SURVEYS INTERVIEWS
INTRODUCTION
Interviewing - verbal questioning - is one of the most common methods of data collection. Interviews and questionnaires together make up the survey method. Interviews are employed as methods of data collection in most research designs, regardless of the underlying methodology. Quantitative researchers employ mostly standardized and semi-standardized interviews. Feminist and qualitative researchers employ mostly un-standardized forms, such as intensive and focused interviewing.
1 DIVERSITY IN INTERVIEWS
1.1 Types of interview
Interviews are, in a way, questionnaires that are presented verbally; they are ‘talking questionnaires’. More importantly, questionnaires constitute the structure of interviews. This is relevant to our discussion in two ways. First, the discussion of questionnaires in the previous chapter applies here also, so there is no need to discuss them again in this chapter. Second, the diversity of questionnaires is reflected in the diversity of interviews: they are as diverse as the underlying questionnaires.
However, the questionnaire structure is only one of the sources of diversity in interviews. Interviews differ also in purpose, the role of the interviewer, sample size and presentation, as we shall see.
Structured and unstructured interviews
Structured interviews employ structured questionnaires which are verbally presented to respondents, with the answers recorded in the questionnaire by the interviewer. When conducting a structured interview, strict adherence to the order and wording of the questions and the instructions is required. The interviewer in expected to perform almost 'like a robot', acting in a neutral manner, keeping the same tone of voice across the interviews, offering a consistent impression to the respondents, using the same style, appearance, prompts, probes and so on, and showing no initiative, spontaneity or personal interest in the research topic. Response categories are fixed and prescriptive. This serves to reduce interviewer bias to a minimum and achieve the highest degree of objectivity and uniformity in procedure. This form of interview is employed in quantitative research.
Unstructured interviews employ unstructured questionnaires (interview schedules) containing a number of open-ended questions, whose wording and order can be changed at, will. The interviewer acts freely in this context, on the basis of certain research points, reformulating questions as required and employing neutral probing. The structure of the interview is flexible and the restrictions minimal, in most cases taking the form of guides rather than rules. This type of interview is mostly used in qualitative and feminist research.
Semi-structured interviews lie somewhere between the structured and unstructured types. They contain elements of both, with some being closer to structured interviews, and others closer to unstructured ones. The degree to which interviews are structured depends on the research topic and purpose, resources, methodological standards and preferences, and the type of information sought, which of course is determined by the research objective. They are employed in qualitative, feminist or quantitative research.
Delphi interviews
This is a version of the ethnographic interview, employing a multi-stage ethnographic approach. The interviewer questions persons who are experts in the area of study, In a Delphi interview, respondents are asked to offer information, pass judgments on the issue in question, and make relevant predictions. The researcher summarizes this information in a logical and sociological context and offers a written summary to the experts for comments and discussion. Eventual comments or identified deviations are discussed and considered again, and the new summary is again handed to the key informants for extra comments. This process is continued until the differences of opinion among the experts are reduced significantly.
Other types of interview
There are many other types of interview which cannot be accommodated in this discussion. A selection of these interview types is given below.
· Analytic interviews: These are designed to analyze concepts and theories following a specific paradigm,
· Biographical interviews: These record a person's life history; they are also known as narrative interviews.
· Convergent interviews: These include two interviewers taking notes independently and writing a report together.
· Diagnostic interviews: These aim to find out specific attributes (feature) of the respondents,
· Dilemma interviews: These evaluate the subject's response to a problem (dilemma) presented by the interviewer.
· Elite interviews: Respondents are well-known, prominent people.
· Ethnographic interviews: Respondents include key informants or experts aiming to study (sub-cultures.
· Group interviews: Several respondents are interviewed at one time.
· Guided interviews: The interviewer takes a guiding not a dominating role.
· 'Hard' interviews: These are based on strict procedures; the interviewer views the responses with doubt and scepticism, and the interview resembles an interrogation.
· Individual interviews: Only one respondent is interviewed at a time.
· 'Inquiring' interviews: The interviewee is seen as an equal informing partner.
· Neutral interview: The interviewer is neutral, factual, encouraging, friendly but also distanced and impersonal.
· Open interviews: These allow flexibility and change in their structure and process.
· Panel interview: These are conducted more than once, employing the same sample.
· Problem-centered interviews: These are focused on a particular problem; they are open, and the interviewee's critical contribution is valued.
· Receptive interviews: The interviewee is the actor and the interviewer an active-supportive listener.
· ‘Soft' interviews: The interviewer assumes a guiding role only.
· Telephone interviews: Interviews are conducted via the telephone.
· Unique interviews: These interviews are conducted only once.
1.2 Interviews in qualitative research
In qualitative research, the most common interviews are the semi-structured and unstructured ones, and more generally those that comply fully with the standards and principles of qualitative research. The methodological parameters and technical elements of qualitative interviewing are briefly listed below.
Reflexivity: Qualitative interviewing employs methods and a process of analysis in which researchers reflect upon their subjective approach to the world, and take into consideration the implications of the knowledge they produce for social life.
Naturalism: Qualitative interviews are directed towards studying reality as it really is, in its own terms, and as it is manifest in everyday life events.
Primacy of the respondent: The respondents are experts who provide valuable information. They are as important as the researcher and not just a source of data.
Absence of standardization: Qualitative interviews are un-standardized interviews, allowing freedom to respondents to express their views without external limitations.
Openness: Qualitative interviews do not use a strictly standardized approach. Rather, they employ a readiness to change, to correct and adjust the course of study as required by the research. Interviewers are expected to engage in upon discussion with the respondent, and to maintain a stimulating, but not dominating, role.
Flexibility: The qualitative researcher follows the course that emerges through the interview.
Life as process: Qualitative interviews find out aspects of personal experience as displayed in everyday life.
Grounded theory: Qualitative interviews in most cases aim at developing a grounded (data-based) theory.
Explication: Findings emerge through the study and are interpreted during the process of interviewing.
These methodological and technical aspects show that qualitative interviews are far from being a soft methodological option or an easy form of research. On the contrary, they require more competence on the part of the interviewer and higher ability on the part of the respondent to verbalize views, opinions and ideas
Table 12.1 Comparison of quantitative and qualitative/feminist interviews
| Criteria | Quantitative Research | Qualitative/Feminist |
| Nature of questions Order of questions Interviewing assumes Interviews are controlled Interviewers usually are Probing/prompting Duration of interviews Details of interviewing Interviewees per study Overall structure Presentation of questions Number of interviewers Nature of interviewing Sample | Mostly structured/semi-structured Prescriptive The primacy of the interviewer By the interviewer Distant from the subject; uninvolved, objective and ethical Controlled Relatively brief Are set at the outset Usually many Rigid Uniform across respondents Usually many Question-answer Random; constructed before the study, and cannot be changed | Mostly open Not binding The primacy of the respondent By no one Close to the subject; engaged, subjective and ethical Not controlled Relatively long Are guided by the situation Usually few Flexible As required; flexible Usually one Discussion Not random; flexible; expanded during the study |
2 THE INTERVIEWER
In quantitative studies interviewers control the interviewing situation, are distant and uninvolved; in qualitative research they are expected to avoid controlling the Interview situation, to be close to the subject and to be engaged.
2.1 The tasks of the interviewer
The role of interviewers in research varies significantly from one paradigm to another. There are some central tasks, however, which interviewers perform irrespective of the underlying paradigm. The most common of these are listed below,
· Approaching the respondent and arranging the time and date of the interview
· Conducting the interview
· Conducting/guiding the interview, complying with the underlying paradigm
· Ensuring they do not influence the way the respondent answers the questions
· Recording the answers accurately
· Establishing and maintaining positive relations with the public
· Observing ethical standards
| Administrative tasks in structured interviewing · Advertise for Interviewers and supervisors. · Print interview schedules. · Select interviewers, supervisors and personnel. · Train interviewers. · Arrange work conditions, payment and duration of employment. · Arrange pre-interview meeting with respondents (where required), · Deal with refusals to take part in pre-interview meetings. · Choose the time and place of the interview. · Initiate the conduct of interviews. · Deal with refusals to take part in the study. · Arrange field supervision. · Ensure proper conduct of interviews and adherence to ethics. · Arrange the return of completed interviews to the researcher. · Phone respondents randomly to verify completion of interviews. · Check interviews for completeness, and eventual faults and errors. |
2.2 Interviewer selection
Because interviewers are so important to the research, investigators employ a very systematic process when selecting them. The criteria usually considered significant are
· honesty, trustworthiness and self-control
· intelligence, maturity and friendliness
· sociability and social acceptability
· carefulness, conscientiousness and ability to concentrate
· accuracy and dependability
· objectivity, and lack of bias or prejudice
· adaptability, independence and initiative
· verbal ability and ability to listen to others carefully
· interest in and familiarity with the research topic
· ability to work with others in a team of experts
| Attributes of qualitative interviewers · personal qualities (e.g. sensitivity, friendliness) · knowledge of the topic and its context · processing ability that would assist with understanding and interpreting meanings and responding accordingly · interest in and commitment to the research · clarity in speech and thought · adaptability, independence and initiative · initiative in exploring the respondent's mind · creativity in asking questions and devising tasks that would interest the respondent · experience with leading qualitative interviews · personal and professional maturity that encourage trust and respect |
3 THE PROCESS OF INTERVIEWING
In quantitative research, interviewing is conducted in a number of stages, with each stage including certain tasks. The various types of interview are conducted in different ways but some commonalities can be identified in all forms of interviewing.
3.1 Finding respondents
In most cases, the interviewers are supplied with a list of the names and addresses of prospective respondents to approach. They will need to familiarize respondents with aspects of the research such as its purpose and sponsor, assure them of anonymity and confidentiality, explain other details of the research (e.g. duration and possible effects of the interview), seek their cooperation, and either conduct the interview or arrange for it to be conducted at another time.
The interviewer must be prepared in cope with failures to meet the prospective respondent and/or with refusals. The person may not be at home, may have moved away or died, be incapable of communicating or refuse to cooperate. In the latter case, the interviewer should be prepared to gain the respondent's trust by providing relevant evidence (e.g. ID card, reference letters, brief publications of previous studies), stress the significance of the research findings, and demonstrate the significance of the respondent's participation in the study.
| Why do respondents refuse to be interviewed? Respondents may, for instance, · not be interested in the research topic · not agree with the research objective · not find the interviewer friendly, serious, or polite · not have time for the interview · not be sure of the real intentions of the interview · not feel at ease with the interviewer, be suspicious or fearful · not find the research consistent with their professional work/commitment · not have the consent of their spouse, who may be against it |
Interviewers are more likely to win the respondents' cooperation if they know the reasons for them refusing to be interviewed as well for agreeing to do so. Using these factors correctly, fairly and ethically helps to succeed in the search for respondents.
| Why do respondents agree to be interviewed? The reasons for agreeing to be interviewed may be: · Interest: They may be interested in talking about contemporary issues. · Need to talk: there can be emotional relief through talking about one's problems. · Feeling important: They may feel honoured to be chosen to speak on important issues. · Altruism: They may see the interview as a way of contributing to the wellbeing of the community. · Compassion: They may see value in helping interviewers 'to do their Job'. · Politeness: They don't want to disappoint the interviewer. · Curiosity: They are interested in the research topic and in the process of interviewing. · Boredom: 'I didn't have anything better to do at that time' · Sense of obligation: They feel indebted to the sponsor or research agency. |
Another responsibility of the interviewer is to arrange the place and time of the interview.
3.2 Conducting the interview
The conduct of the interview – the manner in which questions are to be asked – depends very much on the type of interview. In structured interviewing the questions are asked exactly as instructed. In general, the interviewer is expected to show interest in the research topic and in the respondent, and should avoid bias, leading questions and suggestive questioning.
Recording too takes place as laid down in the instructions. In many cases this means ticking or crossing boxes on the interview schedule; in other cases responses are recorded verbatim, or at least at length. Open-ended questions may be recorded verbatim or summarized, as the research requires. The use of audio-visual equipment makes the work easier, but the respondent must be asked for approval (which by the way is equally important when notes are taken). Some respondents object to the interviewer taking notes and/or to speaking in front of a microphone.
3.3 Ending the interview
Care should be taken to end the interaction between the interviewer and the respondent smoothly and in a friendly atmosphere, in a spirit of trust, cooperation and mutual respect, so that the respondent feels that the contribution made to the research and to society in general has been appreciated. In many cases, the researcher lays down the way of ending the interview by providing a sample format that the interviewer has to memorize and use at the end of the session.
3.4 Field supervision and checks
Despite the strong involvement of interviewers in data collection, or perhaps because of this, its administration is not left entirely up to them. Supervision and checking are essential, particularly when more than one interviewer is employed. Some interviewers may not be as thorough as they should be, and cheating is always a possibility. Usually, the researcher or supervisor checks the returned interviews for completeness, too much or too little probing, and inadequate recording of the data. Apart from this, the supervisor contacts respondents randomly to ensure, First of all, whether the intended interview has taken place, and then whether it was conducted as planned. Checks (or bias, honesty, politeness, objectivity, ethics and interviewer-respondent relationships are part of this procedure.
4 INTERVIEWER-RESPONDENT RELATIONSHIP
The type of relationship that develops between the interviewer and the respondent depends on factors such as background methodology, the research topic and objective and the style of interviewing. However, some common features of the relationship involve criteria that should be adhered to in ail situations.
Background and appearance
Interviewers are chosen to match the respondents in aspects that are thought to make the respondents feel comfortable and at ease; these include social background, gender, race, ethnicity, age and personality. A good example is taken here from the feminist paradigm, which stresses emphatically that women can only, effectively studied by women. As we already know, feminist research has been seen as research not only about women but also by women.
The logic behind this is that successful interviews are associated with positive and effective relationships, and these grow where interviewers and respondents come from a similar background. Background similarity makes the entry to the respondent's world easier; promotes trust, understanding and cooperation; and allows the development of a close and rewarding relationship between the interviewer and the respondent. Further women are equipped with the female 'lens' that allows a more effective study of women than the lens of men. Clothing and grooming are also significant. The interviewer is expected to be dressed neutrally, ideally in a way similar to that of the respondent, and unobtrusively, so that the centre of the interview is the research topic and not the interviewer.
Status of the parties
Interviewing should avoid patronizing the respondents, showing disbelief in statements given, or judging the answers. They should not appear as wise judges but rather as interested researchers who wish to learn from the respondents. Neither should they encourage or discourage certain types of answers. They should be neutral, receptive and eager to know the respondents' views, which for the interviewer are interesting and valuable.
5 PROMPTING AND PROBING
Prompts and probes are very common in interviewing, either to make it easier for respondents to answer questions, or to encourage them to continue with their response. More specifically, a prompt is a part of a question that offers a list of possible answers, from which the respondent is expected to select one or more. An example is the question 'Which religious activities have you attended during the past four weeks?' with the respondent being read or shown a list of such activities to choose from. This makes answers to certain questions easier and more accurate. Prompts are usually developed after careful pre-tests and pilot studies.
Probes are questions or neutral statements that encourage the respondents to extend or amplify a partial, irrelevant or inaccurate response, and/or to stimulate and assist them to answer a question, without affecting the direction of their thinking and without causing bias or distortion. Probes are employed in interviews where open discussion is allowed, such as intensive interviewing. There are at least two types of probe: non-directive probing and summary technique. Examples of probes are:
Controlled non-directive probing entails a specific but non-directive comment or question. For instance, when the respondent gives an incomplete, inadequate or general answer, the interviewer may comment: ‘That’s interesting, tell me more about it!’, or ‘What do you mean by that?’ or ‘What would be an example for that?’, or ‘I see’ or just remain silent. The summary technique encourages respondents to continue their comment by summarizing the respondent’s last statement and waiting for them to add new information. An example would be when the interviewer following, for instance, the respondent’s silence adds: OK, this was the first step…..’, followed by silence.
6 NARRATIVE INTERVIEW (NI)
6.1 Introduction
The narrative interview is both an interview and an autobiographic method of data collection. It is therefore also referred to as the autobiographic narrative interview, or simply autobiographic interview. It entails content analysis, and this helps it to make use of the advantages of three methods.
Narrative interviews tend to be closer to life and more natural than quantitative ones, and assign a relaxed and casual role to the interviewer. More specifically, the NI is a form of communication with people, and in this sense communication refers to everyday life situations and experiences. Hence, the interview is constructed as a natural communication process, whereby the interviewer listens, encourages the interviewee to continue with the story, and does not interfere, interrupt, distract or disturb the interviewee in any way.
The NI places a strong emphasis on the narrative, which is seen as a language game that relates closely to experiences of the storyteller. The narrative is thought to reflect the teller’s thinking processes, cultural patterns and determinants that guide or even dominate his/her life choices. The NI allows the interviewee full freedom of expression without limits posed by questions, the interviewer, time or set conditions. Not bring 'guided' by the interviewer, the interviewee can express any views, opinions and ideas and can concentrate on any points of the topic at will.
6.2 Process of interviewing
The process of narrative interviews entails three elements: the main story, the narrative inquiry and the conclusion.
Stage 1: Introduction
· The interviewee is introduced to the interview situation, and is familiarize with the expectations and overall framework of the interview, including ethical standards, anonymity, confidentiality, and so on.
· The topic is introduced, outlined, and its dimensions explained in detail.
· The interviewer introduces a question that motivates the interviewee to talk about his/her personal life and experiences.
· Arrangements for (audio-video) recording the interview are made.
· The interviewer does not ask direct questions about motives, reasons, causes and the like; instead he/she leaves it up to the informant to refer to such issues if he/she comes to it as a part of the narrative.
Stage 2: Narrative
· The interviewee is encouraged through a specific stimulus to talk freely about life experiences. The interviewer here does not interfere, but remains an 'interested listener', only making remarks that encourage the teller to continue, indicating that the interviewer is listening carefully. This is shown in brief verbal expressions and simple gestures. The interviewee chooses the events (s)he considers most relevant, expands on topics considered relevant and important, stops whenever necessary, presents events in order of importance or as the memory dictates, and is free to decide the order of presentation, the events to talk about (e.g. childhood, adulthood, present time, or pre-war, wartime and post-war years).
· The interviewee is encouraged not only to describe personal and social experiences but also to compare them with other experiences and to explain these events as (s)he understands them. The NI motivates and enables the teller to transport past events to the present, to become aware of the experiences, to travel back to the old times and to relive them once again through telling.
Stage 3: Questions
· When the interviewee indicates the end of the story, the interviewer can ask for more information where gaps have emerged, or for an explanation when statements were unclear, ambiguous or incomplete. A characteristic of this stage is that questions are asked to gain information and not to criticize or pass judgment; the process continues to be a part of the narrative, where the story teller otters his/her wisdom of the researcher.
· If new points surface that require further story-telling, the interviewee is encouraged to begin again talking about this topic, bringing the NI back to Stage 2. Only when this stage is fully exhausted, do they proceed to 'questions' and then to 'explanation'.
Stage 4: Explanation
· At this stage, and after the basic information has been fully described to the satisfaction of the interviewee and the interviewer, more direct questions are asked. Here the interviewee is asked, first, to establish more general and abstract views of the situation and its regularities, identifying recurring events, and developing abstractions and systematic interconnections; second, to describe more general aspects of the issue in question, being given the opportunity to demonstrate his/her capacity in assessing this as well as to offer a more abstract explanation of the situation; third, to explain motives and intentions; and finally, to discuss with the interviewer the meaning of the story.
· If new points surface that require further story-telling, the interviewee is encouraged to begin talking about this topic again, bringing the NI back to Stage 2. Only when this stage is fully exhausted, docs die researcher move to the next step.
Stage 5: Analysis
· The narrative as well as the debate that follows is transcribed. The methods of transcription employed for this purpose are similar to those employed in any other interview, including numbering the text lines and so on, although with a difference. Here, as Pfeifer (2000; 71) notes, emphasis is placed on 'Sound', focusing on vocal emphasis and nuance rather than on written forms. The transcript is expected to reproduce the nature of the discussion as it happened, including pauses, hesitations, fluctuation in the tone of the voice, speed of talk, emotionality and similar features. As a rule, the transcript is prepared by the interviewer, who has a good knowledge of the nature of the interview and the speech in particular.
· The transcribed text is subjected to content analysis. This is known as 'conversation analysis', which is central to this method of data analysis.
The process of analysis is very complex and very demanding, perhaps more demanding than in a simple quantitative structured interview. A more detailed, but still summarized and restricted description of this analysis is given later in this volume. What is important here is that complexity, diversity and intensity in this research model result in a high quality of findings, which is the reason for employing narrative interviewing.
7 INTENSIVE INTERVIEWING / IN-DEPTH INTERVIEWS
This is an unstructured interview technique employed by social researchers as either the sole method of research or in addition to other methods. The technique is also known as 'in-depth' interviews. It is a very sophisticated technique, requiring adequate knowledge of the research topic, extensive experience of interviewing, and an ability to communication effectively so as to establish and maintain trust and rapport with the respondents
The format of intensive interviewing is unstructured and flexible. There are no specific questions to be asked. The interviewer is expected to develop the questions when they are required and as they best fit into the interview situation. For this reason, the actual formulation and order of the questions might differ from interview to interview. Intensive interviews are usually long and may extend over two or more sessions. As a research instrument they have strengths and weaknesses that researchers must be aware of.
Strengths of intensive interviewing
· the rapport in the relationship between the interviewer and the respondents which makes the respondent feel important
· the degree of commitment of the parties and their interest in the relationship and discussion
· the unstructured process of interviewing and the opportunity to cover issues not covered in the interview schedule
· the flexibility, continuity of thought, freedom of probing, and evaluation of behaviour during the interview
· the high quality of information obtained
· the ability of the interviewer to use listening and empathy as significant tools of interaction
Weaknesses of intensive interviewing
· dependence on the skills, values, standards and ideology of the interviewer
· inability to generalize the findings
· inability to facilitate comparisons between cases studied since the elements of the interview may differ from case to case
· high demand on time and resources
Intensive interviews are conducted like any other interview, so the general rules of interviewing introduced above also apply here. The lack of strict boundaries and the freedom of choice in the formulation of questions and the order of their presentation, as well as the primary focus on the interviewee, transform the general rules of interviewing accordingly, and also guide the process of interviewing to a path that leads closer to the world of the interviewee.
8 TELEPHONE INTERVIEWING
Telephone interviewing demonstrates the same structural characteristics as standard interviewing techniques, except that it is conducted by telephone. Although this difference might not appear to be very significant, it does have certain effects on some aspects of the research process. In the first instance, questions have to be constructed in a way that will allow a clear understanding of their content when presented over the telephone. Also, telephone interviews cannot be as lengthy as face-to-face ones. Finally, they have to be introduced in a manner that, on the one hand, will encourage the respondent to take part in the study and, on the other hand, will meet ethical standards. This in turn strengthens the chances of a prospective respondent agreeing to be interviewed. Such introductions contain information similar to that: included in cover letters of mail questionnaires.
Telephone interviewing is employed when the interviews are simple and brief, when quick and inexpensive results are sought, when it is not necessary to approach the respondent face to face, and when sampling inaccuracies (e.g. nonsubscribers and unlisted numbers) are not considered important. Where appropriate, this type of interviewing offers many advantages (e.g. Benini, 2000, Mahr, 1995), but also some limitations.
Advantages of telephone interviewing
· is less labour-intensive than other methods
· allows the study of relatively large samples
· is relatively economical
· produces quick results
· allows more open communication since the respondent is not confronted with the interviewer
· reduces bias since factors such as race, ethnicity, appearance and age do not influence the respondents
· offers more anonymity than other techniques, particularly when random-digit-dialing techniques are used
Limitations of telephone interviewing
· high refusal rate
· inability to determine the identity of the respondent
· limited access to research topics
· relative inability to control the interview fully
· limited access to target population (people without a telephone or with unlisted numbers are not accessible)
Despite these problems, the advantages outweigh die weaknesses of this method, which is a very .useful and popular technique of data collection.
9 THE INTERVIEW CONTEXT
Interviewing is not: a straightforward process in which the interviewee freely surrenders information as required by the questions. Giving answers is a process that is affected by many situational factors, which can affect whether an answer is forthcoming as well as the truthfulness of the response. The nature of the topic, the interviewer and the interview situation are some of these factors. The latter may include aspects such as the way in which the question is presented, the environment in which interviewing takes place and the extent to winch the questions allow comments and interpretation. For instance, an interview regarding drug use by teenagers will produce different results when conducted in the schoolyard, at home in the presence of parents, or in the police station.
Knowing these factors, the interviewer can achieve better and more accurate results, by consider the following factors to be relevant.
· Understanding the question: The clearer the question, the more likely it is to receive an answer. Understanding can be hampered by factors associated with the interviewer, such as the tone of voice adopted, which may divert the respondent's attention, as well as conceptual factors, such as taking the question in general rather than in specific terms, and the respondent's capacity to understand its literal or conceptual content.
· Cognitive processing: Normally, messages are processed, connected with data held in the memory, and direct the respondent to the correct answer. Then the interviewees evaluate the suitability of the answer in terms of the accuracy of the response, as well as the expected assessment of the answer within attitudes, opinions, experiences, and facts. Finally, pieces of information will be put together in form a response. This process can be affected by the manner and context in which the question is presented, by the fact that they fed obliged to give an answer, and other similar factors.
· Precision: Questions requiring a high level of precision (accuracy) are more likely to produce inaccurate answers. This can be due to lack of knowledge and of access to relevant information required to answer the question. This, apart from anything else, may affect the self-esteem of the respondent, which in turn might have an impact on the remaining part of the interview.
· Authority: Respondents are least likely to otter truthful answers to sensitive questions when a person in authority (employer, parent, spouse, teacher, police officer etc.) is present. Neutral environments should be considered.
· Social standards: Respondents usually do not give answers that violate general social standards. They weigh the suitability of their answers against social expectations, and give an answer which will not transgress social or other standards and hence, in their view, will not receive negative social evaluation. Hence, there may be cases where the researcher receives socially desirable but not necessarily correct answers. It is therefore important that the interviewer and the researcher ensure that sensitive questions are formulated in a manner that neutralizes external factors in interviewing. Three factors are of particular importance. These are: social desirability, relating to cultural norms and values, situational expectancy, relating to contextual requirements and expectations; and the role of the interviewee, relating to the perceived roles within the interview setting, such as the definition of the interviewer and interviewee and the implied boundaries.
· Decision bias: Respondents are confronted with a series of situations in which they have to make important decisions. These decisions direct performance in interviews and this can result in an unavoidable bias, such as conformity bias, preference bias, adjustment bias or incompleteness. The respondent answers correctly when the answer entails no bias and no high personal or social costs.
The rule in these and similar circumstances is that the interviewer and the researcher in general should construct the interview in such a way that it will avoid placing respondents in a situation that will compromise their position and hence the truthfulness of their answers.
ADVANTAGES AND LIMITATIONS OF INTERVIEWING
Advantages
· Flexibility: Interviews can be adjusted to meet many diverse situations.
· High response rate: Interviewing attracts a relatively high response rate-
· Easy administration: Interviews do not require respondents to have the ability to read or to handle complex documents or long questionnaires.
· Opportunity to observe non-verbal behavior: Such opportunities are obviously not available when questionnaires or indirect methods are used.
· Less tedium: Less patience and motivation is required than in other methods. Interviews need 'participation', not just: 'response'. Participation involves another person with whom the respondent interacts to complete the task; hence interviewing is often perceived as a cooperative venture rather than a one-sided exercise.
· Control over the environment: Here the interviewer has an opportunity to control the conditions under which the questions are answered.
· Capacity for correcting misunderstandings by respondents: Such an option is very valuable and not available in other forms of data collection, such as questionnaires.
· Control over the order of the questions: Respondents have no opportunity to know what question comes next or to alter the order of the questions they answer. When the order of die questions is significant, an interview is much more useful than a questionnaire.
· Opportunity to record spontaneous answers: The respondent does not have as much time available to answer questions as when questionnaires are employed. When spontaneity is important, interviews offer a real advantage over other methods.
· Control over the identity of the respondent: when interviews are employed, the identity of the respondent is known; this is not so when other methods (e.g. questionnaires) are used.
· Completeness: The fact that the interviewer presents the questions guarantees that all questions will be attempted and the interview will be complete.
· Control over the time, date and place of the interview: Interviews can be conducted exactly as planned, regarding the time and date, and according to specified conditions. Such a guarantee cannot be given when questionnaires are used.
· Ability to handle complexity: More complex questions can be used because the presence of the interviewer can assist in answering them.
· Length: Greater length is possible in interviewing that when other methods (e.g. questionnaires) are used.
Limitations
· Cost: Interviews are more costly and time consuming than some other methods; such as questionnaires.
· Bias: Interviews are affected by the 'interviewer' factor and the possible bias associated with it,
· Inconvenience: Interviewing is less convenient than other methods, such as questionnaires.
· Lack of anonymity: It offers less anonymity than other methods since the interviewer knows the identity, residence, type of housing, family conditions and other personal details of the respondent.
· Sensitivity: It is less suitable than other methods when sensitive issues are discussed. For example, many people prefer to write about sensitive issues rather than to talk about them.
11 PROBLEMS AND ERRORS IN INTERVIEWING
Problems associated with the nature of the method include data recording, evaluation errors and instruction errors.
Recording errors
· selective hearing or vision
· misunderstanding of the respondent
· too-early or too-late registration of the respondent
· incomplete, faulty or illegible responses
Evaluation errors
· Leniency effect, when extremely negative responses are avoided
· Severity effect, when extremely positive responses are avoided
· Projection effect, when personal prejudice and stereotypes are projected onto the respondent, affecting perception and evaluation of responses
· Contact effect, when loss of objectivity caused by knowing the respondent leads to a mild evaluation of responses
· Central tendency effect, when the researcher tends to avoid recording extreme responses
· Reference-group effect, when the researcher develops expectations related to the reference group of the respondent and judges the responses according to these expectations, examples being:
- the grandpa effect, expecting too little from the respondent
- the authority effect, feeling intimidated by the respondent’s position of authority
- the Santa Claus effect, expecting more too much from the respondent (e.g. treating people thought to be more intelligent or sympathetic in a more forthcoming and understanding way etc.)
- the identification effect, generating errors associated with the researcher’s tendency to identify with the respondent and therefore treat him or her more mildly, and vice versa.
Instruction errors
· Replacing non-response with another person’s responses
· Withhold information collected
· Introducing changes in procedure against the researcher’s instructions, for example changing questions or the order of the questions
· Forgery of parts of data
· Showing consent or rejection of responses while collecting data
12 NON-RESPONSE IN INTERVIEWING
Unwillingness and/or inability to take part, is a serious problem especially in quantitative research, where sampling is strict and deterministic, and where replacement of respondents is not permitted. It is true that approaching the respondents persistently until they relent is reported to be a relatively successful method of combating non-response. Offering rewards to 'difficult' respondent is another. In addition, when interviews are employed, it is suggested that consideration of issues such as the following may prove helpful:
· Approach: The more personal, honest, brief and pleasant the approach when visiting the respondent for the first time, the more likely it is that the respondent will participate in the study.
· Explanation: The dearer and fuller the explanation of the survey, including the sponsor, purpose, time required, anonymity and confidentiality, the more likely it is that the respondent will agree to take part in the study.
· Appeal: The more appealing and attractive the presentation of the study, and the more challenging and inviting the description of the topic, the more likely it is that the respondent will agree to be interviewed.
· Honesty: The more honest, polite, frank, reassuring and modest the researcher, the more likely it is that the respondent will take part in the study,
· Respect: The more respected the respondent feels in his or her role as an 'expert' to be consulted rather than as just a source of information, the more likely it is that the interview will take place.
· Trust: The more successful the researcher is in eliminating mistrust, insecurity, fears, confusion, doubts and ambiguity, the more likely the respondent is to agree to take part in the study.
· Impression: The more impressed the respondent is with the overall sincerity and appearance of the study, the more likely it is for the interview to take place.
· Arrangements: The more favourable and convenient the interview arrangements are for the respondent, and the more considerate the approach (e.g. phoning before visiting), the more likely it is for the interview to be completed.
· Friendliness: The more friendly and discreet the researcher, the more likely it is for the respondent to agree to be interviewed.
· Sponsors: Interviews conducted for the government, universities or other institutions are more likely to attract the cooperation of the respondent than interviews conducted for other sponsors or for the personal interest of the interviewer.
· Purpose: The more worthwhile the purpose of the interview, the more likely it is for the respondent to agree to be interviewed.
In brief, the most successful way of reducing the non-response rate is prevention. Knowing how respondents react to elements of the questionnaire or the survey helps researchers to plan its structure so that non-response can be significantly reduced.
13 INTERVIEWING IN THE COMPUTER AGE
New developments in the field of IT have had a positive impact on the practice of interviewing. Interviews can now be held between the interviewee and a computer, and where researchers are also present, their input is often restricted to giving instructions and to helping the respondent. The following are some examples of computer packages that are relevant to interviewing.
Computer-aided personal interview (CAPI): In CAPI, questioning and control of the responses is done through the computer. In this case, the interviewee, sitting in front of the computer, reads the questions appearing on the screen, and enters the responses as advised. The computer processes the responses automatically and prepares them for analysis. This program can be used by single interviewees or with a group of respondents. The researcher simply notes the identity of the respondents leads them to the computer, and after completion of the computer entry, makes the usual closure of the session.
Computer-driven self-completion interview (CODSCI): This is a fully automated interviewing program in which there is no need for a human interviewer. The interview is carried out in a computer session in which the respondent reads the questions from the screen in direct communication with the computer; the computer 'asks' the questions, explains problems and provides help in answering the questions. After completion of the interview, the responses are saved automatically in the memory and added to previous interview data.
Computer-aided telephone interview (CATI): Here the computer is used by the interviewer, who reads the questions to the interviewee over the telephone as they appear on the screen and records the response in the computer. The computer can do more than just present the questions and receive the answers. It can draw the sample, choose the telephone number, dial the respondent through a self-dial system and 'connect' the interviewer with the interviewee.
Computer-integrated survey research (CISUR): This fully integrated computer program is more advanced than the previous ones in that it contains more functions in a wider area of the research process. This is another example of how computers are entering areas of research above and beyond the statistical analysis of data, gradually replacing th human element.
‘The data collector’: This program works on the same principles as the others and have similar goals and functions. Although a data collection device, it has additional features that make it a useful tool for social research. The program assists with the development of questionnaires and interview schedules, and provides measures for quantitative data analysis, such as descriptive statistics, frequency distributions, chi-square, analysis of variance, Mann-Whitney tests and correlations (r, rho and lagged correlations), as well as for qualitative analysis, for instance word counts, word/phrase searches and a variety of other options for searching text.
There are many more computer programs employed in this field. The few listed above are only examples of, what computers can offer when interviewing is the methods of data collection.
14 SURVEY METHODS IN COMPARISON
It has often been asked which survey method is the most appropriate. The correct answer is that all methods are effective and useful depending on the type of information required. They all are suitable for certain topics, certain respondents and certain types of information. Table 12.2 offers a brief overview of the characteristic of these survey methods.
Table 12.2 A Comparison of face-to-face, telephone and mail surveys
| | Type of surveys | ||
| Major considerations | Face-to-face | Telephone | Mail |
| Access to respondents | Limited only by hearing or speech impairment or disability | As in face-to-face plus no access to phone, or unlisted numbers | Limited by disability or illiteracy |
| Length of questionnaire | Long | Medium | Short |
| Complexity of questionnaire | High | High | Low |
| Addressing sensitive topics | Fair | Good | Good |
| Response rate | Very high | High | Low |
| Completion time | High | Low | High |
| Interviewer bias | High | Moderate | Nil |
| Control over respondent's identity | Very high | Low | Very low |
| Allowed complexity of instrument | Very high | Moderate | Very high |
| Opportunity to probe | Very high | High | Nil |
| Asking sensitive questions | Least suitable | Just suitable | Most suitable |
| Costs | Very high | Moderate | Low |
| Inconvenience | High | Moderate | Low |
MAIN POINTS
· Interviews are orally conducted surveys, are very common and appear in many and diverse forms
· Interviews can be structured or unstructured, standardized or un-standardized.
· Interviews are conducted by an interviewer face-to-face with the respondent or by telephone
· In qualitative research, interviews are single and personal, employ open-ended questions and are open and flexible.
· Choosing interviewers who are similar in background to the respondents not only makes entry into the respondents' world easier but also promotes trust, mutual understanding and cooperation and therefore reduces bias and distortion.
· Telephone interviewing produces quick results, can study large samples, is relatively economical, promotes open communication, reduces bias and guarantees more anonymity than face-to-face interviews.
· Errors in interviewing may be associated with inappropriate recording of data or evaluation of responses, or with failure to follow instructions given to interviewers.
· Interviews are employed in quantitative and qualitative research, although in different forms.
· Narrative interviews introduce a topic for discussion and encourage the respondent to offer as much information as possible. They are becoming increasingly popular and are mostly used by qualitative researchers,
· Intensive interviews are mostly unstructured and un-standardized, aiming at an in-depth exploration of the issues in question.
No comments:
Post a Comment