Filetype PDF | Posted on 20 Aug 2022 | 4 years ago
Authentication
digital resources
336x Filetype PDF File size 0.49 MB Source: cebma.org
Methods of data collection IN BRIEF
• Interviews and focus groups are the most
common methods of data collection used PR
in qualitative research: in qualitative healthcare research
• Interviews can be used to explore A
the views, experiences, beliefs and C
interviews and focus groups motivations of individual participants TI
• Focus group use group dynamics to CE
generate qualitative data
1 2 3 4
P. Gill, K. Stewart, E. Treasure and B. Chadwick
This paper explores the most common methods of data collection used in qualitative research: interviews and
focus groups. The paper examines each method in detail, focusing on how they work in practice, when their use
is appropriate and what they can offer dentistry. Examples of empirical studies that have used interviews or focus
groups are also provided.
INTRODUCTION The purpose of this paper is to explore are usually very timeconsuming (often
Having explored the nature and pur these two methods in more detail, in lasting several hours) and can be diffi cult
pose of qualitative research in the previ particular how they work in practice, to manage, and to participate in, as the
ous paper, this paper explores methods the purpose of each, when their use is lack of predetermined interview ques
of data collection used in qualitative appropriate and what they can offer tions provides little guidance on what
research. There are a variety of methods dental research. to talk about (which many participants
of data collection in qualitative research, QUALITATIVE RESEARCH find confusing and unhelpful). Their use
including observations, textual or visual INTERVIEWS is, therefore, generally only considered
analysis (eg from books or videos) and where significant ‘depth’ is required, or
1
interviews (individual or group). How There are three fundamental types of where virtually nothing is known about
ever, the most common methods used, research interviews: structured, semi the subject area (or a different perspec
particularly in healthcare research, are structured and unstructured. Structured tive of a known subject area is required).
interviews and focus groups.2,3 interviews are, essentially, verbally Semistructured interviews consist of
administered questionnaires, in which several key questions that help to defi ne
a list of predetermined questions are the areas to be explored, but also allows
asked, with little or no variation and the interviewer or interviewee to diverge
QUALITATIVE RESEARCH with no scope for followup questions to in order to pursue an idea or response
2
IN DENTISTRY responses that warrant further elabora in more detail. This interview format
1. Qualitative research in dentistry tion. Consequently, they are relatively is used most frequently in healthcare,
2. Methods of data collection in qualitative quick and easy to administer and may be as it provides participants with some
research: interviews and focus groups of particular use if clarification of cer guidance on what to talk about, which
3. Conducting qualitative interviews with tain questions are required or if there are many find helpful. The fl exibility of
school children in dental research l i k e l y t o b e l i t e r a c y o r n u m e r a c y p r o b l e m s this approach, particularly compared to
4. Analysing and presenting qualitative data with the respondents. However, by their structured interviews, also allows for
very nature, they only allow for limited the discovery or elaboration of informa
participant responses and are, therefore, tion that is important to participants but
1* Senior Research Fellow, Faculty of Health, Sport of little use if ‘depth’ is required. may not have previously been thought of
and Science, University of Glamorgan, Pontypridd, Conversely, unstructured interviews as pertinent by the research team.
2
CF37 1DL; Research Fellow, Academic Unit of Primary do not reflect any preconceived theories For example, in a recent dental public
Care, University of Bristol, Bristol, BS8 2AA, 3Dean
4 5
and Professor of Dental Public Health; Professor of or ideas and are performed with little or heath study, school children in Cardiff,
Paediatric Dentistry, School of Dentistry, Dental Health 4
and Biological Sciences, School of Dentistry, Cardiff no organisation. Such an interview may UK were interviewed about their food
University, Heath Park, Cardiff, CF14 4XY simply start with an opening question choices and preferences. A key fi nd
*Correspondence to: Dr Paul Gill such as ‘Can you tell me about your expe ing that emerged from semistructured
Email: PWGill@glam.ac.uk
rience of visiting the dentist?’ and will interviews, which was not previously
Refereed Paper then progress based, primarily, upon the thought to be as highly infl uential as the
DOI: 10.1038/bdj.2008.192
©British Dental Journal 2008; 204: 291-295 initial response. Unstructured interviews data subsequently confi rmed, was the
BRITISH DENTAL JOURNAL VOLUME 204 NO. 6 MAR 22 2008 291
© 2008 Nature Publishing Group
PRACTICE
significance of peerpressure in infl uenc between the two, such as the purpose of a oneoff or, if change over time is of
4
ing children’s food choices and prefer the encounter, reasons for participating, interest, repeated basis, for example
ences. This finding was also established roles of the people involved and how the exploring the psychosocial impact of
primarily through followup question interview is conducted and recorded.8 oral trauma on participants and their
ing (eg probing interesting responses The primary purpose of clinical subsequent experiences of cosmetic den
with followup questions, such as ‘Can encounters is for the dentist to ask the tal surgery.
you tell me a bit more about that?’) and, patient questions in order to acquire suf Developing the interview
therefore, may not have emerged in the ficient information to inform decision
same way, if at all, if asked as a prede making and treatment options. However, Before an interview takes place,
termined question. the constraints of most consultations respondents should be informed about
The purpose of research interviews are such that any openended question the study details and given assur
ing needs to be brought to a conclusion ance about ethical principles, such as
2 2
The purpose of the research interview within a fairly short time. In contrast, anonymity and confi dentiality. This
is to explore the views, experiences, the fundamental purpose of the research gives respondents some idea of what to
beliefs and/or motivations of individu interview is to listen attentively to what expect from the interview, increases
als on specific matters (eg factors that respondents have to say, in order to the likelihood of honesty and is also a
influence their attendance at the den acquire more knowledge about the study fundamental aspect of the informed
9
tist). Qualitative methods, such as inter topic. Unlike the clinical encounter, it consent process.
views, are believed to provide a ‘deeper’ is not to intentionally offer any form of Wherever possible, interviews should
understanding of social phenomena than help or advice, which many researchers be conducted in areas free from distrac
would be obtained from purely quanti have neither the training nor the time tions and at times and locations that are
1 for. Research interviewing therefore
tative methods, such as questionnaires. most suitable for participants. For many
Interviews are, therefore, most appro requires a different approach and a dif this may be at their own home in the
priate where little is already known ferent range of skills. evenings. Whilst researchers may have
about the study phenomenon or where The interview less control over the home environment,
detailed insights are required from familiarity may help the respondent to
individual participants. They are also When designing an interview schedule relax and result in a more productive
particularly appropriate for exploring it is imperative to ask questions that 9 Establishing rapport with
interview.
sensitive topics, where participants may are likely to yield as much information participants prior to the interview is
not want to talk about such issues in a about the study phenomenon as possible also important as this can also have a
group environment. and also be able to address the aims and positive effect on the subsequent devel
Examples of dental studies that have objectives of the research. In a qualita opment of the interview.
collected data using interviews are tive interview, good questions should When conducting the actual interview
‘Examining the psychosocial process be openended (ie, require more than a it is prudent for the interviewer to famil
6 yes/no answer), neutral, sensitive and iarise themselves with the interview
involved in regular dental attendance’
2
and ‘Exploring factors governing den understandable. It is usually best to schedule, so that the process appears
7 Gibson et start with questions that participants
tists’ treatment philosophies’. more natural and less rehearsed. How
6
al. provided an improved understanding can answer easily and then proceed to ever, to ensure that the interview is as
of factors that influenced people’s regu more difficult or sensitive topics.2 This productive as possible, researchers must
lar attendance with their dentist. The can help put respondents at ease, build possess a repertoire of skills and tech
7
study by Kay and Blinkhorn provided up confidence and rapport and often niques to ensure that comprehensive
a detailed insight into factors that infl u generates rich data that subsequently and representative data are collected
enced GDPs’ decision making in relation develops the interview further. 10 One of the most
during the interview.
to treatment choices. The study found As in any research, it is often wise to important skills is the ability to listen
that dentists’ clinical decisions about first pilot the interview schedule on sev attentively to what is being said, so that
treatments were not necessarily related eral respondents prior to data collection participants are able to recount their
8
to pathology or treatment options, as proper. This allows the research team to experiences as fully as possible, without
was perhaps initially thought, but also establish if the schedule is clear, under unnecessary interruptions.
involved discussions with patients, standable and capable of answering the Other important skills include adopting
patients’ values and dentists’ feelings of research questions, and if, therefore, open and emotionally neutral body lan
self esteem and conscience. any changes to the interview schedule guage, nodding, smiling, looking inter
There are many similarities between are required. ested and making encouraging noises
The length of interviews varies 2 The
clinical encounters and research inter (eg, ‘Mmmm’) during the interview.
views, in that both employ similar inter depending on the topic, researcher strategic use of silence, if used appro
personal skills, such as questioning, and participant. However, on average, priately, can also be highly effective at
conversing and listening. However, there healthcare interviews last 2060 min getting respondents to contemplate their
are also some fundamental differences utes. Interviews can be performed on responses, talk more, elaborate or clarify
292 BRITISH DENTAL JOURNAL VOLUME 204 NO. 6 MAR 22 2008
© 2008 Nature Publishing Group
PRACTICE
particular issues. Other techniques that use in public sector marketing, such as researcher gives due consideration to the
can be used to develop the interview the assessment of the impact of health impact of group mix (eg, how the group
13
further include refl ecting on remarks education campaigns. However, focus may interact with each other) before the
group techniques, as used in public 14
made by participants (eg, ‘Pain?’) and focus group proceeds.
probing remarks (‘When you said you and private sectors, have diverged over Interaction is key to a successful focus
were afraid of going to the dentist what time. Therefore, in this paper, we seek to group. Sometimes this means a pre
9 existing group interacts best for research
did you mean?’). Where appropriate, it describe focus groups as they are used in
is also wise to seek clarifi cation from academic research. purposes, and sometimes stranger
respondents if it is unclear what they When focus groups are used groups. Preexisting groups may be eas
mean. The use of ‘leading’ or ‘loaded’ ier to recruit, have shared experiences
questions that may unduly infl uence Focus groups are used for generating and enjoy a comfort and familiarity
responses should always be avoided information on collective views, and the which facilitates discussion or the abil
(eg, ‘So you think dental surgery wait meanings that lie behind those views. ity to challenge each other comfortably.
ing rooms are frightening?’ rather than They are also useful in generating a rich In health settings, preexisting groups
‘How do you find the waiting room at the understanding of participants’ experi can overcome issues relating to disclo
12
dentists?’). ences and beliefs. Suggested criteria sure of potentially stigmatising status
13
At the end of the interview it is impor for using focus groups include: which people may fi nd uncomfortable in
tant to thank participants for their time • As a standalone method, for research stranger groups (conversely there may
and ask them if there is anything they relating to group norms, meanings be situations where disclosure is more
would like to add. This gives respond and processes comfortable in stranger groups). In other
ents an opportunity to deal with issues • In a multimethod design, to explore research projects it may be decided that
that they have thought about, or think a topic or collect group language or stranger groups will be able to speak
are important but have not been dealt narratives to be used in later stages more freely without fear of repercussion,
9
with by the interviewer. This can often • To clarify, extend, qualify or and challenges to other participants may
lead to the discovery of new, unantici challenge data collected through be more challenging and probing, lead
pated information. Respondents should other methods 13
ing to richer data.
also be debriefed about the study after • To feedback results to research Group size is an important considera
the interview has fi nished. participants. tion in focus group research. Stewart and
14
All interviews should be tape recorded Shamdasani suggest that it is better to
12
and transcribed verbatim afterwards, as Morgan suggests that focus groups slightly overrecruit for a focus group
this protects against bias and provides a should be avoided according to the fol and potentially manage a slightly larger
permanent record of what was and was lowing criteria: group, than underrecruit and risk hav
not said.8 It is often also helpful to make • If listening to participants’ views ing to cancel the session or having an
‘field notes’ during and immediately generates expectations for the out unsatisfactory discussion. They advise
after each interview about observations, come of the research that can not that each group will probably have two
thoughts and ideas about the inter be fulfi lled nonattenders. The optimum size for
view, as this can help in data analysis • If participants are uneasy with each a focus group is six to eight partici
4,8 other, and will therefore not discuss
process. pants (excluding researchers), but focus
FOCUS GROUPS their feelings and opinions openly groups can work successfully with as
• If the topic of interest to the few as three and as many as 14 partici
Focus groups share many common fea researcher is not a topic the pants. Small groups risk limited discus
tures with less structured interviews, but participants can or wish to discuss sion occurring, while large groups can
there is more to them than merely col • If statistical data is required. be chaotic, hard to manage for the mod
lecting similar data from many partici Focus groups give depth and erator and frustrating for participants
pants at once. A focus group is a group insight, but cannot produce who feel they get insuffi cient opportu
discussion on a particular topic organ useful numerical results. 13
nities to speak.
ised for research purposes. This discus Conducting focus groups: Preparing an interview schedule
sion is guided, monitored and recorded group composition and size
by a researcher (sometimes called a Like research interviews, the interview
11,12 The composition of a focus group needs schedule for focus groups is often no
moderator or facilitator).
Focus groups were first used as a great care to get the best quality of dis more structured than a loose schedule of
research method in market research, cussion. There is no ‘best’ solution to topics to be discussed. However, in pre
originating in the 1940s in the work of group composition, and group mix will paring an interview schedule for focus
always impact on the data, according to 14 sug
the Bureau of Applied Social Research groups, Stewart and Shamdasani
at Columbia University. Eventually the things such as the mix of ages, sexes and gest two general principles:
success of focus groups as a marketing social professional statuses of the par 1. Questions should move from general
tool in the private sector resulted in its ticipants. What is important is that the to more specifi c questions
BRITISH DENTAL JOURNAL VOLUME 204 NO. 6 MAR 22 2008 293
© 2008 Nature Publishing Group
PRACTICE
2. Question order should be relative to differences of opinions to be discussed should be identified in a way that makes
importance of issues in the research fairly and, if required, encourage reti it possible to follow the contributions
13
agenda. cent participants. of each individual. Sometimes obser
Other relevant factors vational notes also need to be described
There can, however, be some confl ict in the transcripts in order for them to
between these two principles, and trade The venue for a focus group is impor make sense.
offs are often needed, although often tant and should, ideally, be accessible, The analysis of qualitative data is
discussions will take on a life of their comfortable, private, quiet and free from explored in the final paper of this series.
13
own, which will influence or determine distractions. However, while a central However, it is important to note that the
the order in which issues are covered. location, such as the participants’ work analysis of focus group data is different
Usually, less than a dozen predeter place or school, may encourage attend from other qualitative data because of
mined questions are needed and, as with ance, the venue may affect participants’ their interactive nature, and this needs
research interviews, the researcher will behaviour. For example, in a school set to be taken into consideration during
also probe and expand on issues accord ting, pupils may behave like pupils, and analysis. The importance of the con
ing to the discussion. in clinical settings, participants may text of other speakers is essential to the
Moderating be affected by any anxieties that affect understanding of individual contribu
13
them when they attend in a patient role. tions. For example, in a group situa
Moderating a focus group looks easy Focus groups are usually recorded, tion, participants will often challenge
when done well, but requires a complex often observed (by a researcher other each other and justify their remarks
set of skills, which are related to the fol than the moderator, whose role is to because of the group setting, in a way
15
lowing principles: observe the interaction of the group to that perhaps they would not in a one
• Participants have valuable views enhance analysis) and sometimes vide toone interview. The analysis of focus
and the ability to respond actively, otaped. At the start of a focus group, a group data must therefore take account
positively and respectfully. Such an moderator should acknowledge the pres of the group dynamics that have gener
approach is not simply a courtesy, but ence of the audio recording equipment, ated remarks.
will encourage fruitful discussions assure participants of confi dential Focus groups in dental research
• Moderating without participating: a ity and give people the opportunity to
moderator must guide a discussion withdraw if they are uncomfortable with Focus groups are used increasingly in
rather than join in with it. Express 14 dental research, on a diverse range of
being taped.
ing one’s own views tends to give A good quality multidirectional 16 illuminating a number of areas
topics,
participants cues as to what to say external microphone is recommended relating to patients, dental services and
(introducing bias), rather than the for the recording of focus groups, as the dental profession. Addressing a spe
confidence to be open and honest internal microphones are rarely good cial needs population diffi cult to access
about their own views enough to cope with the variation in vol and sample through quantitative meas
13 17
• Be prepared for views that may be ume of different speakers. If observers ures, Robinson et al. used focus groups
unpalatably critical of a topic which are present, they should be introduced to to investigate the oral healthrelated
may be important to you participants as someone who is just there attitudes of drug users, exploring the
• It is important to recognise that to observe, and sit away from the dis priorities, understandings and barriers
researchers’ individual characteristics 14 Videotaping will require more 18
cussion. to care they encounter. Newton et al.
mean that no one person will always than one camera to capture the whole used focus groups to explore barriers to
be suitable to moderate any kind of group, as well as additional operational services among minority ethnic groups,
group. Sometimes the characteristics personnel in the room. This is, therefore, highlighting for the fi rst time differ
that suit a moderator for one group very obtrusive, which can affect the ences between minority ethnic groups.
will inhibit discussion in another spontaneity of the group and in a focus Demonstrating the use of the method
• Be yourself. If the moderator is com group does not usually yield enough with professional groups as subjects in
19
fortable and natural, participants will additional information that could not be dental research, Gussy et al. explored
feel relaxed. captured by an observer to make vide the barriers to and possible strategies for
15 developing a shared approach in preven
otaping worthwhile.
The moderator should facilitate group The systematic analysis of focus tion of caries among preschoolers. This
discussion, keeping it focussed without group transcripts is crucial. However, mixed method study was very important
leading it. They should also be able to the transcription of focus groups is more as the qualitative element was able to
prevent the discussion being domi complex and time consuming than in explain why the clinical trial failed, and
nated by one member (for example, by onetoone interviews, and each hour this understanding may help research
emphasising at the outset the impor of audio can take up to eight hours to ers improve on the quantitative aspect
tance of hearing a range of views), transcribe and generate approximately of future studies, as well as making a
ensure that all participants have 100 pages of text. Recordings should be valuable academic contribution in its
ample opportunity to contribute, allow transcribed verbatim and also speakers own right.
294 BRITISH DENTAL JOURNAL VOLUME 204 NO. 6 MAR 22 2008
© 2008 Nature Publishing Group
no reviews yet
Please Login to review.
