Document(s) 15 of 27

1. Introduction
2. Types of questions
3. Steps in designing a questionnaire/interview guide
4. Checklists
5. Interview skills

I. INTRODUCTION

The quality of research depends to a large extent on the quality of the data collection tools. Interviewing and administering questionnaires are probably the most commonly used research techniques. Therefore designing good ‘questioning tools’ forms an important and time-consuming phase in the development of most research proposals.

Once the decision has been made to use these tools, the following questions should be considered before designing them:

• What exactly do we want to know, according to the objectives and variables we identified earlier?
• Is questioning the right technique to obtain all answers, or do we need additional techniques, such as observations or analyses of records?
• Of whom will we ask questions and what techniques will we use? Do we understand the topic sufficiently to design a questionnaire, or do we need some loosely structured interviews with key informants or a FGD first to orientate ourselves?
• Are our informants mainly literate or illiterate? (If illiterate, the use of self-administered questionnaires is out of the question.)
• How large is the sample that will be interviewed? Studies with many respondents often use shorter, highly structured questionnaires while smaller studies allow more flexibility and may use interview guides or questionnaires with a number of open-ended questions.

II. TYPES OF QUESTIONS

In this module we will concentrate on interactive (usually face-to-face) interviews. Before examining the steps in designing a questionnaire, we need to review the types of questions used in interviews. Depending on how questions are asked and recorded we can distinguish two major possibilities:

• open-ended questions, (allowing for completely open as well as partially categorised answers), and
• closed questions.

1. Completely open-ended questions

Such questions are useful for obtaining in-depth information on:

• facts with which the researcher is not very familiar,
• opinions, attitudes and suggestions of informants, or
• sensitive issues.

For example:

‘What is your opinion on the services provided in the ANC?’ (Explain why.)

‘What do you think are the reasons some adolescents in this area start using drugs?’

‘What would you do if you noticed that your daughter (school girl) had a relationship with a teacher?’

The answers to these questions are written down as closely as possible in the words of the respondents.

Advantages of completely open-ended questions

• Allow you to probe more deeply into issues of interest being raised.
• Issues not previously thought of when planning the study may be explored, thus providing valuable new insights on the problem.
• Information provided in the respondents’ own words might be useful as examples or illustrations, which add interest to the final report.
• Often, re-reading an answer in a later phase of the analysis offers the possibility for different interpretations in relation to other data collected, which would have been impossible if the answer had been pre-categorised.

Risks of completely open-ended questions

• Skilled interviewers are needed to get the discussion started and focused on relevant issues and to record all information collected. A big risk is incomplete recording of all relevant issues covered in the discussion.
• Analysis is time-consuming and requires experience; otherwise important data may be lost.

Suggestions to improve use of completely open-ended questions

• Thoroughly train and supervise the interviewers or select experienced research assistants. (See section V of this module.)
• Prepare a list of further questions to keep at hand to use to ‘probe’ for answer(s) in a systematic way.
• Pre-test open-ended questions and, if possible, pre-categorise the most common responses, leaving enough space for other answers (see 2).

2. Partially categorised questions

In interviews questions are often asked as open-ended questions, but to facilitate recording and analysis, some answers can already be pre-categorised.

For example:

‘How did you become a member of the Village Health Committee?’

In this case the first four categories of answers are known, but there may be other possibilities. Therefore there is a category ‘other’ where other answers can be recorded. During the analysis, these responses can still be further categorised.

For open-ended questions, more than one answer is usually allowed. The interviewers will have to be trained to wait for additional answers

Advantages of pre-categorised answers

• Answers can be recorded quickly, and
• Analysis is easier.

Risks of pre-categorised answers

• If one pre-categorises too early, a lot of interesting and valuable information may never be recorded, or may end up in the category ‘other’.
• Interviewers may try to force the information into the categories that are listed and, by merely ticking these, additional valuable information will be lost.
• Interviewers may stop after receiving the first answer, whereas more than one response could be applicable.
• Sometimes, if the respondent hesitates in answering, the interviewer may be tempted to present some possible answers, thereby causing bias.
• Frequently, questionnaires have very little space for recording full responses under ‘other’, forcing the interviewer to write down a response that summarises the respondent’s answer, thereby losing valuable information.

Suggestions to minimise risks associated with pre-categorised answers

Clear guidelines have to be provided to interviewers on important issues. For example:

• If a question leads to an interesting discussion, it should be written down as completely as possible, in addition to being coded. Space should be reserved to record these discussions.
• Interviewers should be trained to solicit discussion when questions allow respondents to choose more than one option. The different options may be elaborated in subsequent questions.
• In case of non-response, the interviewer should repeat or rephrase the question without providing options for answers. The interview guidelines should provide suggestions for further probing which all interviewers should follow (see section V).
• Adequate space should be provided so that ‘other’ responses can be recorded as close as possible to the respondents’ own words. Otherwise categorisation of these responses may be difficult afterwards.

3. Closed questions

Closed questions are most commonly used for background variables such as age, marital status or education, although in the case of age and education you may also take the exact values and categorise them during data analysis (see Module 13).

Closed questions may be used to get the respondents to express their opinions or attitudes by choosing rating points on a scale.

For example:

What is your opinion on the following statement:

‘Women who have induced abortion should be severely punished.’

Closed questions may also be used if one is only interested in certain aspects of an issue and does not want to waste time obtaining more information than one needs.

For example, a researcher who is only interested in the sources of protein in a family diet may ask:

‘Did you eat any of the following foods yesterday?’ (Circle yes if at least one item in each set of items is eaten.)

Using attitude scales in face-to-face interviews with literate respondents is most objectively carried out if the various options for each answer are provided on different cards. The respondents can be asked to put the cards in the order preferred by them while making their choice. If the researcher only reads the options, the respondents might not consider all options equally and the scale will not accurately measure the attitudes.

Advantages of closed questions:

• It saves time
• Comparing responses of different groups, or of the same group over time, becomes easier.

Risks of closed questions:

• In case of illiterate respondents, the interviewer may be tempted to read the list of possible answers in the given sequence, thereby influencing the choice of response and introducing bias.
• If there is no question to elaborate on the informant’s reasons for choosing a certain rating, uniformity in rating may still be deceptive, as there may be considerable variation in reason for choosing the same ratings.

Suggestions to minimise risk associated with using closed questions:

• Develop picture codes can be used for illiterates as well as literates (e.g., five, four, three, two and one stars indicating a 5-point scale).
• First present the extremes and then the values in between so that the respondent is straight away aware of the range of answers.
• Ensure inclusion of follow up questions to elaborate on reasons for choosing a given rating.

III. STEPS IN DESIGNING A QUESTIONNAIRE/ INTERVIEW GUIDE*

Designing a good questionnaire always takes several drafts. In the first draft we should concentrate on the content. In the second, we should look critically at the formulation and sequencing of the questions. Then we should scrutinise the format of the questionnaire. Finally we should do a test-run to check whether the questionnaire gives us the information we require and whether interviewers as well as respondents feel at ease with it. Usually the questionnaire will need some further adaptation before we can use it for actual data collection.

* For the sake of simplicity we take semi-structured questionnaires used (containing completely open as well as partially pre-categorised and closed questions) in face-to-face interviews as an example. The same steps apply to designing other ‘questioning tools’.

Step 1: Content

Take your objectives and variables as a starting point.

Decide what questions will be needed to measure or (in the case of qualitative studies) to define your variables and reach your objectives.

When developing the questionnaire, you should reconsider the variables you have chosen and, if necessary, add, drop or change some. You may even change some of your objectives at this stage.

Step 2: Formulating questions

Formulate one or more questions that will provide the information needed for each variable.

Take care that questions are specific and precise enough so that different respondents don’t interpret them differently. For example, a question such as: ‘Where do community members usually seek treatment when they are sick?’ cannot be asked in such a general way because each respondent may have something different in mind when answering the question:

• One informant may think of measles with complications, and say he goes to the hospital, another of cough, and say he goes to the private pharmacy
• Even if both think of the same disease, they may have different degrees of seriousness in mind and thus answer differently
• In all cases, self-care may be overlooked

The question therefore, as a rule, has to be broken up into different parts and made so specific that all informants focus on the same thing. For example, one could:

• Concentrate on illnesses that have occurred in the family over the past 14 days and ask what has been done to treat these from the onset; or
• Concentrate on selected diseases, ask whether they have occurred in the family over the past x months (chronic or serious diseases have a longer recall period than minor ailments) and what has been done to treat each of them from the onset.

Check whether each question measures one thing at a time.

For example: the question, ‘Do you think that the war situation leads to mental problems that require treatment by health staff?’ brings three topics, which should be split up in

• mental problems resulting from the war,
• treatment required, and
• who should provide the treatment.

Avoid leading questions.

A question is leading if it suggests a certain answer. For example: the question, ‘Do you think that people have to give bribes at hospital X to be seen by a doctor?’ hardly leaves room for ‘no’ or for other options. A better question would be: ‘Have you recently visited hospital X?’ This would be followed by a series of other probing questions such as, ‘By whom were you seen?’ ‘What were the complaints?’ ‘How much were you asked to pay?’ ‘Are the fees fixed?’ ‘Do they include an examination by the doctor if the condition of a patient is serious?’ ‘Do all patients have equal access to a doctor in case of need?’ ‘Was this what you expected?’

Sometimes, a question is leading because it presupposes a certain condition. For example: ‘What action did you take the last time your child had diarrhoea?’ presupposes the child has had diarrhoea. A better set of questions would be: ‘Has your child ever had diarrhoea?’ (If yes:) ‘When was the last time?’ ‘Did you do anything to treat it?’ (If yes:) ‘What?’

Avoid words with double or vaguely defined meanings or that are emotionally laden. Concepts such as dirty (clinics), lazy (patients), or unhealthy (food), for example, should be omitted.

Ask sensitive questions in a socially acceptable way:

Questions relating to abortion, sexual practices of adolescents, or AIDS and mental illness in the family are usually sensitive. Such questions should be formulated in such a way that the question does not judge or embarrass the respondent. Making the subject more socially acceptable and then asking a question about the respondent’s own preference or practice could lead to a more accurate response. For example: ‘Many teenagers have had abortions for unwanted pregnancies. Do you know girls who had this problem? Have you ever had an abortion?’

Another way to deal with sensitive questions (as indicated by the respondent’s hesitation in answering the question) is by asking the question indirectly. For example you could ask: ‘If your friend was considering abortion for her daughter who became pregnant while in school, what would you advise her?’

A common weakness in questionnaires is the inappropriate transformation of research questions into interview questions. You cannot ask informants: ‘Would a woman’s educational status influence her health?’ That is exactly what you have to find out by relating individual women’s level of education to a number of health conditions.

Step 3: Sequencing the questions

Design your interview schedule or questionnaire to be ‘informant friendly’.

• The sequence of questions must be logical for the informant and allow, as much as possible, for a ‘natural’ conversation, even in more structured interviews.
• At the beginning of the interview a limited number of questions concerning ‘background variables’ (e.g., age, education, marital status) may be asked.
• As informants may be reluctant to provide ‘personal’ information and may become worried about confidentiality, or bored by a list of unrelated and, to them, senseless questions, you should restrict yourself to an essential minimum. You may postpone questions on religion until later when cultural questions are being asked. Socio-economic status/ occupation/ income questions can also better be postponed until later when you can link them to problems (e.g., low service utilisation).
• Start with an interesting but non-controversial question (preferably open) that is directly related to the subject of the study. This type of beginning should help to raise the informants’ interest and lessen suspicions concerning the purpose of the interview.
• Pose more sensitive questions as late as possible in the interview (e.g., questions pertaining to income, political matters, sexual behaviour, or stigma experienced in case of stigmatising diseases).
• Use simple, everyday language.
• If interviews are carried out in English (or any other secondary language), local terminology should be used for crucial concepts that do not have the exact equivalent in the secondary language.

Step 4: Formatting the questionnaire

When you finalise your questionnaire, be sure that:

• A separate, introductory page is attached to each questionnaire, explaining the purpose of the study, requesting the informant’s consent to be interviewed and assuring confidentiality of the data obtained.
• Each questionnaire has a heading and space to insert the number, date and location of the interview, and, if required, the name of the informant. You may add the name of the interviewer, to facilitate quality control.
• Layout is such that questions belonging together appear together visually. If the questionnaire is long, you may use subheadings for groups of questions.
• Sufficient space is provided for answers to open-ended questions, categories such as ‘other’ and for comments on pre-categorised questions.
• Boxes for pre-categorised answers are placed in a consistent manner (e.g., on the right half of the page). (See example in Module 13, Annex 13.1.)

If you use a computer, the right margin of the page should be reserved for boxes intended for computer codes. (Consult an experienced facilitator when designing your questionnaire.)

Your questionnaire should not only be INFORMANT - but also RESEARCHER FRIENDLY!

Step 5: Translation

If interviews will be conducted in one or more local languages, the questionnaire should be translated in order to standardise the way questions will be asked.

After having it translated you should have it retranslated into the original language by a different person. You can then compare the two versions for differences and make decisions concerning the final phrasing of difficult concepts.

Self-administered (written) questionnaires

All steps discussed above apply to written questionnaires as well as to guides/questionnaires used in interviews. For written questionnaires, however, clear guidelines will have to be added on how the answers to questions should be filled in.

Self-administered questionnaires are most commonly used in large-scale surveys using predominantly pre-categorised answers among literate study populations.

As a response rate of 50% or less to written questionnaires is not exceptional, these tools will rarely be used in small-scale studies. In exploratory studies which require intensive interaction with informants in order to gain better insight in an issue, self-administered questionnaires would, more over, be inadequate tools.

However, written questionnaires may sometimes be useful in small-scale studies on sensitive topics.

Even then, they are usually combined with other tools:

For example, during a FGD on sensitive issues such as sexual behaviour, a facilitator cannot always be sure that all participants dare to speak freely. A short, written questionnaire with open questions may then be used to explore remaining sensitive questions or worries of participants without revealing their identities.

Sometimes, a self-administered questionnaire seems the only possibility:

For example, on one of the smaller islands in the Indian Ocean, the shortage of nurses was acute and public complaints about the functioning of health facilities abounded. A relatively high number of nurses had been leaving the profession over the past ten years and the vacancies could not all be filled as less and less young people were choosing nursing as a career. The research team that investigated the problem (all MOH staff) appreciated the sensitivity of the topic. Aware that everyone on the island knew each other, it was decided to use a flexible self-administered questionnaire which members of the research team personally delivered at each health facility, stressing the importance of filling it in. In that way they hoped to ensure anonymity as well as increase participation in the study. (Still, the response rate was only slightly over 50%.)

IV. CHECKLISTS

Checklists can be used to systematically observe human behaviour or the condition of specific equipment (e.g., fridge, expiring dates of medicines, or completeness of records). Observations can be relatively open or can be predetermined comparisons of reality against fixed standards. Sometimes the aim is systematic content analysis (e.g., newspaper articles, health information system). The objectives of the study determine the content of a checklist. A checklist includes all the items or points that must be considered during an observation in the field, or when extracting data from existing records.

An observation checklist for rating the cleanliness and use of Blair/Ventilated Improved Pit Latrines (VIP) in a study on hygiene practices is presented below, as an example:

V. INTERVIEW SKILLS

1. The interviewer-informant relationship and interview conditions

An interviewer needs to have the skills of a detective. (S)he should carefully, step-by-step, delve for the truth. Here the comparison stops, because the truth researchers are looking for, in general, has nothing to do with criminality. To the contrary, our respondents are possible partners who with their information help solve a shared problem, or at least help to better understand why people behave as they do.

To turn an informant into a partner, a researcher has to invest in the relationship. First of all, (s)he has to be clear to the informant about the purpose of the interview and the study. Enough information should be given to raise the interest of informants and to enable them to judge whether they would like to participate or not. Consent has to be obtained before the interview. (See Module 10A.) On the other hand, not too many details should be given about what will be asked and why. Otherwise, the researcher runs the risk informants become selective in what they tell, and conceal information in order to ‘help’ or please him/her (interviewer bias).

When sensitive topics are being explored anonymity should be ensured, for example, by not including a name in the interview notes. If a second interview would be useful, one may ask a name and address, but make sure this will never appear in the report.

Further, before the research starts, the possibility of involving informants in the discussion of the results and recommendations should be considered and discussed with informants. Feedback sessions with informants are usually rewarding for both informants and researchers. If it is impossible to organise such sessions for all participants, an interviewer can at least summarise the main results of the interview at the end and check if this is what the informant meant to say. Usually the informant will react, elaborate on the responses already given, or sometimes withdraw a statement if it seems too personal.

Partnership between interviewer and informant implies that the interviewer will try to minimise the social distance between him/herself and the informant. Interviewers should try to blend in the environment. In order to do so:

• Clothing of interviewers should be culturally acceptable and as simple as possible (no fancy dresses, high heels or tight jeans in rural areas).
• Sitting arrangements for interviewer(s) and informant(s) should preferably be at the same height (no straight chair when the informant is sitting on a mat on the floor) and beside each other, forming an angle of 90 degrees, rather than opposite each other.
• Gender relations have to be respected. When interviewer and informant are of opposite sex, more physical distance will usually be required than when they are of the same sex. In some societies, a male researcher may have to ask consent from a woman’s husband or father before he can interview her, or may have to employ a female research assistant to interview women. One has to ask advice from key informants and try out what works, as ‘opposite sex interviews’ sometimes also produce interesting information that in ‘same sex interviews’ would not so easily be obtained. Differences in age and culture, if exploited with care, may have the same surprising effect of generating useful information on sensitive topics.
• A general rule is that, through his or her behaviour, an interviewer should show interest in what the informant says, be at ease (never in a hurry) and make the informant feel at ease. This implies that the interviewer should never show any disapproval of the information received during the interview. Otherwise the informant will close up. The interviewer should only try to listen and understand WHY people do what they do, even if the practice seems dangerous. Only after the interview or after the research is finished, can the interviewer try to address problems identified through the interviews.

For example: A research team headed by a Regional Medical Officer (RMO) carried out a study into reasons for low utilisation of delivery services in a district in Tanzania. When interviewing Traditional Birth Attendants (TBAs), the team discovered that TBAs delivered services in case of highly complicated deliveries which could be dangerous to the mothers. The RMO was surprised about the practices of the TBAs and asked many details about what they did with respect to various complications, but did not show his surprise. After finalising the study he organised a training course for TBAs in which he took their level of knowledge as a point of departure and explained why some practices were dangerous and could better be replaced by safer alternatives.

The environment should also be supportive of the interview situation. Anything that disturbs (noise, other people listening, a formal surrounding) should if possible be avoided.

For example: A hospital-ward, with nurses running in and out, may not be a good place to ask a patient’s treatment history including traditional treatment, or his/her opinion about the treatment in the hospital. An adolescent home is not an appropriate environment for an interview with adolescents about their sexual behaviour, even if the parents are absent. A quiet place outside would be better.

If it appears impossible to prevent others from joining the interview, even after having made clear that one would like to speak undisturbed with the informant, there are different possibilities.

For example:

• If the information is personal but not very sensitive, one may just continue, hoping that others may get bored and disappear.
• Depending on the sensitivity of the data and who joins, one can turn the interview into a mini focus group discussion, asking indirect instead of personal questions (e.g., ‘how would a husband react if he found out that his wife was using the pill to prevent pregnancy without his knowledge’, instead of ‘how would your husband react. . .’).
• One may try to make a new appointment and hope the informant will be able to assure more privacy then.
• Or one may tell the visitors that they will be visited later.

Tape recording may be an enabling or a disturbing environmental factor in the interview. If the informant accepts the argument of the interviewer that (s)he will not be able to remember everything that is said nor write it down, it is an asset. Playing with it at the beginning of the interview and letting informants hear themselves may help to overcome hesitation among informants who have never experienced the use of a tape recorder. Informants should be asked for their consent before use, and be assured confidentiality. If they hesitate or refuse, the only solution is to take notes. This is in any case advisable, as the tape-recorder may not always function correctly. Usually, informants forget quickly that the recorder is on, but if they appear disturbed despite their consent that it be used, it should be stopped.

2. The interviewer’s tasks

Apart from the introduction to the interview, which includes building a partnership that needs to last throughout the interview and even beyond, the interviewer has a number of other tasks:

— Posing questions

— Evaluating answers and probing for elaboration or more precision in case of a superficial or invalid answer

— Noting down answers

— Leading the discussion, but at the same time encouraging the informant to give spontaneous information relevant to the topic by letting him/her talk

Unless the interview is carried out in pairs, the interviewer will have to carry out all these tasks at the same time.

(1) The introduction to the interview should be written as the interview guide is being developed, even if there is only one researcher, but certainly if a research team conducts the interviews. This will ensure consistency, which prevents bias due to interviewer variation.

We have already paid much attention to
(2) the art of posing questions in the preceding parts of this module. Not yet covered is the issue of what to do when an answer is unsatisfactory. Clearly, the quality of an interview will be determined by the skills of the interviewer in
(3) evaluating answers and diagnosing what went wrong and why, so that corrective action can be taken on the spot.

An answer can be unsatisfactory in many ways:

• There may be no answer at all, either because the informant did not understand the question or because the question touched on sensitive information that the informant hesitated to divulge.

  Possible remedy: If you suspect that the question was not understood, you may say: ‘Perhaps, I was not so clear. What I meant to ask was . . .’, followed by the question, phrased slightly more elaborately. If you suspect the informant is hesitant to answer, then repeat that if (s)he wishes (s)he can skip this question. If the informant has no objection to continuing, stress again that the information will remain in confidence. Then repeat the question.

• The answer may be incomplete, or unclear.

  Possible remedy: Repeat what was said and encourage the informant to continue with phrases such as: ‘I do not fully understand what you mean. Could you elaborate a bit?’ or: ‘Could you mention more possibilities?’

  Or, if the informant has started to give an example but did not finish, you can probe with questions such as: ‘When was that?’, ‘Why did you do that?’, ‘How often does that happen?’,

  ‘Why do you think so?’, etc.

• The answer may be irrelevant to your question.

  Possible remedy: Say something such as ‘That is not exactly what I meant to ask about’ and repeat the question, slightly elaborated.

• You suspect the informant is not telling the truth (invalid answer), either because various parts of the answer contradict each other or because the informant knows the ‘desirable’ answer and gives it to please the interviewer. Both can happen at the same time.

  For example: in an interview about sexual behaviour, a young man reports that he has used condoms last month. Later he says that he only uses condoms with casual partners. When asked to elaborate a bit on casual partners (who they are, where he finds them, how often, when the last time was) he states that he has not had a casual partner over the past three months, since he has a regular partner. Probably the first answer was a ‘socially acceptable’ statement to please you, but the contradicting data may also mean that the truth (actual sexual behaviour) is quite complicated. It may be ‘socially unacceptable’ to his new girlfriend that he still has other sexual contacts; he may be denying those contacts, and the first answer (that he used condoms with a casual partner last month) may be right.

  Possible remedy: Summarise the three contradicting statements, and ask the informant: ‘How should I interpret this: You said that you have used condoms last month. You also said that you use condoms only with casual partners, and that since three months you have not had casual partners as you have a regular partner now. . .’ Then wait and see what comes. Avoid the slightest sign of disbelief or irritation. Make clear by an attentive and interested look that you would appreciate it highly if he told something more about himself and when he does and does not use condoms with his different girlfriends.

Note that when something goes ‘wrong’ in an interview the interviewer should always take the blame. (S)he is the one who ‘did not ask the question clearly enough’, or who ‘did not understand the answer’, and so is asking the question again. The interviewer is the one who receives a ‘present’ in the form of information and has to show by word and body language that (s)he appreciates this highly. Nodding, humming, or repeating the last few words may all be perceived as encouragement. A silence from the side of the interviewer may also work as encouragement for the informant to continue.

(4) Noting down answers should never go at the cost of the cost of the eye contact with the informant, even if the information collected is not sensitive. Scribbles in a small notebook are preferable over more extensive notes, provided they are elaborated straight after each interview. It is advisable to take brief notes even when a tape recorder is used, just in case the recording fails. Elaborating notes after an interview takes more time than the interview itself, and the full transcription of a taped interview may take between 5-10 times the duration of the interview, and often longer.

(5) Keeping control over the interview without imposing oneself is a skill each researcher has to learn. Even though an interviewer may encourage the provision of (often valuable) spontaneous information, some spontaneous information may distract from the main issues. The interviewer has to take care that there is not only a good beginning but also a satisfactory end to the interview. In between, all topics should have been adequately covered. Otherwise the information provided by different informants will not be comparable or, in case of key informants, not be comprehensive. Therefore an interviewer will also sometimes have to stop informants who go enthusiastically off track. This can be done politely: ‘Thank you, this is interesting, but do you mind if I go back to the previous (or next) question? This (specify) is not yet fully clear to me.’

At the end, the interviewer should not only summarise the interview, which may lead to a valuable new discussion, but also respond to questions that came up during the interview, give advice (if necessary or asked for) and give an opportunity for further questions of the informant. Such ‘after-interview’ discussions and questions should always be recorded, like all spontaneous information, because discussions can shed light on complicated, not yet fully clear issues from many preceding interviews. (Remember the researcher’s function of detective!)

3. Training the research team/assistants

Those who for the first time conduct a face-to-face interview will need training in the application of all the advice provided above. Good quality interview data form the heart of an HSR study. Obtaining such data presupposes more than appropriate knowledge and skills; also specific attitudes (e.g., not imposing one self) and insight (e.g., appreciating signs of discomfort in an informant, and recognising evading or, to the contrary, highly illuminating answers) are required.

The learning process therefore preferably consists of reading and listening/observing, as well as doing. ‘Doing’ may involve a written exercise, a role-play, or actual field practice during a pre-test. The following exercise may be helpful:

REFERENCES

Briggs CL (1992) Learning how to ask. A sociolinguistic appraisal of the interview in social science research. (In particular chapter 3, analysis of communicative blunders) Cambridge: Cambridge University Press. (5^th edition)

Lee RM (1993) Doing Research on sensitive topics. London: Sage Publications.

Pretty JN, Guyt I, Thompson J, Scones IL (1995) Participatory Learning & Action. A Trainer’s Guide. London: International Institute for environment and Development (IIED) (In particular Chapters 4 and 5 on semi-structured interviewing, diagramming, ranking and scoring.)

Rubin HJ, Rubin IS (1995) Qualitative Interviewing. The art of hearing data. Thousand Oaks: Sage.

Seidman I (1998) Interviewing as qualitative research. A guide for researchers in education and the social sciences. New York: Teachers College Press. (In particular chapters 5-7.)

Spradley JP (1979) The ethnographic interview. New York: Holt, Rinehart and Winston.

Annex 10B.1: MATERNAL RECORD*

* This questionnaire which was made for use by non-literate health workers was shared by Dr. Peter Lamptey.

Trainer’s Notes

Module 10B: DESIGN OF RESEARCH INSTRUMENTS; INTERVIEW GUIDES AND INTERVIEW SKILLS

Introduction and discussion

• The introduction should be straightforward but interactive, providing participants with an opportunity to comment on obviously poor questions and come up with suggestions for improvement.
• The handling of half open-ended, partly pre-categorised questions, which appear in many questionnaires, may need special attention. Participants should be aware of the danger of bias if these questions are not asked uniformly with guidelines for probing.
• The formatting of questionnaires should be illustrated with examples.

Exercise: Interview training

1. Written exercise: Ask each participant to write an introduction for one of the interviews his or her team will be developing for its study. Then ask each team to meet in a small group and critique each other’s introductions.
2. Role-play: Ask six participants to volunteer to participate in the role play - three as interviewers (health care staff working in an area with a high percentage of unsupervised deliveries) and three as mothers who have delivered in the past 6 months. Announce that the task of the three interviewers is to interview the mothers to determine the reasons for the low percentage of deliveries supervised by the health centres and possible ways to increase coverage. Ask the three interviewers to go out of the room and each separately jot down ideas on the questions he or she might ask the mothers in five minutes interviews. While the interviewers are doing this, discuss the roles of the three mothers with the participants who have volunteered to play them and the rest of the class. (One mother may have given birth at home and honestly will tell why; another may have given birth at home and is hesitant to disclose why; and the third may have given birth in the hospital, but is not an easy talker; the interviewer will have to draw the information out of her to get a complete picture.)

   Once the ‘stage is set’, ask each of the interviewers to interview one of the mothers in turn, while the rest of the class watches. When the three interviews are completed, lead a discussion on the attitudes of the interviewers, their skills, what has been done well and suggestions for improving the interview. The ‘mothers’ can be asked what their feeling were about the interviews and what the interviewers could have done to obtain more accurate and complete information.

   Another possibility is to take a ‘real life’ situation and let three pairs of participants interview each other on a topic of common interest. Discuss this first with the participants and ask for their suggestions for a topic.

Group work

• All facilitators should be aware that the quality of the data collection tools determines the quality of the data with which the participants return from the field. Skilful guidance of the groups is therefore essential.
• If participants are relatively inexperienced in research, the first version of an interview guide/ questionnaire is often too general and has too many closed questions. It is extremely important that the groups try their tools out in a ‘real life’ situation during the workshop.
• The time needed to develop the data collection tools may exceed four hours. Usually groups continue working in the evening. There should be two opportunities in the workshop program for the teams to revise the data collection tools: after the exercise on reviewing the data collection tools in class and after the exercise in the field, so that they can include what they learned from their own experience and the comments of facilitators and other course participants. In their own research situations the teams will then do full pre-tests (see Module 14).

Exercise: Review of data collection tools

• If possible, have all groups review and critique the data collection tools of all other groups, with special attention to the tools of one group. In plenary, the group that has the primary responsibility for reviewing a particular group’s tools should comment first. Then other groups should be asked to give any additional suggestions, after which the group whose tools have been discussed can reply, if necessary.
• When groups have many different data collection tools (questionnaires, checklists, schedules for FGDs), it may not be possibly to discuss all tools of all groups in two hours. Then two groups could swap their tools and discuss them in mini plenaries of the two groups. Facilitators, however, should read and comment on the data collection tools of all groups.

Document(s) 15 of 27