view qualita ve evidence as of ‘lower status’ than sta s cal data. Torrance
argues that ‘qualita ve approaches to research are marginalized’ in the debate
about how research might serve policy (2011: 569). Similarly, Silverman feels
that in government research ‘there is li le doubt that quan ta ve data rule the
roost’, with many research-funding agencies viewing qualita ve researchers as
‘journalists or so scien sts’ (2011: 35). To the extent that such limited views of
what counts as research ‘evidence’ do con nue to exist among research users or
commissioners, there remains a danger that ques ons essen al to social policy
will be either misconceived or misunderstood.

The func ons of qualita ve research

To consider the par cular role of qualita ve methods in providing the kinds of
informa on and understanding needed in social research, it is useful to consider
some of the broader func ons of social inves ga on. These have been defined
in various ways, depending on the purpose of the classifica on. In more
theore cal research, for example, dis nc ons are o en made between the
func ons of theory building, hypotheses tes ng and content illumina on.
Alterna vely, in applied research, the policy-making cycle is some mes used to
define the different types of research needed during the key stages of policy-
making – that is, formula on, where research to understand a social problem
and/or generate ideas for a policy or programme is required; implementa on,
where research focuses on the process of implemen ng a policy and the
barriers and facilitators to doing so effec vely; and appraisal, which is
concerned with assessing impacts.

From these and other such defini ons, it is possible to iden fy a broad,
although comprehensive, classifica on of the poten al func ons of social
research as follows:

• Contextual – describing the form or nature of what exists

• Explanatory – examining the reasons for, or associa ons between, what

exists

• Evalua ve – appraising the effec veness of what exists

• Genera ve – aiding the development of theories, strategies or ac ons.

Different forms of research can contribute to each of these func ons in quite
dis nct ways, as discussed later in this chapter. Here, we are concerned with the
kinds of evidence that qualita ve research can provide within each of these
broad categories.

Contextual research is concerned with iden fying what exists in the social
world and the way it manifests itself. A major feature of qualita ve methods is
their facility to describe and display phenomena as experienced by the study
popula on, in fine-tuned detail and in the study par cipants’ own terms. It
therefore offers the opportunity to ‘unpack’ issues, to see what they are about
or what lies inside, and to explore how they are understood by those connected
with them. Such evidence can be used to:

• Map the range of elements, dimensions, classes or posi ons within a social
phenomenon, for example

how do parents define ‘good behaviour’ in their children; how do their

children define it?
 what percep ons of poli cs and poli cians do young people hold?

what dimensions are contained within the concept of a ‘standard of

living’ or ‘a good standard of living’?

• Display the nature or features of a phenomenon, such as

how does racism manifest itself?

how is ‘obesity’ experienced?

what does ‘community par cipa on’ involve?

• Describe the meaning that people a ach to an experience, event,

circumstance or other phenomenon

what does it mean to be a grandparent?

how does it feel to have a criminal record?

what is the significance of a marriage or civil partnership ceremony to

those involved?

• Iden fy and define typologies, models and groups

what characterises different groups amongst people who experience

social exclusion?

what are the different models for organising online support networks?

what defines different approaches to voca onal counselling for young

unemployed people?

These func ons of qualita ve research have been called descrip ve or

exploratory by other authors (Marshall and Rossman, 2011; Robson, 2011), and
indeed both descrip on and explora on are key features of contextual research.
The essen al purpose is to inves gate and capture interpreta ons of social
phenomena as experienced and understood by par cipants.

Explanatory research is concerned with why phenomena occur and the forces
and influences that drive their occurrence. Because of its facility to examine
subjects in depth, qualita ve research provides a unique tool for studying what
lies behind, or underpins, a decision, a tude, behaviour or other phenomena. It
also allows associa ons that occur in people’s thinking or ac ng – and the
meaning these have for people – to be iden fied. These in turn may indicate
some explanatory – even causal – link. Such features make it possible to
iden fy:

• the factors or influences that underlie a par cular a tude, belief or

percep on, for example

what are the underlying factors leading to racism?

what influences people’s beliefs about climate change?

what shapes people’s views about poetry?

• the mo va ons that lead to decisions, ac ons or non-ac ons

why do people ‘decide’ or choose not to have children?

what leads people to become involved in volunteer/community

ac vi es?

what underlies different pa erns of public response to economic

recession?

• the origins or forma on of events, experiences or occurrences

why does homelessness occur?

how do different systems for managing and controlling household

income and expenditure evolve?

what are the barriers that inhibit the use of preven ve health services?

• the contexts in which phenomena occur

in what circumstances does housing evic on take place?

what condi ons give rise to the stability of marital, civil and cohabi ng
partnerships?

in what environments/circumstances does social unrest occur?

The role of qualita ve methods in seeking and providing explana on is widely

recognised within a range of different epistemological approaches (Giddens,
1984; Layder, 1993; Miles and Huberman, 1994; Yin, 2012). There is, however,
debate about whether ‘causes’ of social phenomena can be truly detected, with
some arguing that cause and effect in social enquiry can only be specula ve (see
Chapter 1 and Chapter 10). Even assuming that is so, qualita ve methods s ll
have a crucial role in iden fying the important influences and in genera ng
explanatory hypotheses.

Evalua ve research is concerned with how well things work, an issue that is
central to much policy-related and organisa onal inves ga on. In order to carry
out evalua on, informa on is needed about both processes and outcomes and
qualita ve research contributes to both. Because of their flexible methods of
inves ga on, qualita ve methods are par cularly adept at looking at the
dynamics of how things operate. They can contribute to an understanding of
outcomes by iden fying the different types of effects or consequences that can
arise from a policy, prac ce or system and the ways in which they occur.
Qualita ve research can reveal the many factors that shape a programme or
service, which may not be accessible through quan ta ve methods (e.g. history,
organisa on and culture, personali es, poli cal dynamics, social interac ons
and rela onships between stakeholder). Evalua on research has also made
considerable use of qualita ve methods in scru nising the logic and
assump ons underlying the design of par cular programmes – an approach
referred to as ‘logic modelling’ (W.K. Kellogg Founda on, 2004).

Qualita ve methods can thus be used to:

• Describe in detail what a programme consists of and what assump ons and
theories underpin the belief that it will lead to par cular outcomes, for
example

What does a programme of support for young mothers involve in

prac ce? What are all the inputs? What are the mechanisms through
which it is expected that the programme will deliver its planned results?

Under what circumstances do those delivering a cardiac screening

programme believe that this will lead to reduced incidence of heart
a acks?

What are the risks to the successful delivery of a smoking cessa on

programme? What external factors might prevent it from achieving its
target outcomes?

• Iden fy the factors that contribute to successful or unsuccessful delivery of

a programme, service or interven on, for example

What factors contribute to successful rese lement for people who are
homeless?
 What makes the support services provided by an organisa on effec ve
or ineffec ve for recipients?

What factors facilitate or prevent pharmacists from carrying out

preventa ve health checks with customers?

• Iden fy the effects of taking part in a programme or ini a ve on

par cipants and how such effects occur, such as

What impact does a welfare to work programme have on the

employment ac vity of its par cipants?

How do driver educa on schemes change behaviour?

What are the psychological consequences of holding asylum seekers in

deten on centres?

• Examine the nature of the requirements of different groups within the target
popula on

How should health promo on programmes be designed to engage with

the needs and interests of different ethnic groups?

How do different groups of older people respond to home care services?

What are the requirements of different cons tuencies of people taking

part in adult literacy schemes?

• Explore a range of organisa onal aspects surrounding the delivery of a

programme, service or interven on

What characteris cs and behaviours are required of personal advisers for

the effec ve delivery of debt counselling services?

How is funding most effec vely used by organisa ons providing

community development schemes?

What organisa onal structures are needed to support social work

interven ons for adop on?

• Explore the contexts in which interven ons are received and their impact on
effec veness

How do the personal circumstances of lone parents affect their

par cipa on in employment-related programmes?
 How does the nature of rela onship breakdown affect receipt of
paren ng support services?

How do different personal or domes c circumstances affect secure

parole arrangements?

Various forms of evalua on have been iden fied (Stern 2005) which broadly
depend on the

• purpose of the appraisal – whether for development, assessment of

effec veness, accountability, monitoring, etc. (Pa on, 2012)

• the nature of the system/programme/organisa on/policy being inves gated

– whether newly formed, na onal or local, user or system led, objec ves,
etc. (Bamberger etal., 2012).

One of the most widely used dis nc ons in evalua on is between forma ve and
summa ve modes of evalua on, a concept originally introduced by Scriven
(1967) with much subsequent development. In brief, forma ve evalua ons are
designed to provide informa on that will help to change or improve a
programme/system and its delivery or organisa onal effec veness. Summa ve
evalua on is concerned with the impact of an interven on, programme or
policy in terms of effec veness and the different outcomes that have resulted.

In terms of methods for evalua on two principles are broadly agreed. First,
many evalua ons require a mix of both sta s cal and qualita ve enquiry in
order to measure and understand outcomes and modes of effec ve delivery
(Bamberger etal., 2010; Greene etal., 2005). Second, the methods used need to
be suited to the primary objec ves of the evalua on:
Methodological appropriateness means matching the evalua on design to the evalua on
situa on taking into account the priority ques ons and intended uses of primary intended
users, the costs and benefits of alterna ve designs, the decisions that are to be made, the
level of evidence necessary to support those decisions, ethical considera ons and u lity.
No design should be lauded as a gold standard without regard to context and situa on.
(Pa on, 2012: 298)

Thus the role of qualita ve research in evalua on is integral and the way it is
used will depend on the evalua ve ques ons to be answered. But whatever the
purpose, the need for evalua on to be done ‘with’ rather than ‘to’ those most
involved is increasingly recognised. In this context Abma and Widdershoven
(2011) have wri en about the poten al for qualita ve researchers to contribute
to a more ‘interac ve’ form of evalua on, in which researchers and prac oners
engage in ongoing dialogue from which both sides learn. They suggest that
qualita ve evalua on provides an understanding of the programme from the
insider perspec ves of par cipants and other stakeholders.

Genera ve research is concerned with producing new ideas as a contribu on

either to the development of social theory or to the refinement or s mulus of
policy solu ons. Because qualita ve research seeks to capture emergent
concepts and because its coverage is not overly prescrip ve, the poten al for
original or crea ve thoughts or sugges ons to emerge is high. It also allows
ideas to be generated through, and then placed in, the ‘real’ contexts from
which they arise. It therefore has the poten al to:

• develop new concep ons or understandings of social phenomena

the nature of religious fundamentalism

the func ons of an ‘online community’ as perceived by users

changing concep ons of ‘family’ and family rela onships

• develop hypotheses about the nature of the social world and how it
operates

the nature of the UK’s drinking culture and how this compares with other
countries

the nature of paren ng in the twenty-first century

the nature of the social structures that lead to ‘workless’ households

• generate new solu ons to persistent social problems

innova ve schemes to provide effec ve support for older people living

alone

ways of interrup ng cycles of disadvantage

pupil-generated interven ons to stop bullying at school

• iden fy strategies to overcome newer or recently defined phenomena or

problems

factors that reduce the risk of social unrest occurring in a community

ways of restoring declining local economies

 mechanisms to encourage greater preserva on of scarce environmental
resources

• determine ac ons that are needed to make programmes, policies or services

more effec ve

changes that are needed to help improve the quality of care in hospitals

ways of encouraging car owners to make greater use of public transport

mechanisms for the early detec on of child abuse.

Par cipatory Ac on Research (PAR), introduced in Chapter 1, can be framed as a

par cular approach to genera ng new contribu ons to social and poli cal
debate which builds on the idea that par cipants are experts on their own lives.
The core principles of PAR are that ‘it is built upon the no on that knowledge
genera on is a collabora ve process in which each par cipant’s diverse
experiences and skills are cri cal to the outcome of the work’ (Brydon-Miller
etal., 2011: 387), that there will be clear benefits to those involved from the
research, and that the outcome will have a wider applica on due to the rigorous
nature of the principles of good research being systemically applied to the
research problem (Brydon-Miller etal., 2011). PAR has thus tried to break down
the barriers between the ‘insider’ of the research ‘expert’ and the ‘outsider’ of
the research subject, flipping this around to ask cri cally who really is the
‘expert’ at understanding the par cipant’s circumstances.

As was noted earlier, the role of qualita ve methods in contribu ng to social

theory has a well-honoured heritage. The contribu on of qualita ve research to
developing and evalua ng policy has evolved and been more fully recognised in
recent years, with government departments, other public sector organisa ons
and research trusts suppor ng qualita ve research to inform the kinds of issues
listed above.

The following chapters of this book will describe in detail the art of
conduc ng qualita ve research such that the various func ons described above
are fulfilled to an op mum level. But it is important to emphasise again here
that it is because of the exploratory, interac ve and interpre vist nature of
qualita ve enquiry that it can make the kinds of contribu ons it does. Indeed all
the defining features described in Chapter 1 allow qualita ve research to
provide evidence of a unique kind such that knowledge and understanding of
social phenomena, and the contexts in which they arise, is extended.
Qualita ve research as an independent
research strategy
Later in this chapter we describe the various circumstances in which a
combina on of qualita ve and quan ta ve methods might be used in the
conduct of social research. However, as will be evident from the preceding
discussion, there are many occasions when a qualita ve approach will be the
only approach needed to address a research ques on. A number of authors
have iden fied the kind of circumstances in which this might be so (Denzin and
Lincoln, 2008; Marshall and Rossman, 2011; Pa on, 2002; Walker, 1985). There
is general agreement that the factors that determine whether qualita ve
methods should be used at all are centrally related to the objec ves of the
research. But there are also other factors, primarily related to the subject ma er
under inves ga on, that may necessitate the sole use of qualita ve inquiry, at
least in the first instance. These arise when the phenomena being studied hold
certain features:

• Ill defined/not well understood – qualita ve research is some mes used as

a prelude to sta s cal enquiry when the subject ma er needs to be more
clearly understood or defined before it can be measured. However, there are
perhaps more circumstances where qualita ve research is needed to
provide greater understanding of the nature of an issue or problem, but
where measurement of its extent is not at that me of interest. This can
arise with newly developing social phenomena, such as cyber-bullying within
schools; where previous knowledge or understanding has not fully explained
occurrences or circumstances that are known to be widespread (for
example, teenage use of drugs); or where refinements to understanding are
needed (for example, changing understandings of ci zenship in a
mul cultural society). The open and genera ve nature of qualita ve
methods allow the explora on of such issues without advance prescrip on
of their construc on or meaning as a basis for further thinking about policy
or theory development.

• Deeply rooted – there are subject areas in which the phenomena that need
to be studied will be deeply set within the par cipants’ personal knowledge
or understanding of themselves. These may be related to the origins of long-
standing values or beliefs (for example, beliefs about personal autonomy or
assisted suicide); to the forma ve influences on par cular a tudes or
behaviours (for example, gender roles); or to responses to events that have
 been very distressing, joyous or emo onal (for example, feelings about
becoming a parent for the first me). The nature of such phenomena makes
it likely that par cipants will need very delicate and responsive ques oning –
and me – to explore the issues for themselves. This will also allow them to
move below ini al or stylised responses to reach inner knowledge that has
either been suppressed, or has remained largely unconscious.

• Complex – similar issues arise in the study of complex subject ma er where

there is a need to understand phenomena which are innately intricate or
conceptually difficult to relate. The complexity may lie in the nature of the
subject itself – for example, technical ma ers like fiscal policy or
philosophical ques ons like the nature of spirituality. Alterna vely, it may be
that the intricacy relates to the level of unpacking that is needed to
formulate a posi on, view or belief. This can o en be the case, for example,
where cogni ve processes such as judgements or decisions are the focus of
the study (for example the ways verdicts are formed in criminal
proceedings). Again, par cipants will need me to reflect both on the issue
itself and on their own thinking and will require facilita ve ques oning to
help them in the process.

• Specialist – a related point concerns the collec on of informa on from

individuals or groups that have a singular or highly specialised role in society.
Examples would be public figures, leading professionals or ‘experts’ or senior
representa ves of organisa ons. If their views are being sought from the
vantage of their par cular posi ons, then the nature of the informa on is
likely to require exploratory and responsive ques oning. This is partly
because the nature of the subject coverage is likely to be complex and/or
involve aspects of system process but also because their perspec ves may
well be uniquely held.

• Delicate or intangible – certain subjects in social research are difficult to

capture because they are so fragile in their manifesta on. Again this might
be because of the nature of the phenomenon itself, which is either ethereal
or unseeable (for example the ‘culture’ of a community); or it might relate to
the elusive nature of feelings or thoughts that an event or circumstance
provokes (for example, empathic response to other people’s grief). Here,
carefully framed and responsive ques oning or observa on is needed to
help par cipants uncover and relay the delicacy of their percep ons and
responses.
• Sensi ve – as will be discussed in Chapter 7, it is hard to predict the subject
ma ers that might prove distressing or emo ve to individual par cipants.
Virtually any subject ma er could turn out to raise sensi vi es, depending
on the circumstances or experiences of the person concerned. But there are
also subjects which, by their very nature, are likely to generate emo onal
and o en painful responses. Some obvious examples would include
rela onship breakdown, physical or sexual abuse, bereavement or life-
threatening illness. Predetermined ques oning of such subjects is possible
and can provide extremely valuable informa on about the prevalence of
par cular views, behaviours and experiences. However, there are arguably
both prac cal and ethical limits to what more structured ques oning can
achieve in terms of understanding such sensi ve areas. Certainly any in-
depth inves ga on of such ma ers will require finely tuned ques ons that
are responsive to the par cular circumstances of the individual, as well as
sensi ve facilita on to help people to describe feelings or emo ons that
have been distressing or previously gone unexpressed.

The features described above are some of the main determinants of using
qualita ve research as an independent mode of research enquiry. In all cases
they are the kinds of subject ma er that are difficult to address in structured
surveys, at least without some detailed qualita ve work first. Again, the crucial
issues in the choice of research methods are the objec ves of the research and
the nature of the informa on that the research needs to provide.

In this context, it is perhaps important to warn against the other factors that
can inappropriately influence the choice of qualita ve research as the sole
method to be used. Some mes restricted budgets or mescales lead to a choice
of small-scale qualita ve methodology when this is not suited to the type of
informa on required – for example, where a funder really wants to know how
prevalent a par cular experience is, but commissions a very small number of
focus groups rather than a survey because they need the research to be done
quickly and (rela vely) cheaply. In other circumstances, the par cular
orienta on of a researcher or a funding body may influence the use of
qualita ve methods rather than the research ques ons that need to be
addressed. It is therefore important that both research-funders and researchers
themselves ensure that there is good fit between the specifica on for the
enquiry and the methods used to yield the informa on required.
Combining qualita ve and quan ta ve
methods
Nowadays, qualita ve and quan ta ve methods are so o en to be found in
combina on, par cularly in applied research, that it might be temp ng to think
that debates about the logic and process of ‘mixing methods’ were a thing of
the past. Certainly, the view that quan ta ve and qualita ve approaches are so
different in their philosophical and methodological origins that they cannot be
effec vely blended in applied research is diminishing, although it has by no
means disappeared (see for example discussion in Cresswell, 2011). As Mason
puts it ‘mixing methods has come to be seen as a good thing, and … research
funders are increasingly thought to look favourably upon research proposals
involving more than one method for genera ng and analysing data’ (2006: 3).
However, even among those who recognise the benefits of both, different
perspec ves persist on when, how and why qualita ve and quan ta ve
methods should be combined.

Approaches to mixing methods

Recent literature on mixing methods (e.g. Brownlie, 2011; Flick, 2009; Mason,
2006) has focused on what it means, philosophically and epistemologically, to
combine methods. For example is one method seen as more important than the
other in terms of the evidence that is needed? Are the different methods
addressing dis nct but related ques ons? And are the findings produced by one
method intended to validate those produced by the other or to extend
understanding?

Flick (2009) argues that most quan ta ve research textbooks s ll tend to see
survey research as central and qualita ve research as preliminary, or as ‘playing
a more illustra ve part’, with statements in qualita ve interviews ‘tested and
“explained” by their confirma on and frequency in the ques onnaire data’
(2009: 25). Similarly, Mason (2006) argues that the most commonly used logic
for mixing methods remains the desire among researchers to add either breadth
(from quan ta ve research) or depth (from qualita ve data) to their analysis. In
both cases, one method remains dominant and the other is seen as
‘embellishment’ of the argument, rather than an essen al component of it.
Mason suggests that such an approach risks polarising methods and does not
use either to its best advantage.
 A more effec ve way of combining quan ta ve and qualita ve methods is to
see them as equal but separate, suited to answering different ques ons about
the same or related topics. This perspec ve is frequently embodied in applied
social research, where projects o en require measurement of some kind but
also greater understanding of the nature or origins of an issue. Each of the two
research approaches is seen as providing a dis nc ve kind of evidence and,
used together, they can offer a powerful resource to inform and illuminate
policy or prac ce.

To illustrate this, Box 2.1 provides an example of how qualita ve and

quan ta ve methods could contribute differently to a study about the nature of
homelessness and the types of interven ons required. The qualita ve research
is intended to address ques ons surrounding the nature of homelessness, how
or why it arises, and to appraise ways in which different forms of preven ve or
rehabilita ve interven on can be made most effec ve. Meanwhile quan ta ve
research is concerned with the measurement of levels of homelessness, their
distribu on among the popula on, the extent to which homelessness services
are used and future levels of provision required. In other words, both the aims
and the outputs of the two methods in this study are seen as being of a quite
different nature but provide powerful evidence in combina on.

A third purpose in mixing methods is to see quan ta ve and qualita ve

approaches as ‘triangula ng’ each other – that is, describing a social
phenomena from different perspec ves, with each perspec ve tes ng and
adding to or valida ng the other. Simply put, the idea is that qualita ve and
quan ta ve research can each make up for the ‘methodological blind spots’ of
the other to provide a ‘fuller’ picture of the phenomenon being studied.

There has been a long-standing debate about the extent to which

triangula on actually offers researchers a meaningful way of verifying their
findings. There are many strands to these discussions (see for example Denzin
and Lincoln, 2011; Fielding and Fielding, 1986; Flick, 2009; Hammersley and
Atkinson, 2007; Mason, 2006; Seale, 1999; Silverman, 2010) but two key points
are recurrent in the challenges to its valida ng func ons. First, there is cri cism
from an ontological perspec ve that there is no single reality or concep on of
the social world to ascertain and that a emp ng to do so through the use of
mul ple sources of informa on is fu le. Second, it is argued on epistemological
grounds that all methods have a specificity in terms of the type of data they
yield and thus they are unlikely to generate perfectly concordant evidence. As a
result of these concerns, there is now some consensus that the value of mixing
methods lies in extending understanding through the use of mul ple
perspec ves or different types of ‘readings’ – that is revealing ‘the different
dimensions of a phenomenon’ and enriching ‘understandings of the mul -
faceted nature of the social world’ (Gilbert, 2008: 128). In other words, the
‘security’ that using mul ple methods provides is by giving a fuller picture of
phenomena, not necessarily a more certain one.

Box 2.1
STUDY OF HOMELESSNESS: CONTRIBUTIONS OF
QUALITATIVE AND QUANTITATIVE METHODS TO
DIFFERENT RESEARCH FUNCTIONS

This perspec ve is well illustrated in a study of people’s a tudes towards and

experiences of emo onal support – that is, the support people seek when they
undergo difficul es in their lives – in which both qualita ve and quan ta ve
data were collected (Brownlie, 2011). Quan ta ve data were collected from
ques ons in the Bri sh Social A tudes survey exploring issues such as who they
talk to if they are feeling worried, stressed or down, and a tudes towards
talking about emo ons. Qualita ve data were gathered via a series of follow-up
in-depth interviews with survey respondents. In-depth interviews explored the
role and importance of talk alongside other forms of support, the basis of ‘trust’
in others (whether professionals, family or friends), and percep ons of how
views and experiences of emo onal support may have changed over me.

For the authors of this research, the process of bringing together qualita ve
and quan ta ve findings was not seen as ‘an exercise in triangula on, a way of
“nailing” the reality … In fact, it illustrates the opposite: … the impossibility of
closing down all possible readings, imagined or otherwise, of that experience’
(Brownlie, 2011: 472). While some of the qualita ve and quan ta ve findings
from the study resonated with each other, others appeared to conflict. The
researchers argued that these tensions reflected the complexity of the subject
(emo onal lives) and helped deepen understandings of how we think about
emo ons and their expression. Mason (2006) and Brownlie (2011) both argue
for a reflexive approach to mixing methods, that looks for ‘some kind of
intersec on, or interplay’ (Mason, 2006: 9) of the dis nc ve ways of seeing the
world that qualita ve and quan ta ve methods each offer. This recognises that
the different dimensions we seek to research may exist in ‘messy tension’ rather
than in a neatly integrated or triangulated fashion.

Sequencing of quan ta ve and qualita ve methods

Decisions about the role and status of quan ta ve and qualita ve methods will
o en have implica ons for the order in which different methods are employed
within a study – or sequencing as it o en termed (e.g. Flick, 2009; Miles and
Huberman, 1994). Very simply, qualita ve research may precede sta s cal
enquiry, accompany it, or may be used in some form of follow-up study.

Examples where qualita ve research could precede survey or other

quan ta ve research include:

• Studies where the subject ma er is new, underdeveloped or complex and

qualita ve research can help to define terminology, concepts or subjects for
inves ga on. This might include genera ng the ‘real-life’ language in which
subsequent survey ques ons should be framed;

• Qualita ve research can be used to generate hypotheses that can then be

subject to sta s cal tes ng – for example, by genera ng ideas about the
rela onships between par cular phenomena;

• Similarly, qualita ve research can help iden fy the defining features of

different groups within the study popula on in order to inform sample
segmenta on. As discussed in Chapter 10, typologies are a key output of
qualita ve research. They can also be used in subsequent sta s cal analysis
 to measure the size of par cular segments and their distribu on within
other variables.

A more specific but related use of qualita ve methods is in ‘cogni ve tes ng’ to
inform survey design. Cogni ve tes ng involves using qualita ve methods to
assess how people respond to survey ques ons. Through the use of probing or
‘think aloud’ techniques, poten al problems with survey ques ons – including
problems rela ng to the comprehension of par cular terms or phrases, as well
as issues around recall, bias and sensi vity – are uncovered and can be
addressed in the final survey instrument (see Willis, 2005, for more detailed
discussion of cogni ve tes ng). Cogni ve tes ng is arguably not a ‘true’ example
of ‘mixed methods’ research, because the purpose of the qualita ve research is
only to aid in the design of ques on form and content for a sta s cal enquiry.

Qualita ve and quan ta ve methods can also be used in tandem to study the
same or related phenomena where:

• There is a need to examine both the number and nature of the same
phenomenon.

• Different types of informa on are required about a par cular phenomenon.

An example might be levels of spending in different areas of service
provision. Both qualita ve and quan ta ve research would have common
ground in iden fying areas in which spending is greatest. Quan ta ve
research might then be used to provide a profile of expenditure in different
areas over a number of years or to compare the characteris cs of the high-
and low-spending areas. Qualita ve research meanwhile might explore the
processes through which expenditure is controlled in areas with different
levels of spend, or the factors that have led to changes in pa erns of
spending.

• There is a need to understand the context in which par cular phenomena

occur. For example, qualita ve research might explore the cultural
requirements of specific ethnic minority groups in rela on to health service
delivery, while quan ta ve research examines pa erns of take-up and
demand for par cular services amongst these same groups.

• The factors underlying a par cular phenomenon are unknown and/or too
complex to be captured fully through quan ta ve methods. For example,
although quan ta ve research will be able to iden fy barriers to service use
at a global level – that is, awareness, accessibility, cost, convenience and so
 on – qualita ve research is o en be er placed to explain the origins of these
barriers or exactly how they deter people from service use.

Finally, qualita ve research can be used to follow-up on quan ta ve research,

par cularly where this presents findings that need further explana on or where
more detail or depth about a phenomenon is needed. It can be par cularly
useful where there are subgroups that are too small for any detailed sta s cal
analysis, or where the group in ques on appear to have an important
perspec ve on the subject of enquiry that has not been fully explored or
explained in the quan ta ve research.

There is also a case to be made for using qualita ve and quan ta ve

approaches in some kind of interac ve sequence to extend learning or
knowledge about an issue. For example, qualita ve research might be used as a
follow-up to a survey to provide greater understanding of the factors underlying
a problem. Indicators of those factors, already exis ng in the survey data set,
could then be used for subsequent modelling or sta s cal tes ng. In any such
uses, the important requirement is to recognise the linkages between the two
sources of informa on and to maximise their associa on.

Mixing qualita ve methods

The concept of a ‘mixed method’ approach to research is most o en discussed
in the context of combining qualita ve and quan ta ve methods, but the same
principles apply to using more than one qualita ve method to carry out an
inves ga on because each brings a par cular kind of insight to a study. For
example, interviews are o en used in combina on with observa on to provide
an understanding of how events or behaviours arise, as well as to reconstruct
perspec ves on their occurrence. Similarly individual interviews and focus
groups are o en used in the same study. For example, focus groups might be
used as an ini al stage to raise and begin to explore relevant issues which will
then be taken forward through in-depth interviews; or might be used a er in-
depth interviews to discuss the issues at a more strategic level. A design
combining, say, individual interviews and some later conversa on analysis, or
ini al documentary analysis with subsequent in-depth interviews might be used
for similar reasons. All these methods and their different contribu ons are
described in detail in the following chapter (Chapter 3). But as with all decisions
about the choice of methods, the objec ves of the study and the nature of the
data required to meet them will be central to the use of two or more qualita ve
approaches. It will also be affected by the epistemological orienta on of the
researcher and their views on the integrity of different methods for inves ga ng
the central phenomena under study.

This chapter has explored some of the many uses of qualita ve research, both
as an independent method of inves ga on and in combina on with quan ta ve
research. There has been a seismic shi in a tudes to qualita ve methods in
recent years, partly as a result of greater apprecia on of what they can do but
also because of a need for greater and more refined understanding of social
issues. The poten al for an ever-widening use of qualita ve methods in applied
research is extensive.

KEY POINTS
• Un l the la er part of the twen eth century the use of qualita ve methods
was much more evident in research that was concerned with developing
social theory than in more applied se ngs. This was par cularly so in social
policy research where there had been some resistance to trea ng qualita ve
research findings as ‘evidence’. Although there has been considerable
growth in the use of qualita ve research within this sector in recent
decades, there are s ll areas where it is underu lised.

• A broad classifica on of the func ons of research is described, based on the

nature of the informa on or understanding it brings. This is categorised as:
contextual research which describes the form or nature of what exists;
explanatory, examining the reasons for, or associa ons between, what
exists; evalua ve, appraising the effec veness of what exists; and genera ve
– aiding the development of theories, strategies or ac ons. Qualita ve
research, like sta s cal enquiry, has a specific role to play in contribu ng to
each of these func ons.

• There are circumstances in which qualita ve research may be the sole or

principal method needed to address a research ques on. These are centrally
related to the nature of the research informa on or evidence required.
There are also factors related to the subject ma er under inves ga on,
specifically where it is ill-defined or not well understood; deeply rooted;
complex; specialist; delicate, intangible or sensi ve.

• The poten al for combining qualita ve and quan ta ve research is

considerable. Several authors have provided useful frames of reference for
op mising the strengths of the two approaches in harness. When using
qualita ve and quan ta ve research in combina on it is important to
 recognise that each offers a different way of knowing about the world, and it
should not be expected that the evidence generated from the two
approaches will replicate each other. Rather, the benefits of mixing methods
may result from the ability of each method to add to a more nuanced
picture of complex social phenomena.

KEY TERMS
Theore cal research is concerned with the aim of tes ng, genera ng or
enhancing theore cal or academic thinking within a par cular discipline.
Applied research is concerned with using the knowledge acquired through
research to contribute directly to the understanding of a contemporary issue.
Applied social research is o en equated with social policy research, which has
the objec ves of developing, monitoring or evalua ng policy and its related
prac ce.

Forma ve evalua ons are designed to provide informa on that will help to
change or improve a programme or policy, either as it is being introduced or
where there are exis ng problems with its implementa on. Summa ve
evalua on is concerned with the impact of an interven on or policy in terms of
effec veness and the different outcomes that have resulted. Qualita ve
methods can contribute to both.

Triangula on involves the use of different methods and sources to check the
integrity of, or extend, inferences drawn from the data. More generally, mixed
methods research will extend understanding through the use of mul ple
perspec ves or providing different types of ‘readings’.

Further reading
Flick, U. (2009) An Introduc on to Qualita ve Research, London: Sage,
especially Chapters 2 and 3.

Silverman, D. (2011) Interpre ng Qualita ve Data, London: Sage, especially

Chapters 1 and 13.

Online resources
 Spencer, L., Ritchie, J., Lewis, J. and Dillon, L. (2003) Quality in Qualita ve
Evalua on: a Framework for Assessing Research Evidence, Government Chief
Social Researcher’s Office, available at: h p://www.hm-
treasury.gov.uk/d/data_magentabook_qqe_suppguidance_181212.pdf (reissued
version published 2012). (accessed 13 January 2013)

Mason, J. (2006) Working Paper: Six Strategies for Mixing Methods and
Linking Data in Social Science Research, ESRC Na onal Centre for Research
Methods, available at:
h p://www.socialsciences.manchester.ac.uk/morgancentre/reali es/wps/
(accessed 13 January 2013)

Sage interview with Alan Bryman – in which he discusses, among other

things, his views on the importance of developing a good knowledge of different
methods and on the recent increased interest in mixed methods research.
Available at: h p://www.youtube.com/watch?v=bHzM9RlO6j0 (accessed 13
January 2013)