의학교육연구에서 포커스그룹의 활용: AMEE Guide No. 91
Using focus groups in medical education research: AMEE Guide No. 91
RENE´ E E. STALMEIJER1, NANCY MCNAUGHTON2 & WALTHER N. K. A. VAN MOOK1,3 1Maastricht University, the Netherlands, 2University of Toronto, Canada, 3Maastricht University Medical Centre, the Netherlands
Focus groups in this Guide are defined as ‘‘. . . group discussions organized to explore a specific set of issues . . . The group is focused in the sense that it involves some kind of collective activity . . . crucially, focus groups are distinguished from the broader category of group interview by the explicit use of the group interaction as research data’’ (Kitzinger 1994, p. 103).
포커스그룹의 정의
A popular data-collection technique used in qualitative research is the ‘‘focus group’’, originally called ‘‘focused group interview’’ which was initially described by Merton & Kendall (1946). Focus groups in this Guide are defined as:
. . . group discussions organized to explore a specific set of issues . . . The group is focused in the sense that it involves some kind of collective activity . . . crucially, focus groups are distinguished from the broader category of group interview by the explicit use of the group interaction as research data (Kitzinger 1994, p. 103).
의학교육에서의 포커스그룹Focus groups in medical education
For example, focus groups have been a method of choice for performing
- needs assessments (e.g. MacDonald et al. 2007; Telner et al. 2008),
- program evaluation (e.g. McIntosh et al. 2008; Stergiopoulos et al. 2010),
- exploratory data collection (e.g. Bombeke et al. 2012; Cleland et al. 2012),
- explanatory data collection (e.g. Smithson et al. 2010; Duvivier et al. 2012), and
- design and validation of questionnaires (e.g. Wade et al. 2012; Riquelme et al. 2013; Strand et al. 2013).
Where historically, focus groups were used as part of a mixed methods approach in which both quantitative and qualitative data was being collected, the use of focus groups as the principal method of investigation has increased in the last decade (e.g. Stalmeijer et al. 2009; Mann et al. 2011; Slootweg et al. 2013) (see Box 1).
포커스그룹 정의하기 Defining focus groups
포커스그룹의 역사
The history of focus groups
Focus groups are generally seen to have emerged in the 1940s when they were first used by Paul Lazarsfeld. The technique was further developed within sociology by Merton & Kendall (1946) during the Second World War to test the reactions of people to propaganda and radio broadcasts. They later grew to be an established research method in the field of marketing and organizational development (Barbour 2007).
Focus groups came into the education realm in the 1970s during a time of growing interest in participatory approaches to carrying out research (Freire 1970).
However, focus groups as a research method of choice did not become prevalent until the mid-1980s (Coˆte´-Arsenault & Morrison-Beedy 2005).
패러다임
Paradigmatic fit
의학교육이 다루는 분야
Medical education as a field of inquiry is committed to pursuing scientific, social, and cultural questions related to medical training and practice as well as issues relevant to the health professions more broadly.
구성주의적 패러다임에 가장 잘 맞음. 포커스그룹연구에서 도출된 지식은 실증주의적 용어인 타당도, 신뢰도, 일반화가능도 등의 개념과 맞지 않음.
Focus groups as a method fit most commonly within a constructivist paradigm which views reality (ontology) as socially negotiated or constructed and knowledge (epistemology) as a product of the social and co-constructed interaction between individuals and society. More importantly, focus groups, as a method of data gathering, fit under a methodological umbrella concerned with how people make meaning from their experiences in the world (phenomenology, see Box 2 and Glossary). The researcher engaging in focus groups is interested in participants’ ideas, interpretations, feelings, actions and circumstances. The knowledge that focus group research produces is therefore not measurable according to such precepts as validity, reliability or generalizability which all belong to ideas and values posited within a positivist paradigm.
포커스그룹 정의내리기 Defining focus groups
포커스그룹은 다음과 같은 특성을 갖는다.
Focus groups involve:
- – a discussion within a (small) group of people is the focus of the research,
- – a discussion within the group is focused on a certain topic,
- – a group led by a researcher/moderator/guide who stimulates active engagement of participants in a discussion,
- – an interaction between group members which is used to gain depth in the exploration of the topic of discussion,
- – an understanding that this interaction is also a focus of the analysis
포커스그룹과 집단면담의 차이
Focus group versus group interview
There is a fundamental difference between the two research techniques with the critical point of distinction being the role of the researcher and his/her relationship to the researched (Smithson et al. 2000 cited in Parker & Tritter 2006, p. 25).
- ‘‘In group interviews the researcher adopts an ‘investigative’ role: asking questions, controlling the dynamics of group discussion, engaging dialogue with specific participants. This is premised on the mechanics of a one-to-one, qualitative, in-depth interview being replicated in a broader (collective) scale’’ (p. 26).
- In a focus group, the researcher takes on a peripheral role acting as a moderator or facilitator; that is, facilitating the group discussion between participants not between her/himself and the participants. ‘‘It is the inter-relational dynamics of the participants that are important, not the relationship between the researcher and the researched (Parker & Tritter 2006, p. 26).
Why and when to use focus groups?
왜 사용하는가 Why use focus groups
As mentioned earlier the main reason for using focus groups is to gather information from different participants’ points of view.
잘 밝혀지지 않은 분야의 연구에 적합(Exploratory)
So, one aim of focus groups is to record, understand and explain the meanings, beliefs and cultures that influence the participants’ feelings, attitudes and behaviors (Rabiee 2004). Focus groups are thus particularly appropriate for exploratory research, i.e. research in poorly understood or ill-defined topics (Kitzinger 1995).
예비 데이터를 확신, 강화하는 용도로 사용가능함 (Explanatory)
A second aim of focus groups is to further strengthen and confirm preliminary data from studies that possibly used other research tools, i.e. an explanatory design study. Although focus groups are more often used for exploratory and explanatory purposes, they can also be used as confirmatory tools (Stewart et al. 2007).
언제 사용하는가 When to use focus groups
연구 시작 전, 연구 중간, 연구 이후에 사용할 수 있음.
Focus groups can be used prior to, during and after other investigations or research. However, since focus groups are particularly appropriate for research in poorly understood or ill-defined topics (Kitzinger 1995), they are frequently used early in a research project, and are often even employed as a starting point, to lay the foundation for subsequent research using other research techniques such as surveys. Focus groups can also be used after other research methods in order to help further explore the data collected, to gather in-depth information or to refine or interpret previously gathered data; in other words to study associations that need clarifications, elaboration or ‘‘salvaging’’ (Powell & Single 1996).
언제 사용하지 않아야 하는가 When not to use focus groups
연구문제가 민감한 문제나 개인적 정보에 대한 것이어서 여러 사람 앞에서 말하기를 꺼려하는 주제일 경우. 예상되는 참가자들이 권력상의 차이로 인해서 포커스그룹동안 거의 말을 하지 않을 것이 예상되는 경우. 이 두 가지 모두 일대일 면담이 낫다.
Cases in which focus groups might not be the best method of data collection are studies in which research questions are directed at gathering potentially sensitive or personal information that people might not want to share within a larger group. Also, studies situated in research settings which are characterized by large power differentials between potential participants are advised not to use focus groups since the power differential might cause participants to stay silent within a focus group setting in fear of repercussions for sharing their opinion. In both cases, one-on-one interviews are preferred over focus groups (Barbour 2007).
Challenges to the focus group method
'견고한 질적 자료'가 부족하다는 우려, 그리고 전체 집단을 대표하지 못한다는 우려
The first concern cited is the lack of ‘‘hard quantitative data’’ produced, and the second relates to the composition of groups that may not necessarily be representative of a larger or the whole population (Stewart et al. 2007).
"다른 방법으로는 도달하지 못하는 곳에 갈 수 있음"
With respect to the first concern, unlike constructivist research approaches, a positivist research perspective seeks quantitative data that can be proven to be ‘‘true’’ and therefore can be reliably applied universally across multiple sites (generalizable). In fact, when properly employed focus groups can ‘‘reach the parts that other methods cannot reach’’ (Kitzinger 1995).
연구자는 수합된 정보의 깊이와 풍부함에 관심을 두게 되며, 이것이 곧 다른 맥락에서도 "진실"임을 의미하진 않음
The researcher is interested in the depth and richness of the information collected and is not suggesting the findings are ‘‘true’’ in other contexts.
질적 연구에서 '대표성'이란 구체적인 맥락과 주제 영역에 대한 것이지 인구집단에 대한 것이 아님.
The second concern about group composition also reflects a misunderstanding about the objectives of qualitative methods in which ‘‘representativeness’’ is tied to the specific contexts and topic areas and not to the representation of a population. These differences will become clearer in the following sections on sampling strategies and decisions about focus group formats. Also raised as a concern is the notion that qualitative research is ‘‘vague’’, or lacks rigor.
포커스그룹을 연구방법으로 선택하는 것을 방법론적, 패러다임적 이해에 맞게 정당화할 수 있어야 한다. 의학교육연구는 더 이상 '무엇을 했는가'만을 기술하는 것으로 충분하지 않다.
Therefore, it is important to be able to rationalize your choice of focus groups as a method according to methodological and paradigmatic understanding. Medical education research has matured and we are no longer in an age when stating what you did is enough to satisfy research standards
준비 Preparing for focus groups
(1) Who are my potential participants and how many should I include?
(2) How should I compose the groups; who should be in them?
(3) How big should the groups be?
(4) How will I compose my questions to explore and answer the key research question
샘플링 Sampling
'질문'과 '현상'에 부합하는 표본을 구성해야 한다. 'what'과 더불어 'how'를 고민해야 한다.
As with all qualitative methods, a sample must ‘‘fit with the question’’ and ‘‘fit with the phenomenon’’ being investigated (Crabtree & Miller 1999). However, as well as considering ‘‘what’’ is sampled the researcher must consider ‘‘how’’ to go about assembling meaningful groups.
질적연구의 표본수집은 집단을 대표하기보다는 집단의 다양성을 대표해야 한다.
‘‘The purpose of qualitative sampling is to reflect the diversity within the group or population under study rather than aspiring to recruit a representative sample’’ (Barbour 2007). The ‘‘focus’’ of focus groups is the emergence of opinions, meanings, feelings, attitudes and beliefs about a topic area and so it is the dynamics within any group as much as the answers provided to questions that will provide the researcher with essential data.
두 가지 방법이 있다.
In this respect, sampling is considered by some qualitative researchers to always be ‘‘purposeful’’. Patton identifies sixteen kinds of purposeful sampling strategies (Crabtree & Miller 1999), however for the purposes of focus groups these can be narrowed down. The two most common approaches are referred to as ‘‘theoretical’’ (Glaser & Strauss 1967; Mays & Pope 1995) sampling and ‘‘purposive’’ (Kuzel 1992) sampling.
이론기반 표본수집 Theoretical sampling
Theoretical sampling is described by Glaser & Strauss (1967) as the ‘‘process of data collection for generating theory whereby the analyst jointly collects, codes and analyses the data making decisions about what data to collect next and where in order to develop theory as it emerges’’.
포커스그룹을 어떻게 구성할것인가에 대한 결정은 포커스그룹을 진행하는 동안 드러나는 개념에 의해서 명확해진다. 귀납적이고 반복적인 전략을 통해서 연구가 진행됨에 따라서 포커스그룹의 구성과 멤버쉽이 바뀔 수 있다.
In other words, decisions about focus group composition serve to further elucidate concepts that emerge during the focus groups themselves. This is an inductive and iterative strategy in which composition and membership in a focus group may change as the research progresses.
목적기반 표본수집 Purposive sampling
목적기반 표본수집에서는 자료가 수집된 이후 비교를 위한 선택 기준을 활용한다. 목적이 무엇인가를 염두에 두고, 샘플은 이 목적에 맞는 집단을 선택하고 그렇지 않은 사람들은 배제한다.
Purposive sampling anticipates the use of selected criteria in making comparisons once the data have been generated (Barbour 2007). It starts with a purpose in mind and the sample is thus selected to include people of interest and exclude those who do not suit the purpose.
목적기반 표본수집을 하면 참가자 수가 늘어나야 한다는 오해를 흔히 하지만, 한 명의 참가자가 여러 기준을 만족시킴으로써 다양성을 확보하면 예상보다 적은 수의 참가자로도 가능하다.
There is a common misunderstanding that purposive sampling necessarily inflates the number of participants involved. However, as Barbour (2005) suggests, each participant may potentially meet several of the desired criteria in terms of diversity making multiple comparisons possible with fewer participants than at first might be apparent (Barbour 2005).
포화 Saturation
It is considered good practice in qualitative research to sample until saturation is achieved. This refers to a time when no new ideas about your topic or problem emerge from the various focus groups. Strauss & Corbin (1998) suggest that ‘‘saturation should be concerned with reaching the point where it becomes ‘‘counter-productive’’ and that ‘‘the new’’ which is discovered does not necessarily add anything to the overall story, model, theory or framework (p. 136).
그룹 구성: 같지만 다르게? Group composition: same but different?
Decisions about heterogeneous versus homogeneous groups as well as issues of power relations within groups all factor into the possibility of gathering rich focus group data. Other considerations include the degree of familiarity among the participants – strangers versus friends, colleagues versus professional peers and the level of compatibility among the participants (Crabtree & Miller 1999). Focus groups are essentially social gatherings in which one’s comfort with sharing is an important consideration. So, as suggested by Crabtree & Miller (1999, p. 115), the best focus group participants ‘‘will have some degree of personal or professional investment in the topic under examination either as a consumer, provider or policy maker’’ ensuring that they will have something to say on the topic under examination.
균질 구성 Homogeneous group composition
참가자들의 '배경'이 유사해야지 그들의 '태도'가 유사해서는 안된다.
‘‘Focus groups should be homogeneous in terms of background and not attitudes’’ (Morgan 1988, cited in Barbour, p. 59).
예상되는 장, 단점
- Similar contexts may also promote a sense of safety in expressing conflicts or concerns (Crabtree & Miller 1999).
- Disadvantages include the possibility of ‘‘group think’’ or the lack of diversity in ideas as well as hidden agendas or power struggles within a group.
비균질 구성 Heterogeneous group composition
장점
As the name implies a heterogeneous sample brings together participants from diverse backgrounds and experience in order to stimulate discussion and provide new insights into the topic area. Introducing a range of differences in a group may facilitate ideas and potentially conflicting perspectives into conversation may inspire group members to consider the topic under discussion in a different light (Crabtree & Miller 1999, p. 115).
장점: 사전에 설정된 가정이 없다, 솔직할 수 있다, 동일한 의견을 따라 움직일 가능성이 낮다, 비밀이 지켜진다.
One of the advantages of heterogeneous group compositions in cases where the participants do not know each other is that everyone comes to the meeting without pre-set assumptions about the other people in the group. Another advantage is that with this anonymity comes the possibility of more candid input on emotional or highly charged topics. A heterogeneous group is also less likely to be swayed toward consensus agreement by a dominant member who they may never see again. Finally, the preservation of confidentiality is more likely in a disparate group of individuals who are unlikely to cross paths.
단점
Furthermore, a clear disadvantage of a diverse group composition is the possibility of power imbalances and lack of respect for differing opinions (Crabtree & Miller 1999, p. 115).
단점
Apart from issues of expertise, one dominant person can effectively destroy a productive and open group dynamic.
익명성이 오히려 단점으로 작용할 수도 있음. 의견을 표현할 때 과격해지거나 다른 사람의 발언을 저해할 수 있다. 그러나 약간의 의견불일치나 긴장은 꼭 나쁜 것만은 아니다.
The very anonymity that allows for the freedom of expressing ones thoughts can also become a destructive and silencing force for the rest of the group requiring sophisticated moderation. However, a little bit of disagreement and tension is not necessarily a bad thing in a focus group and can be used to help clarify what lies beneath opinions and perspectives.
그룹의 수 Number of groups
Most researchers agree that there is no magic number of focus groups for the successful completion of your data collection.
Crabtree & Miller (1999, p. 118) suggest that when focus groups are to be the sole source of data collection a minimum of four to five focus groups is recommended. Barbour suggests that nominal three or four focus groups are advisable if you want to conduct across group analysis looking for patterns and themes.
Focus groups are often singular events with a particular configuration of participants unlikely to be called to meet a second time. However, there are exceptions to this depending on the topic and the overall intent of the study.
그룹의 크기 Size of groups
The optimal size of a focus group is agreed to be between six to ten participants (Morgan 1996; Crabtree & Miller 1999; Barbour 2005; Krueger & Casey 2000) although as with other elements of qualitative research methods this varies depending on the research context and topic area.
그룹의 크기는 다양성이 드러날 만큼 크면서, 개개인의 의견이 충분이 드러날 만큼 작아야 한다. 말하고 싶어하는 사람이 말할 기회를 가지지 못하고 옆사람과 작게 이야기한다는 것은 그룹이 너무 크다는 하나의 신호이다.
Coˆte´-Arsenault & Morrison-Beedy (2005) suggest that group size depends not only on the topic but also on other factors such as gender, and developmental levels of the participants (p. 175). The groups should be large enough to allow for varying opinions and perspectives and small enough to allow each individual to participate fully and be heard (Coˆte´-Arsenault & Morrison- Beedy 2005; Krueger & Casey 2009). When a group exceeds a dozen people there may be a tendency for the group to fragment. Participants who want to speak may not have the opportunity to and so begin sharing their views by whispering with their neighbors. This is always a signal that the group is too large.
그룹이 너무 크면 연구자도 힘들다. 그러나 너무 작으면 참가자가 말을 해야한다는 압박을 느끼고 포커스그룹보다는 집단면담식으로 될 가능성도 크다.
For social science (and health sciences) research Barbour (2007) advocates for a maximum of eight participants per group for a number of reasons. In terms of moderating groups (picking up and exploring new leads as these emerge), she suggests that with the requirement of researchers to identify individual voices, seek clarifications and further explore any differences in views that merge make larger groups exceedingly demanding. Also, in terms of analysis, focus groups are subject to verbatim transcription and detailed and systematic scrutiny meaning that the data set will be rich without being overwhelming. A minimum number of three or four participants is possible (Kitzinger & Barbour 1999; Bloor et al. 2001) and for some topics may be preferable. However, if a group is too small each participant may feel the pressure to speak, turning the session into more of a group interview rather than focus group dynamic (also see later on running a focus group).
세션의 길이 Length of focus group session
1시간~1.5시간
Although there is no hard and fast rule about how long a focus group should run, it is best to plan for between one hour to one and a half hours depending on the topic and the degree of interaction and engagement by the participants. If
2시간은 넘지 않게
However, there is a point of exhaustion for both participants and focus group facilitators so it is not recommended to extend a session more than two hours.
질문 준비하기 Creating questions for your focus groups
We recommend preparing a list of questions that will help you as the researcher and for the moderator to guide the discussion within the focus group. This list is known as a discussion guide, an interview guide, or a questioning route (Krueger & Casey 2009).
In their 2009 guide, Krueger & Casey stress the importance of preparing a questioning route, and suggest the following steps in designing your guide:
First, brainstorm together with a few people that are familiar with and knowledgeable about your subject. The aim of this phase is to explore and then focus in on ‘‘key questions – those questions that will drive the study’’ (p. 52).
질문을 개방형으로 만들고, 단순화, 구어체로 만든다. 문구 뿐만 아니라 질문의 순서도 중요하다. 일반적인 것에서 구체적인 것으로, 긍정적인 것에서 부정적인 것으로, 힌트가 없는 것에서 있는 것으로 하는 것이 좋다.
The next step is to phrase questions so that they are openended, simple and conversational in nature. These types of questions allow participants to decide the direction of their response, decide when to join the conversation and keep the discussion going. Not only phrasing but the sequencing of questions is important. Krueger & Casey (2009) suggest that general questions should come before specific questions, positive questions before negative questions, and un-cued questions before cued questions.
각 질문에 걸릴 시간을 예상하는 것도 중요하다.
In the additional steps described by Krueger & Casey they stress the importance of estimating time needed for responses for each question and the possibility of needing to revise the questions when necessary. Box 5 provides an overview of the categories of questions relevant to ask within focus group research, whilst Box 6 provides special considerations when working with on-line focus groups.
Running a focus group
모더레이터의 역할 The role of moderator
역할
The role of the moderator is a demanding and challenging one, and moderators will need to possess good interpersonal skills, be good listeners, non-judgmental and adaptable.
역할
The main responsibility of the person running a focus group is to facilitate discussion and exchange of ideas between participants.
한 사람이 moderator의 역할을 하고 다른 사람이 기록을 하는 것이 좋다. 또한 그룹간 일관성이 있어야 하므로 역할과 책임을 주의해서 준비해야 한다.
However, it is recommended that just one moderator facilitates and the other takes notes and checks the recording equipment during the meeting. There also needs to be consistency across focus groups, so careful preparation with regard to role and responsibilities is required (Gibbs 1997).
미팅이 시작되면 moderator의 역할이 중요해진다. 그룹의 목적에 대해 명확히 설명하고, 사람들을 편안하게 해주며, 사람들의 상호작용을 촉진해야 한다.
Once a meeting has been arranged, the role of moderator or group facilitator becomes critical, especially in terms of providing clear explanations of the purpose of the group, helping people feel at ease, and facilitating interaction between group members.
논쟁을 유도하기도 한다.
During the meeting, moderators may need to promote debate, perhaps by asking open questions.
모든 사람에게 발언 기회를 줘야 한다. 동시에 지나치게 동의하는 모습을 보여서도 안된다. 그러면참가자들은 moderator의 기분을 맞추려고 할 것이기 때문이다. 개인적 의견을 피력하는 것을 지양해야 한다.
Moderators also have to ensure everyone participates and gets a chance to speak. At the same time, moderators are encouraged not to show too much approval (Krueger 1988), so as to avoid participants’ attempts to please the moderator. Moderators must avoid giving personal opinions so as not to influence participants towards any particular position or opinion.
상반된 역할? Moderator – conflicting roles?
PI가 moderator를 하는 것은 적절치 못하다.
In many circumstances it is not appropriate for the principal investigator to act as the moderator. The danger that lies in this role allocation may be that the principal investigator is too focused on the research question and potential personal theories underlying this question.
그룹이 모르는 사람을 moderator로 지정할 수도 있다.
Another main consideration in the decision to appoint someone unknown to the group as a moderator
Appointing an unknown moderator avoids influencing the discussion based on personal knowledge or experience (Gibbs 1997).
옵저버의 역할 The role of the observer
moderator와 observer를 따로 두는 것이 좋다.
When possible it is a benefit to have both a moderator and an observer (other member from the research team or research assistant) take part in the running of a focus group. An observer offers another set of eyes and ears and is valuable in picking up non-verbal nuances in participant reporting that may be missed by the moderator.
포커스그룹이 끝나면 moderator와 observer가 모여서 추가적인 관찰내용과 관련된 생각들을 공유하는 시간이 필요하다.
Following a focus group, it is good practice to allow time for both the moderator and the observer to jot down additional observations and thoughts related to the focus group which can then be included in follow-up research meetings (Gibbs 1997).
스타일 Moderator styles
두 가지 스타일이 있고 각각이 적합한 연구목적이 있다.
Generally speaking moderators can take on two broad styles, the directive and the non-directive style.
- The directive moderator style is most appropriate when the questions to investigate are numerous and focused (e.g. technical documents, new program or questionnaire to assess) and when we want to better understand intriguing and specific data collected through another process such as a survey or interview (explanatory design). The advantage of a high degree of moderator control is the specificity of the data that emerges.
- The non-directive moderator style is however more suited for exploratory research, i.e. to find new research avenues, brainstorming, or broadening and deepening understanding about the research area.
좋은 포커스그룹의 비밀은 moderator가 주도권을 갖지 않는 것이다.
The secret to a good focus group is that it is not moderator-dominated (Gibbs 1997).
Group dynamics
However, some (more disruptive) group dynamics cannot be predicted. Below are several suggestions for moderators on how to deal with the more disruptive group dynamics.
파괴적인 구성원 Countering dominating or disruptive group members
It’s a good idea to ‘‘go around the group’’ occasionally in order to counter the tendency of the group to accept one person’s view as the group consensus. In the introduction, it is also useful to ask them not to be concerned about agreement with other people in the group.
For example, ‘‘So and so has told us why she feels that medical students should have a strong science training before entering medical school. Does anyone have another view about this?’’
말을 하지 않는 구성원 Shy or silent participants
It is a moderator’s responsibility to ensure that everyone has an opportunity to share their views.
In order to invite participation, it is not out of place to directly ask a participant who has not contributed.
The reasons for silence by certain group members may relate back to group composition and power relations within the group and this should be taken into account when composing your group (Gibbs 1997).
묻는 말에만 대답하는 그룹 Groups in which participants only answer directly to the moderator and do not open up to exchange of ideas with each other
A trick to open up conversation between participants is to cast your eyes around the group when the person who is answering the question is responding.
Analyzing focus group data
The majority of the data is generated when the audio records are transcribed verbatim, but besides that the moderator and the observer have gathered valuable observational data. Therefore, it is important that the moderator and observer debrief after each focus group discussion to share their experiences and add an additional layer of data on the spoken words produced by the participants (also see section titled ‘‘The role of the observer’’).
분석을 위한 팁 General tips for qualitative data analysis
자료의 질 Data quality
The quality of your data analysis is inseparably linked to the quality of your data.
자료의 질은 다음과 같은 요소에 의해서 영향을 받는다.
The quality of focus group data will be reliant on a number of factors.
- First of all, the number of participants in the focus group will be of influence: both too few and too many participants can potentially result in just a shallow discussion.
- Second, the quality of the sampling procedure will be of influence: were the right people invited to answer the research question? Was the group composition favorable to an in-depth discussion?
- Third, the quality of your questions and the questioning route determine data quality. Therefore, focus groups need a preparatory period in which the research team discuss and design the questioning route.
- Finally, the skills of the moderator will determine to what extent relevant topics were sufficiently explored and whether all participants will have been able to have a meaningful contribution in the discussion (Barbour 2007; Krueger & Casey 2009).
분석 소프트웨어? Data analysis software: yes or no?
The quality of your data analysis is determined by the quality of the researcher(s) performing the data analysis and not the quality of the software program that is used to perform the analysis (Kidd & Parshall 2000; Pope & Mays 2009).
목적을 잊지 않기 Keeping your eye on the ball
Given the fact that focus group research produces a lot of data, often 30–50 pages per focus group, it is important that during the analysis you keep your purpose and/or your research question in mind so that you do not get overwhelmed (Krueger & Casey 2009).
‘‘The coding frame should be flexible enough to incorporate themes introduced by focus group participants as well’’ (p. 117).
분석틀 Analytical frameworks
All kinds of analytical frameworks for analyzing focus group data can be used. These frameworks should be aligned with the methodology (e.g. grounded theory and phenomenology) (Creswell 2013) and may also be informed by a specific focus (e.g. discourse analysis and conversation analysis).
연역적 vs 귀납적 Deductive versus inductive data analysis
Analyzing focus group data is an iterative process between at least two researchers or team members involved in the process. One can choose to analyze data deductively or inductively. A deductive approach involves reading your transcripts to which you apply a predetermined set of themes or coding structure
After an initial reading of the transcripts independently by each research team member, the group comes together to compare notes and begin the building process. This cycle of reading and meeting to discuss the data continues until the group is satisfied that they have a coherent story related to the participants’ views on the topic or issue under study. As a note, there will be information that you cannot ‘‘put’’ anywhere. The idea or comment(s) may sit outside the rest of the themes or codes. This is important. Do not throw out or eliminate data because it does not fit; save it somewhere so that you can come back to it at a later date (Krueger & Casey 2009).
Focus group specific tips for qualitative data analysis
자료분석의 종류 Types of data analysis relevant for focus group data
This richness is partly connected to the fact that focus groups produce three levels of data:
(1) data about individuals,
(2) data about the group discussion and
(3) data about group interaction (Onwuegbuzie et al. 2009).
All these levels of data are potential avenues for analysis yet few focus group studies pay attention to all levels of data.
속기록에 숨겨진 내용을 보기 Look beyond the transcripts
다음과 같은 내용들을 보아야 함.
What the researcher should not overlook is all other potential data sources surrounding focus group research.
- Firstly, when working together with an observer the potential to collect observational data about the group interaction is present.
- Secondly, it is usually worthwhile to use a small questionnaire to collect demographic data of the participants. In this way, the time of participants and the moderator is optimally used and the questionnaire provides an additional data source.
- Thirdly, the observer could also record non-verbal communication by participants and the interaction between participants to provide an additional dimension to the data transcription and interpretation.
- Fourthly, when performing multiple focus groups, researchers might decide to perform analysis where they compare the discussion between groups but also focus on the discussion within a single group.
합의와 의견불일치, 어디서 오는가 Consensus and disagreement and where it comes from
An important aspect of analyzing focus group data is identifying the extent to which agreement or disagreement occurred within the group and how perspectives arose or were modified within the group process (Kidd & Parshall 2000).
침묵도 자료다 Silences are also data
동의의 표현일수도 있지만, 잘 몰라서 침묵하는 것 일수도 있음.
An underused type of data is the presence of silences within the focus group discussion. Silence could indicate several things, e.g. consensus about a certain topic but also nonfamiliarity with an issue. It is therefore worthwhile to analyze at what points in the data silences arose and to supplement this with observational data to get a more in-depth understanding of the nature of the silence
Quality and ethics in focus group research
Quality of focus group research
다음의 네 가지 요소를 갖춰야 함.
Good qualitative research should be credible, transferable, dependable and confirmable (Frambach et al. 2013).
Credibility
‘‘Credibility is the extent to which the study’s findings are trustworthy and believable to others’’. Practices that are described to ensure the credibility of a study are the use of data, method, and/or researcher triangulation (see Glossary), a prolonged engagement with the data and member checking (see Glossary). Depending on the aim of the focus group study, the relevance of a member check might differ
Transferability
The transferability of a study is determined by the extent to which its findings can be transferred to another context. In order for an audience to judge transferability, researchers are advised to produce thick descriptions (Glossary) of the context under study, to explain the sampling strategy used, and to discuss the extent to which the finding of the study resonate with empirical and theoretical work already published.
Dependability
‘‘Dependability is the extent to which the findings are consistent in relation to the contexts in which they were generated’’. This means that researchers will continue to collect data in a given setting until ‘‘saturation is met’’ (Glossary) indicating that no new themes resulted from the inquiry. This requires that collection and analysis go hand in hand as to assure meaningful and in-depth data collection towards answering the research question (iterative data collection and analysis).
Confirmability
To demonstrate confirmability of the research, the researcher needs to provide insight into how he/she came to certain decisions and conclusions during the research process (audit trail). The concern here is that the participants and settings were not the main source of the findings but the researchers’ potential biases. The researchers therefore need to show reflexivity (Glossary). Furthermore, the researcher needs to consciously search for data and literature that might disconfirm the findings and also discuss the findings with peer researchers (peer debriefing).
Ethics in focus group research
‘‘Ethics or moral philosophy involves systematizing, defending, and recommending concepts of right and wrong conduct’’ (Fieser 2009). When applying this to the field of qualitative research this means protecting the interests of the participants on the one hand, without compromising the aim of the research data for the good of others on the other (Orb et al. 2001)
They also postulated that the paucity of literature on ethics maybe due to the assumption that qualitative research is harmless to the participants, and they noted that medical research committees sometimes have difficulty making judgment on research proposals submitted for their judgment (Gauld & McMillan 1999; Morse 2001; Richards & Schwartz 2002).
잠재적 위험 Potential risks in qualitative research
Principally, participants are fully autonomous, and usually share information on a voluntary basis. A balanced relationship between researcher and participant facilitates disclosure, trust, and awareness of as well as respect for potential ethical issues (Orb et al. 2001). Nevertheless, it can easily be envisaged that when probing into rationales during focus groups it is difficult to avoid touching upon issues that may provoke anxiety and distress in certain participants.
‘‘Old wounds’’ may unexpectedly open (Orb et al. 2001). However, some anticipation regarding problematic focus groups scenarios is necessary and influences the composition of the groups (Barbour 2007).
연구자의 직업적 배경 뿐만 아니라 개인적 배경(성별, 연령, 인종, 사회계층)도 중요하게 영향을 줄 수 있다.
Evidently, the professional background of a researcher can also impact on the focus group, but personal characteristics (such as gender, age, ethnicity, and social class) are considered equally important (Richards & Emslie 2000).
Writing up focus group research
‘‘Qualitative researchers today acknowledge that the writing of a text cannot be separated from the author’’ (Creswell 2013). Therefore the authors of a qualitative research study need to make clear how they were involved in the research and why certain decisions were made
Methods section
The methodology and focus group rationale
To demonstrate credibility and trustworthiness of your data the choices for methodology, design and focus groups as a method given your research question need to be explained. If multiple methods for data collection were used their intended purpose in the research should be described.
Describing
For reporting of qualitative research it is important to “paint the picture” of where, how and from whom data were collected. This is necessary for the reader to be able to judge the transferability of the results to their own setting (Denzin & Lincoln 2005). This includes reporting how participants were chosen (sample), recruited and identified for characteristics that made them valuable for answering your research questions. Furthermore, the number and characteristics of the group composition should be explain; homo- or heterogeneous, number of groups, number of participants per group and how long the discussions lasted.
Further consideration should be given to the extent the discussion was structured, semi-structured on not structured, as well as how the questioning route was designed and used. These explicit descriptions of the research process help readers to paint the outlines of the context in which the data were gathered (Morgan 1996).
Who performed the research?
Not only the characteristics of the participants but also of those of the researchers involved in the study, the moderator and the observer/research assistant should be described. Paradigmatically speaking, qualitative research acknowledges the influence that the researcher has on the research process (Bunniss & Kelly 2010; Bergman et al. 2012). Therefore, it is important for the audience to know who performed what part of the research and what their backgrounds are. The next step is that the research team reflects on the influence that they might have had on data collection and data analysis, this process is called reflexivity (Malterud 2001).
Analysis
With regard to the analyses, the researchers need to describe which data analysis procedures they used and which principles informed their analysis, as well as who was involved in the process, to what extent theory was used to inform data analysis, and to what extent member checking was applied (Malterud 2001). As mentioned earlier, it is important to identify your methodology and how your use of focus groups is supported by the underlying precepts. If there are guidelines that informed your research design, these will assist in the analysis of your data. Finally, if software for data analysis was used, the software package and its version should be mentioned.
Results section
Presentation of quotes: do’s and don’ts
Depending the methodology, the role the focus groups had in the research design, the type of data analysis you chose to apply, and the word-limit provided by the journal, various presentations of the results are possible. By presenting verbatim quotes, the researcher gives the audience insight into “the data from which the patterns and constructs arose during analysis” (Holloway & Wheeler 2010). Richardson (1990) describes three types of quote presentation: (1) short eye-catching quotes indicating a short paragraph from the transcript demonstrating a theme, (2) embedded quotes, are short(er) in-text quotes and (3) longer quotations. Especially, the latter is very much dependent on the style of the journal. For focus group research, it can be valuable to both demonstrate quotes from individuals in the group but also group interactions showing how the discussion between participants evolved.
Visual representations
Depending on the methodology and the findings researchers might decide to present a visual depiction of their findings. Grounded theory, e.g. aims to build a theory grounded in the data. A visualization of the concepts represented within this theory might help the audience to get a better overview of the interaction of the various themes within the theory.
Discussion section
The aim of this section is to reflect on the results in the light of already published empirical and/or theoretical work. As such, the researcher tries to contribute to the knowledge within the field. With the discussion, the researcher might also demonstrate the transferability and confirmability of the research. Therefore, often one will see qualitative papers where reflexivity is both part of the “Methods” section and of the “Discussion” section.
Confirmability: The extent to which the findings are based on the study’s participants and settings instead of researchers’ biases.
Credibility: The extent to which the study’s findings are trustworthy and believable to others.
Deductive analysis: Reading your transcripts to which you apply a predetermined set of themes or coding structure.
Dependability: The extent to which the findings are consistent in relation to the contexts in which they were generated.
Epistemology: Theory of knowledge. What are the origin, nature, and limits of knowledge about reality?
Inductive analysis: Reading your transcripts for emerging themes and trying to articulate what concept/definition/meanings of the main topic arises from the data.
Methodology: Strategic approach to answer the research question and to gain knowledge. What is the research design?
Grounded theory: Systematic, qualitative procedure used to generate a theory that explains, at a broad conceptual level, a process, an action, or an interaction about a substantive topic
Ethnography: “(…)The study of social interactions, behaviors, and perceptions that occur within groups, teams, organizations, and communities” (Reeves et al. 2008)
Phenomenology: “A philosophy which explores the meaning of individuals’ lived experience through their own description. The research approach adopted is based on this philosophy” (Holloway & Wheeler 2010)
Action research: “A cyclical approach to research in which researchers are, or collaborate with, practitioners to effect change or use an intervention, evaluate it and modify their practice in the light of evaluation. The process goes on until the optimum situation has been achieved” (Holloway & Wheeler 2010)
Mixed methods: The collection, analysis and integration of both qualitative and quantitative data in a single study.
Ontology: Theory of the view on reality. What is the nature of physical and social reality?
Paradigm: An interpretative framework, which is guided by “a set of beliefs and feelings about the world and how it should be understood and studied” (Guba 1990).
Positivism: A paradigm which aims to find general laws and regularities based on observation and experiment parallel to the methods of the natural sciences (there is one truth and it can be observed) (Holloway & Wheeler 2010).
Post-positivism: Paradigm stating that there is one truth but it can never be truly observed. Pays attention to falsification and probabilities (Creswell 2013).
Critical theory: Paradigms which aims to critique and change society as a whole, aimed at factors that constrain and exploit individuals (Illing 2007).
Constructivism: Paradigm which states that knowledge and all meaning is not discovered but socially constructed. Meaning is not created but constructed out of the world that is already there (Illing 2007).
Purposive sampling: “Sampling individuals and sites for study which are thought to purposefully inform an understanding of the research problem and central phenomenon in the study” (Creswell 2013).
Reflexivity: An attitude of attending systematically to the context of knowledge construction, especially to the effect of the researcher, at every step of the research process (Malterud 2001; Mauthner & Doucet 2003).
Saturation: Also known as “informational redundancy” (Lincoln & Guba 1985) indicating that everything of importance to the research agenda of the project has been obtained. We can speak of data saturation (sampling to redundancy) and theoretical saturation (no new concepts or dimensions are emerging during data analysis).
Transferability: How well the study’s findings inform medical education contexts that differ from that in which the original study was undertaken.
Theoretical sampling: “sampling individuals or texts whom the researchers predict (based on theoretical models or previous research) would add new perspectives to those already represented in the sample” (Kuper et al. 2008).
Triangulation: Using different perspectives on the same research question to either validate findings or provide a richer understanding of the topic at hand. Examples of triangulation are: (1) methods triangulation (using several methods to answer the same research question), (2) theoretical triangulation (using several theoretical frameworks to create a broader understanding of the findings) and (3) researcher triangulation (multiple researchers playing a part in data collection and or analysis) (Flick 2004; Walsh 2013).
Using focus groups in medical education research: AMEE Guide No. 91.
Author information
- 1Maastricht University , the Netherlands .
Abstract
'Articles (Medical Education) > ⓧ의학교육연구(From Misc.)' 카테고리의 다른 글
| 섬과 군도: 프로그램적 연구와 기회주의적 연구의 조화(Perspect Med Educ, 2020) (0) | 2021.08.27 |
|---|---|
| Likert 스케일, 측정의 수준과 통계의 '법칙' (Adv in Health Sci Educ, 2010) (0) | 2015.04.23 |
| 우리나라 보건의료에서 일차의료 연구의 필요성과 활성화 방안(JKMA) (0) | 2014.12.04 |
| 공공기관 자료의 의학연구 활용원칙과 윤리지침 (JKMA) (0) | 2014.12.04 |
| 의학 학술지 및 종합 데이터베이스 누리집 이용에서 새로운 개념과 출판윤리의 이해(JKMA) (0) | 2014.12.01 |