What is the method used to choose members of a group to participate in an experiment?

If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Contents -

4.1 Introduction

During the planning phase you thought about which community members would best be able to provide the information you want, or, if you are looking at issues within your organisation or department, which staff members can provide the information. As well as these questions there are many other decisions you will need to make when selecting your participants. This section provides you with a list of issues that you will need to consider before making the final decisions regarding study participants.

4.2 Method of selection (sampling)

There are many methods for selecting your participants, and the type of sampling will depend on how you will use the information. Focus group results cannot usually be used to describe how an entire population would respond to the same questions, so the type of sampling used in studies designed to describe whole populations is not really necessary (for a discussion of these methods, just refer to a simple text on sampling for epidemiological studies, e.g., Smith and Morrow, 1991).

The common (and simplest) method for selecting participants for focus groups is called "purposive" or "convenience" sampling. This means that you select those members of the community who you think will provide you with the best information. It need not be a random selection; indeed, a random sample may be foolish. For example, if you are investigating why leprosy patients do not always present for medication, it would seem more "convenient" and more useful to select those patients, relatives and staff involved in the leprosy programme. A random sample of the whole community may not provide you with a single person with leprosy!

4.3 Who can provide the best information?

Do not forget to think carefully about all aspects of the problem and be creative when deciding who can provide you with the best information. People in positions of power and authority, or with technical skills, are not necessarily the best people to talk to if you are interested in community attitudes and beliefs. Sometimes people who are less obvious can be extremely useful. Try to think of all the members of your community that could have some knowledge or influence on the problem.

If you do not understand the community well enough to know who can be of most use, do not be afraid to ask local health staff, local leaders or simply members of the community that you have access to. Never rely only on your own ideas about a problem, particularly where you are studying people's attitudes and beliefs. You could be on the wrong track completely by viewing things from your own experiences.

4.4 What will the composition be in each focus group?

As focus groups are discussions among people with similar characteristics, it is important to ensure that participants in any one group have something in common with each other. The reason for this is simple. People talk more openly if they are in a group of people who share the same background or experiences. For example, suppose you are interested in sexual practices in a project concerned with community education to prevent HIV/AIDS. A group that included both young single women and older married women might not be very successful; the young women may feel obliged to discuss "acceptable" practices rather than their true range of experiences and behaviours. Participants with different backgrounds and experience can restrict the openness of discussion within the group. Given this, you need to think about the status of participants in the community, their socio-economic status, educational background, religion, sex, age, and so on, considering which characteristics might most influence a free and natural discussion.

4.5 How many groups are necessary?

In general, once the focus groups cease to provide you with new information, then you do not need to conduct any more sessions. Sometimes this may occur after only two or three sessions with each grouping of participants; sometimes you may need to run six, seven or more before you are satisfied. If this is the first time your team has used focus groups, then you need to allow also for a few practice sessions that may not provide you with the quality of information you require.

You should group "types" of people together. This is probably obvious, but nevertheless worth mentioning. Say, for example, in a study of leprosy, you have identified as target groups for focus group discussions local health workers, traditional healers, adult patients, caretakers of young people with leprosy, and other members of families with leprosy patients. It would be most appropriate to conduct focus groups separately for each group.

However, do not get too complicated in your selection process. This is a very easy mistake to make! In the above example, you already have identified five separate groups of participants. If you now decided that sex, education and residence might all inhibit discussion, and so decided to interview women and men separately, to interview those with and without formal education, and to interview rural and urban dwellers separately, and you aim to hold three focus groups for each group of participants, you'd end up with 120 focus groups! Use your common sense about the criteria for selection. Ask yourself some basic questions. Will separating leprosy patients according to education, for example, really provide you with more clues to understanding their presentation for therapy?

4.6 How many participants do we want to select?

After deciding who it is you want to include in the project, you need to decide how many people you will want to contact for each session. Focus groups work well with around four to twelve people. Groups with more than eight can be difficult to control, but the decision on how many you want in each group will depend on how your particular community groups together and conducts discussions in natural community settings.

If you have decided on eight participants for each group, it is still advisable to invite ten people, in case some do not arrive at the session. Be careful though not to over-recruit. In many communities it would not be acceptable to turn away participants who had already arrived.

4.7 How do we contact the participants?

This will depend, again, on the community with which you are working. Simply observe the local custom in your area. This usually involves contacting local leaders first, providing an explanation of the study, and gaining permission to work in that village or location. It could also involve meeting with local health workers. Provided you approach such people appropriately, they will usually be happy to help you to locate individuals for the focus groups.

How much notice you give the participants of the focus groups will vary according to the logistics of gaining access to the community. It is ideal to notify the participants the week before and then provide a reminder the day before. In many situations this is not possible, and in some cases participants have been successfully recruited one hour before the session! You need to consider your participants' daily routine and take into account the ease or difficulty for them to attend the session. They are really making a sacrifice to assist you, and this should always be recognised and allowed for.

When the participants are contacted for the first time, they should be provided with information about the study (without actually discussing the focus group questions or directly stating the aim of the study, as this may reduce the quality of the session), about why they have been selected, and how the results will be used. For example, you might introduce a study on perceptions of disease in an area with a high prevalence of schistosomiasis by explaining that you are interested in the health problems of the study community, that in order to really understand these you need to talk to as many people as possible, that you hope to learn from their own experience of health and illness, and that the information you gather will be used to help formulate plans to try to ensure better health. At this time you will also need to check whether anything will need to be provided to help the participant attend, like child care or transport. Personal contact by the project team is strongly recommended as this can show the participants that their contribution is considered important.

Section 5: Developing the question line

5.1 Introduction

The question line or question guide is a list of questions you plan to ask the participants during the focus group sessions. It needs to be very carefully thought out, and you need to be ready and willing to change it several times if necessary. The more practice you have at writing questions, the better you will become, and the sooner you will be able to have a question line that does not need many changes. However, it must be remembered that you cannot possibly get to the very centre of people's beliefs, attitudes and opinions in one or two attempts. If you find that the original questions were exactly correct, be suspicious of yourself! Maybe you are not keeping your mind open enough to what the respondents are really telling you! Be prepared to rewrite after each focus group. At the beginning of the study, every focus group will provide you with new information, and so you will want to incorporate things that you learn into the question line in order to raise those issues with the next focus group.

This section will discuss the basic issues of writing questions and developing the guide. You may also find it helpful to consult more detailed texts on focus group research and writing questions for qualitative research (see, for example, Bernard 1988; Krueger 1988; Morgan 1988; Stewart and Shamdasani 1990).

5.2 Get to know the community

Before you begin writing questions from your objectives, you need to have some understanding of the community you will be working with. The amount of time you spend on this will depend on how familiar you are with the people, their culture and beliefs. If you are new to the area, while you are in the planning phase you can use some time to get acquainted with the community.

Depending on the research question, you can talk to people informally about any issues related to your study. It is also wise to learn about the best ways to enter the community, how groups work naturally within the community, and other issues related to conducting focus groups. Try to develop a dictionary of local terms related to your topic; this will help you later on during the focus groups and in analysis. Of course, you can also use focus groups to build up a vocabulary of local terms, e.g., the variety of words that are used for fever, or cough, or diarrhoea.

5.3 Who writes the question line?

This will depend on the size of your project and whether or not you will be working through translators. It is recommended that everyone involved in the project contribute to the question line. The people who will be making the decisions based on the results should see how the questions will be asked, as should the team conducting the focus group sessions. Everyone needs to agree that the questions relate to the objectives.

If you are working in a second language, then the moderator/translators will need to translate the questions from your language into the language of the community. This requires quite a good understanding of the project objectives. If you are not familiar with the community, local staff should also be used to guide you on the acceptability of the questions to the participants.

5.4 General principles to writing questions

The structure of the questions

Questions are asked in focus groups in a very different way from survey questionnaires. Surveys usually ask respondents questions to gain concise and direct information, often with closed questions and fixed choices. In focus groups, responses should be flexible and should encourage people to reply at length. As a result, the sessions are not tightly structured. Your primary aim is to try to understand a new area, or investigate people's attitudes, opinions and beliefs, and you will not be able to know what the range of answers will be. You need to avoid forcing people into answering questions in particular ways, for you cannot predict how they will want to answer a particular question. Focus groups and other related qualitative research methods such as in-depth interviewing allow respondents to answer in any way they want, and questions do not give away the type of answer required. These types of questions are called "open-ended" questions. For example:

"How do you feel about the leprosy programme?" (to adults receiving multi-drug therapy (MDT) for leprosy).

This question allows respondents to discuss the things they are happy or dissatisfied with. It does not draw attention to any particular aspect of the programme, and does not indicate to the respondent how the discussion should progress. What they actually choose to talk about first can show you what is the most important issue for them, although this may not be the case for sensitive or embarrassing topics.

Say you were interested in knowing whether people had difficulty getting to a clinic which dispensed the drugs. The general questions may not lead to discussions about access to drugs, and you may still want to explore that question in detail. This will require questions with a little more structure, pointing to an aspect of the problem. For example:

"You are all on regular medication for leprosy. Do you find you are able to get the necessary drugs?"

This question allows the respondents to answer in any way they choose, but it does point to a particular aspect of the programme that relates more closely to the objectives. Suppose instead you had asked, "What problems do you have getting to the clinic?", it would have indicated to the respondents immediately that getting to a clinic was a problem. It forces people to discuss access to the clinic, when access might only be a minor area of concern. This type of question could give you a very false idea about the importance of access in the minds of the patients. For them, the more important issue might be privacy-they may prefer to present to a different clinic, where access is a greater problem, in order to avoid others knowing that they are being treated for leprosy.

It is important to avoid questions that will give you a "yes"/"no" answer. It may be necessary on occasion to get a quick "yes" or "no" which can then be explored further, but generally, these are not good questions because they do not encourage lively discussion. As in individual interviews, the best questions are those that encourage people to open up. You want them to tell you about a particular issue or experience. For example:

"A number of you have said that it's too hard to go to the clinic. Could you tell me what things make it so hard?"

Wording the questions

Although writing the questions may seem easy once you have decided on information you want, it may not be quite that simple. Sometimes questions can be confusing, or they may be misinterpreted. At other times participants may simply not know what you are talking about: the focus group may be the first time they have ever given any thought to the issue. Do not assume people know what you are talking about, or that they have well developed ideas about the topic.

There are a few simple rules to follow when you are wording the questions:

• Make sure the questions are easily understood by the respondents

• use simple language,
• be sure the meaning of the question is clear,
• keep questions short,
• do not have several parts to each question.

• Do not word the questions so that people are made to feel guilty or embarrassed.

• Do not indicate any judgement about what is acceptable and what is not.

• If translating questions, always back translate, that is, have the questions translated back into your own language by someone not associated with the project. This allows you to check that the meaning or intent of the question is the same as the original. Sometimes you may need to do this a number of times to ensure that the translation is correct. Do not assume a first translation is accurate, no matter how good your translators are.

• Avoid using too many "why" questions. These questions give people the idea that there should be a sensible, logical answer. Sometimes people cannot provide such answers and may invent a quick response that sounds reasonable. "Why" questions also may sound like an interrogation, which could make them defensive and so answer in what might be a socially acceptable manner.

Number of questions

Focus groups usually last from one to two hours. One and a half hours often is the ideal length. Participants tend to get tired after this. It will depend, however, on the success of the discussion, how lively it is, and how interested the group is to continue. It is recommended that you write your question guide with just over one hour in mind, but let your participants know that it may last a bit longer.

Focus groups are supposed to allow an exploration of attitudes and beliefs in a lot of detail. This is not possible if you have a question guide that covers a large range of topics. If you attempt to do this (and it is very tempting!) you will find during the session that you are rushing through the questions and dealing with each superficially. In addition, it will be hard to follow up on leads and clues given by participants as they respond to questions, and the focus group question line will appear rather like an interview schedule for a survey. We recommend that you limit a single focus group to two or three main ideas, to allow time to really investigate them well. If you feel uncomfortable without a question guide that is rather lengthy and detailed, then experience and pre-testing will show you clearly exactly what is possible and what is not!

Pre-testing

In a pre-test, you take the prepared question guide and try it out to see whether it is understood in the way you intended it to be. You cannot know before you test how people will respond to the questions, and pre-testing is an essential part of designing the question guide. Do not look at the pre-test as the final stage of writing questions; it may be just the beginning! Often these have to be re-written and re-tested. This is quite normal and should be allowed for when you are making your timetable for the project.

Pre-testing can also be used to check the effectiveness of staff training of those who are to be involved in conducting the focus groups. It also provides them with much needed field practice, and helps develop confidence.

On-going question development

Throughout the project, the results of each focus group will give you more information and therefore a better idea of the issues that are important to the participants. With each new piece of information you may need to change the question guide a little so that it can better address the problem and focus in on the main issues. The number of changes varies quite a lot, but in some projects focus group questions have changed with almost every new focus group session. This process-the flexible collection of data, ongoing analysis, and the feeding-back of information while the research is in progress-is referred to as iterative, and is an important feature of all qualitative research methods.

The following two boxes (1 and 2) give examples of a couple of question lines, to give you some idea of the kinds of questions that are used for focus groups. Note that there are only a few for each. Within each focus group, the moderator will build on responses to these questions, through prompts, probes and various other follow-up questions, to encourage further discussion around each topic (see Part II, Section 10.2).

In the first example (Box 1), you are seeking to gain information on what people do when their children are sick. You are interested, firstly, in symptom complexes and diagnostic categories (that is, what sicknesses they perceive), and then in how they treat them and the range of treatment options available to them, including self-medication and folk therapies, traditional healers, village health workers, traditional and government midwives, rural health units, and hospital services. You will want to know where people go for different symptoms and illnesses, and what influences their decisions to go to one service or another. If you are concerned with a particular disease, then you might focus on this disease at some point during the focus group. However, at the beginning, you would ask people very general questions about illnesses and treatments.

The focus group question line is not concerned with biomedical categories of disease, but rather with locally recognised symptom complexes and named illnesses. The moderator will use a range of techniques (probes, prompts, repetition, etc.; see Part II, Section 10), to explore variations of experiences and understandings within the focus group, to build up a vocabulary of locally identified illnesses (e.g., "cough", "fever", "spirit illness", "stomach problems") and to gain a preliminary understanding of the patterns of resort to treatment, including return visits to individual healers or the concurrent or sequential use of healers from various sectors, i.e., folk and professional. The moderator would not word the questions as formally as they are listed. For example, suppose a mother mentioned "normal diarrhoea" (Abdullah Sani et al. 1990). The moderator would not ask, "What are the signs and symptoms of normal diarrhoea?", but rather would ask:

"I wonder if you could tell me a bit more about "normal diarrhoea". For example, could you tell me how you would recognise that your child had this?"

and would then probe to ask;

"Is this the only kind of diarrhoea? Could you tell me what other kinds of diarrhoea children get?"

and also;

"You've mentioned _______, _______, and ______ as different kinds of diarrhoeal disease which children get. Is this all? Are there other words you can use to describe diarrhoea?"

The general order of questions, however, would be followed.

In this discussion, you may be interested in exploring the significance of social networks and relationships within a community in influencing decision-making in the event of illness, particularly malaria. One way to begin discussion is to simply let people know that you are interested in their health, including malaria, and wonder whether they'd tell you about cases of malaria that they have experienced. Where there has been a recent case, you can use this as a starting point for discussion. Hypothetical questions about illness can also help you to encourage an exchange of ideas about diagnosis and treatment, to explore the variability of knowledge and the concordance of ideas within the group. Where contradictory advice or opinions emerge, these need to be explored as fully as possible. You should note also the nature of the interactions of the group, including points at which some may defer to others (e.g., daughters to mothers).

The example in Box 2 seeks to gather a wide range of data on malaria, including its cause, diagnosis, treatment, and prevention. Because of its focus on a single disease, the moderator commences questioning on the disease, but each question is very general. This kind of exploratory focus group might help to identify more specific topics for later group discussions or individual interviews.

Section 6: Managing the information you collect

6.1 Introduction

There are a number of ways to collect and store the information from focus groups. We will look at some of the ways of managing your information, keeping in mind that you may not have access to some of the common technologies used for focus groups, such as computers, cassette recorders or even typewriters. You can still conduct successful focus group projects with a pencil and paper. However, this may limit the size and completeness of your study. It is just a matter of balancing what you want with what you've got!

6.2 Recording the focus group discussions

It is possible to use only one method to record information from focus group discussions, or you can use a combination of methods. The size of the project, and what you are planning to do with the results, may determine the choice you make here.

Taking notes

The most simple method is for the observer of the focus group to take notes of the session, getting as much detail as possible. Without good shorthand skills, it is not possible to get a word-for-word account of the discussion, so the team will need to decide the best method for themselves to get as much information down as possible. This may be just making a summary sentence of each response from each participant, or jotting down key words and then writing down verbatim any striking and important comments. If the note-taker has some shorthand skills, that will help but it is not essential. Practice at note-taking is also important.

Immediately after the session, the observer will go back over the notes and add in further detail in order to give a full and clear account of the session. Observers must only write down what they are sure about. They are not supposed to make their own judgements at this time about what the participants meant, only what they said. If it is important to make a judgement or interpretation of a comment, then it can be written in the notes, but the entry should indicate that it is a comment from the observer, and not from the participants themselves. The example (Box 3) indicates the note-taker's comments in square brackets ([]); these include comments regarding non-verbal behaviour and comments regarding interpretation of information. Direct quotes from participants are marked with quotation marks.

As the example indicates, the observer also makes observation notes about body language, how the group works together and anything else of interest that cannot be defected from other methods of recording the information.

Cassette recording

If possible, the use of a cassette recorder is ideal. Although it is certainly not essential, it does allow a more complicated and thorough analysis of the information if necessary. It also allows you to check sessions you did not attend, assess the performance of the moderator, and check translations, and it provides an accurate and permanent record of the session. It provides a true account that can be listened to again if there is confusion over the meaning of a comment made during a focus group session.

The use of the recording is up to you, but it can provide the basis for a full transcription of the focus group discussion. This full transcription is a written or printed account of the entire session. It is made by writing out the session from the tape recording. This is rather a slow process, but it will greatly increase the quality of your information. It also means that you have a document that can be far more easily analysed in detail, and reduces the influence the observer has on recording the responses. If the only method for recording the session is the observer making summary notes, then from that moment on you will need to treat your information with more caution as it may be the observer's interpretation of the participants' meaning rather than their actual words.

Video recording

This type of recording is sometimes used in first-world industrial contexts. We discuss it only because many texts on focus groups refer to video recordings. Video cameras can be very distracting to participants in focus group sessions, but this will depend on how much exposure the community has had to such technologies. If you have access to a video camera and playback screen, and if you are sure that the camera can be used without distracting the participants from the session, then it can be very helpful. Not only will a video recording provide an account of what the participants said, it also makes a record of how the comments were made. It provides a record of body language (or non-verbal signs) which can be used during the analysis of the information. Body language will be discussed in detail in Part II, Section 11. Video recordings were used in a study to evaluate family planning programmes, and this work indicates they can be useful but not essential in enhancing the quality of data collected (Khan, Anker et al. 1992).

Videos are sometimes used not to record the focus group, but to present information that can be the subject of discussion among focus group participants. For example, in the development of methods for focused ethnographic interviews in ARI (acute respiratory infections), video tapes of infants with fast breathing have been shown to participants of focus groups (e.g., Sison-Castillo 1990).

It is possible to use any combination of these methods. A common approach is to make a cassette recording for later transcription, and for the observer to record aspects that cannot be heard on the recording (e.g., who is speaking, non-verbal communication). The observer also makes summary notes about what the group has said for the immediate debriefing session, and can note key issues raised which will assist in easy reference to the record on tape.

6.3 Preparing the information for analysis

After every focus group it is very important to begin to prepare your information for analysis. For sessions that rely only on note-taking, this means that all notes that are made need to be expanded into the fullest possible record as soon as you can. The moderator will be able to help the note-taker fill out the notes. If you conduct another focus group session before you write down your notes from the previous session, you can easily get confused over who said what and in which group they said it! Build this activity into your detailed plan.

If you are using a cassette recording then the tape will need to be written down (transcribed) as soon as possible. It can take up to a full day for an office assistant to write down all the responses from the entire session. As you will want to read the transcription as soon as possible after the session to help you in question guide development, you will want to have that available quickly. If you are working in a second language, then you need to allow time for translation as well. It is a good idea in this situation to have the recording written down in the language in which the focus group discussion was conducted, and then have that document translated. This allows for better control of the accuracy of translation, as you can check it at each stage, and you will be able to retain and refer to native language terms which might be important to the study. It is also far easier for the translator to work on a written account.

The same applies to the video recording. But instead of making only a transcript of the spoken word, you can easily add in any observations from body language directly onto the transcript.

Whatever recording method you use, it is important to prepare material for analysis within 24 hours of the session. It must be prepared while the session is still clear in your mind.

6.4 Storing the information

Information should be stored in whatever storage system is already in use in your of office or department. There is no need to develop anything special for focus group material. You do need to be aware, however, that if you are making transcripts and analysing the information at the most complex level described here, you will need quite a bit of space to store the information. A single 90-minute cassette recording of a focus group may produce a 15-page transcript. You will also need to allow space for the documents you produce while analysing the information, and for the basic demographic data which you collected when you registered each participant.

Even if you are fortunate enough to be using a computer, information will still need to be stored carefully. Ensure that someone is in charge of maintaining a good storage system as an untidy and illogical filing system could quickly produce many unnecessary frustrations. And remember, if you are using a computer, make backup files: it's easy and common to lose data unless thou do this!

Contents -

How do you choose participants for an experiment?

What is the process of selecting participants called?

What is it called when the participants of an experiment are put into groups?

What sampling method is used for experiments?