CHAPTER

10
STRUCTURED
INTERVIEWING
Introduction 208 Question order 217
Probing 219
The structured interview 208
Prompting 220
Reducing error due to interviewer variability 208
Leaving the interview 221
Accuracy and ease of data processing 210
Training and supervision 221
Other types of interview 210
Other approaches to structured interviewing 222
Interview contexts 212 The critical incident method 222
More than one interviewee 212 Projective methods, pictorial methods,
More than one interviewer 212 and photo-elicitation  223
In person or by telephone? 212 The verbal protocol approach 226
Computer-assisted interviewing 214 Problems with structured interviewing 226
Conducting interviews 215 Characteristics of interviewers 226
Know the schedule 215 Response sets 227
Introducing the research 215 The problem of meaning 228
Rapport 216 Key points 229
Asking questions 216
Recording answers 217
Questions for review 229

Clear instructions 217

CHAPTER OUTLINE

Once sampling issues have been taken into consideration, the next stage of the survey research process
(see Figure 9.1) involves considering whether to administer the questionnaire face-to-face or to rely on
self-completion. This chapter deals with the first option, the structured interview, while Chapter 11
addresses issues relating to self-completion. A further option to consider is whether to administer the
questionnaire by email or online; this will also be covered in Chapter 11.
The structured interview is one of a variety of forms of research interview, but it is the one that is most
commonly employed in survey research. The goal of the structured interview is for the interviewing of
208 10 Structured interviewing

respondents to be standardized so that differences between interviews in any research project are
minimized. As a result, there are many guidelines about how structured interviewing should be carried out
so that variation in the conduct of interviews is small. This chapter explores a number of aspects of this:

• the reasons why the structured interview is a prominent research method in survey research, with
consideration of the importance of standardization to the process of measurement;

• the different contexts of interviewing, such as the use of more than one interviewer and whether the
administration of the interview is in person or by telephone;

• various requirements of structured interviewing, including establishing rapport with the interviewee,
asking questions as they appear on the interview schedule, recording exactly what is said by
interviewees, ensuring there are clear instructions on the interview schedule concerning question
sequencing and the recording of answers, and keeping to the question order as it appears on the
schedule;

• problems with structured interviewing, including the influence of the interviewer on respondents and
the possibility of systematic bias in answers (known as response sets).

Introduction
The interview is a common occurrence in social life, In the business research interview, the aim is for the
because there are many different forms of interview. interviewer to elicit from the interviewee or respondent,
There are job interviews, media interviews, social work as he or she is frequently called in survey research, all
interviews, police interviews, appraisal interviews, to manner of information: the interviewee’s own behaviour
name a few. And then there are research interviews, or that of others; attitudes; norms; beliefs; and values.
which represent the kind of interview that will be covered There are many different types or styles of research inter-
in this and other chapters. These different kinds of inter- view, but the kind that is primarily employed in survey
views share some common features, such as the eliciting research is the structured interview, which is the focus
of information by the interviewer from the interviewee of this chapter. Other kinds of interview will be briefly
and the operation of rules of varying degrees of formality mentioned in this chapter but will be discussed in greater
or explicitness concerning the conduct of the interview. detail in later chapters.

The structured interview

The research interview is a prominent data collection Reducing error due to interviewer
strategy in both quantitative and qualitative research. The
variability
social survey is probably the chief context within which
business researchers employ the structured interview The standardization of both the asking of questions and
(see Key concept 10.1) in connection with quantitative the recording of answers means that, if the interview is
research, and it is this form of the interview that will properly executed, variation in people’s replies will be
be emphasized in this chapter. The reason why survey due to ‘true’ or ‘real’ variation and not due to the inter-
researchers typically prefer this kind of interview is view context. To take a simple illustration, when we ask
that it promotes standardization of both the asking of a question that is supposed to be an indicator of a con-
questions and the recording of answers. This feature cept, we want to keep error to a minimum, an issue that
has two closely related virtues from the perspective was touched on at the end of Chapter 9. We can think of
of quantitative research: the minimizing of error that the answers to a question as constituting the values that
could result from interviewer variability; and the ease a variable takes. These values, of course, exhibit varia-
of data processing provided by the standardized record- tion. This could be the question on skill development and
ing of answers. training among employees that was a focus of Chapter
 The structured interview 209

10.1 KEY CONCEPT

What is a structured interview?
A structured interview, sometimes called a standardized interview, involves the administration of an interview
schedule: a set of questions designed to be asked by an interviewer. The aim is for all interviewees to be given
exactly the same context of questioning. This means that each respondent receives exactly the same interview
stimulus as any other. The goal of this style of interviewing is to ensure that interviewees’ replies can be
aggregated, and this can be achieved reliably only if those replies are in response to identical cues. Interviewers
are supposed to read out questions exactly as they are printed on the schedule, and in the same order. Questions
are usually very specific and often offer the interviewee a fixed range of answers (this type of question is often
called closed, closed ended, pre-coded, or fixed choice). The structured interview is the typical form of interview in
social survey research.

9 at certain points. However, some respondents may be to such things as confusing terms or ambiguity (an issue
inaccurately classified in terms of the variable. There are that will be examined in Chapter 12, ‘Asking questions’),
several possible reasons for this. we want to be able to say as far as possible that the varia-
Most variables will contain an element of error, so that tion that we find is connected with true variation between
it is helpful to think of variation as made up of two com- interviewees and not to variation in the way a question
ponents: true variation and error. In other words: was asked or the way the answers were recorded. Inter-
viewer variability can occur in either of two ways: first,
variation = true variation + variation due to error
intra-interviewer variability, whereby an interviewer is
The aim is to keep the error component to a minimum, not consistent in the way he or she asks questions and/
since error has an adverse effect on the validity of a mea- or records answers; second, when there is more than one
sure. If the error component is quite high, validity will interviewer, there may be inter-interviewer variability,
be jeopardized. In the structured interview, two sources whereby interviewers are not consistent with each other
of variation due to error—numbers 2 and 5 in Tips and in the ways they ask questions and/or record answers.
skills ‘Common sources of error in survey research’—are Needless to say, these two sources of variability are not
likely to be less pronounced, since the opportunity for mutually exclusive; they can coexist, compounding the
variation in interviewer behaviour in these two areas problem even further. In view of the significance of
(asking questions and recording answers) is reduced. standardization, it is hardly surprising that some writ-
The significance of standardization and of thereby ers prefer to call the structured interview a standardized
reducing interviewer variability is this: assuming that interview (e.g. Oppenheim 1992) or standardized survey
there is no problem with an interview question owing interview (e.g. Fowler and Mangione 1990).

TIPS AND SKILLS

Common sources of error in survey research
There are many sources of error in survey research, in addition to those associated with sampling. This is a list of
the principal sources of error:

1. a poorly worded question;

2. the way the question is asked by the interviewer;
3. misunderstanding on the part of the interviewee;
4. memory problems on the part of the interviewee;
5. the way the information is recorded by the interviewer;
6. the way the information is processed, either when answers are coded or when data are entered into the computer.
210 10 Structured interviewing

Accuracy and ease of data processing answers to a certain question. A number will then be
allocated to each category of answer so that the answers
Like self-completion questionnaires (discussed in Chapter can be entered into a computer database and analysed
11), most structured interviews contain mainly questions quantitatively. This general process is known as coding,
that are variously referred to as closed, closed ended, pre- and will be examined further in Chapter 11.
coded, or fixed choice. This issue will be covered in detail in Coding introduces yet another source of error. First, if
Chapter 11. However, this type of question has consider- the rules for assigning answers to categories, collectively
able relevance to the current discussion. With the closed known as the coding frame, are flawed, the variation that
question, the respondent is given a limited choice of pos- is observed will not reflect the true variation in inter-
sible answers. In other words, the interviewer provides viewees’ replies. Secondly, there may be variability in the
respondents with two or more possible answers and asks ways in which answers are categorized by the rater (the
them to select which one or ones apply. Ideally, this proce- person doing the coding). As with interviewing, there
dure will simply entail the interviewer placing a tick in a can be two sources: intra-rater variability, whereby the
box by the answer(s) selected by a respondent, or circling rater varies over time in the way he or she applies the
the selected answer, or a similar procedure. The advan- rules for assigning answers to categories; and inter-rater
tage of this practice is that the potential for interviewer variability, whereby raters differ from each other in the
variability is reduced: there is no problem of whether the way they apply the rules for assigning answers to catego-
interviewer writes down everything that the respondent ries. If either (or both) source(s) of variability occur, at
says or of misinterpretation of the reply given. If an open least part of the variation in interviewees’ replies will not
or open-ended question is asked, the interviewer may not reflect true variation and instead will be caused by error.
write down everything said, may embellish what is said, The closed question sidesteps this problem neatly,
or may misinterpret what is said. because respondents allocate themselves to categories.
However, the advantages of the closed question in the The coding process is then a simple matter of attach-
context of survey research go further than this, as we ing a different number to each category of answer and
will see in Chapter 11. One advantage that is particu- of entering the numbers into a computer database. It is
larly ­significant in the context of the present discussion is not surprising, therefore, that this type of question is
that closed questions greatly facilitate the processing of often referred to as pre-coded, because decisions about
data. When an open question is asked, the answers need the coding of answers are typically undertaken as part of
to be sifted and coded in order for the data to be analysed the design of the schedule—that is, before any respon-
quantitatively. Not only is this a laborious procedure, dents have actually been asked questions. There is very
particularly if there is a large number of open questions little opportunity for interviewers or raters to vary in the
and/or of respondents; it also introduces the potential recording or the coding of answers. Of course, if some
for another source of error, number 6 in Tips and skills respondents misunderstand any terms in the alternative
‘Common sources of error in survey research’: it is quite answers with which they are presented, or if the answers
likely that error will be introduced as a result of vari- do not adequately cover the appropriate range of pos-
ability in the coding of answers. When open questions sibilities, the question will not provide a valid measure.
are asked, the interviewer is supposed to write down as However, that is a separate issue. The chief points to reg-
much of what is said as possible. Answers can, therefore, ister about closed questions for the moment is that, when
be in the form of several sentences. These answers have compared to open questions, they reduce one potential
to be examined and then categorized, so that each per- source of error and are much easier to process for quanti-
son’s answer can be aggregated with other respondents’ tative data analysis.

Other types of interview

The structured interview is by no means the only type of terms have been employed by writers on research meth-
interview, but it is certainly the main type that is likely ods to distinguish the diverse forms of research inter-
to be encountered in survey research and in quantita- view. Key concept 9.1 represents an attempt to capture
tive research generally. Unfortunately, a host of different some of the major terms and types.
 Other types of interview 211

10.2 KEY CONCEPT

Major types of interview
• Structured interview: see Key concept 10.1.
• Standardized interview: see Key concept 10.1.
• Semi-structured interview. This is a term that covers a wide range of instances. It typically refers to a context
in which the interviewer has a series of questions that are in the general form of an interview schedule but is
able to vary the sequence of questions. The questions are frequently somewhat more general in their frame of
reference than that typically found in a structured interview schedule. Also, the interviewer usually has some
latitude to ask further questions in response to what are seen as significant replies. Sometimes such interview
schedules will include closed questions, typically to gather data on variables such as gender, age, income, and
job status, which may be needed as background or for analysis of responses to more open questions.
• Unstructured interview. The interviewer typically has only a list of topics or issues, often called an interview
guide or aide-mémoire, that are covered. The style of questioning is usually informal. The phrasing and
sequencing of questions will vary from interview to interview.
• Intensive interview. This term is employed by Lofland and Lofland (1995) as an alternative term to the
unstructured interview. Spradley (1979) uses the term ethnographic interview to describe a form of interview
that is also more or less synonymous with the unstructured interview.
• Qualitative interview. For some writers (e.g. Mason 1996), this term seems to denote an unstructured interview,
but more frequently it is a general term that embraces interviews of both the semi-structured and unstructured
kind (e.g. Rubin and Rubin 1995).
• In-depth interview. Like the term qualitative interview, this one sometimes refers to an unstructured interview
but more often refers to both semi-structured and unstructured interviewing.
• Focused interview. This is a term devised by Merton et al. (1956) to refer to an interview using predominantly
open questions to ask interviewees about a specific situation or event that is relevant to them and of interest to
the researcher.
• Focus group. This is the same as the focused interview, but interviewees discuss the specific issue in groups.
See Key concept 21.1 for a more detailed definition.
• Group interview. Some writers see this term as synonymous with the focus group, but a distinction may be
made between the latter and a situation in which members of a group discuss a variety of matters that may be
only partially related.
• Oral history interview. This is an unstructured or semi-structured interview in which the respondent is asked to
recall events from his or her past and to reflect on them (discussed in Chapter 20). There is usually a cluster of
fairly specific research concerns to do with a particular epoch or event, so there is some resemblance to a
focused interview.
• Life history interview. This is similar to the oral history interview, but the aim of this type of unstructured
interview is to glean information on the entire biography of each respondent (see Chapter 20).

All of the forms of interview outlined in Key concept asking of questions and recording of answers makes
10.2, with the exception of the structured interview respondents’ replies difficult to aggregate and to pro-
and the standardized interview, are primarily used in cess. This is not to say that they have no role at all.
connection with qualitative research, and it is in that For example, the unstructured interview can have a
context that they will be encountered again in Part useful role in helping the researcher to develop the
Three of this book. They are rarely used in connection fixed-choice alternatives with which respondents are
with quantitative research, and survey research in par- provided in the kind of closed question that is typical of
ticular, because the absence of standardization in the the structured interview.
212 10 Structured interviewing

Interview contexts
In an archetypal interview, an interviewer stands or sits In person or by telephone?
in front of the respondent, asking the latter a series of
A third departure from the archetype is that interviews
questions and writing down the answers. Although this
may be conducted by telephone rather than face-to-face.
archetype is the most usual context for an interview,
While telephone interviewing is quite common in such
there are several possible departures from it.
fields as market research, it is less common in business
research. Telephone interviewing can be carried out
More than one interviewee using landline, mobile or, increasingly, web-based appli-
cations such as Skype. See Research in focus 10.3 for an
In the case of group interviews or focus groups, there is
example of the use of telephone interviewing and Chap-
more than one, and usually quite a few more than one,
ter 20 for a more detailed discussion of the use of Skype
respondent or interviewee. Nor is this the only context
in interviews.
in which more than one person is interviewed. Bell et al.
There are several advantages of telephone over per-
(2001) carried out interviews with two managers in the
sonal interviews:
same company, both of whom had been involved in
the implementation of the people-management initia-
• On a like-for-like basis, they are far cheaper and
tive Investors in People. The managers, who had often quicker to administer. This arises because, for personal
had different roles in relation to the initiative or been interviews, interviewers may have to spend a great
involved with it at different stages of its development, deal of time and money travelling between respond-
were together able to build a chronological under- ents. This factor will be even more pronounced when
standing of its implementation. Similarly, in Bryman’s a sample is geographically dispersed, a problem that is
research on visitors to Disney theme parks, not just cou- only partially mitigated in personal interview surveys
ples but often their children took part in the interviews by strategies such as cluster sampling. Of course,
as well (Bryman 1999). However, it is very unusual for telephone interviews take time, and hired interview-
structured interviews to be used in connection with ers have to be paid, but the cost of conducting a
this kind of questioning. In survey research, it is almost telephone interview will still be lower than that of con-
always a single individual who is the object of question- ducting a comparable personal one.
ing. Indeed, in survey interviews it is advisable to dis-
courage as far as possible the presence and intrusion of • The telephone interview is easier to supervise than the
personal interview. This is a particular advantage
others during the course of the interview. Investigations
when there are several interviewers, since it becomes
in which more than one person is being interviewed tend
easier to check on interviewers’ transgressions in the
to be exercises in qualitative research, though this is not
asking of questions, such as rephrasing questions or
always the case.
the inappropriate use of probes by the interviewer (see
the section on ‘Probing’ later in the chapter). Probes
More than one interviewer are stimuli introduced by the interviewer to elicit fur-
ther information from the interviewee when the lat-
This is a relatively unusual situation in business research,
ter’s response is inadequate, either because it fails to
because of the considerable cost that is involved in dis-
answer the question or because it answers the question
patching two (or indeed more than two) people to inter-
but there is insufficient detail.
view someone. Bechhofer et al. (1984) describe research
in which two people interviewed individuals, with one • Telephone interviewing has a further advantage: some
interviewer doing most of the talking and the other tak- evidence (which is not as clear-cut as one might want)
ing more extensive notes. However, while their approach suggests that, in personal interviews, respondents’
achieved a number of benefits for them, their inter- replies are sometimes affected by characteristics of the
viewing style was of the unstructured kind that is typi- interviewer (for example, class or ethnicity) and
cally employed in qualitative research, and they argue indeed by his or her mere presence (implying that the
that the presence of a second interviewer is unlikely to interviewees may reply in ways they feel will be
achieve any added value in the context of structured deemed desirable by interviewers). The remoteness of
interviewing. the interviewer in telephone interviewing removes
 Interview contexts 213

10.3 RESEARCH IN FOCUS

A telephone survey of dignity at work
Berg and Frost (2005) used results from a telephone survey of over 500 low-skill, low-wage workers in hospitals in
the USA in order to explore the factors that affect workers’ perceptions of dignity at work. They also wanted to find
out if union representation and/or changes in worker representation influenced workers’ perceptions of dignity at
work. The researchers identified three main dimensions to the concept of dignity at work: fair treatment,
intrinsically satisfying work, and economic security. The kinds of jobs that their sample were engaged in included
food service workers, housekeepers, and nursing assistants doing tasks that the authors describe as ‘dead end
jobs with little or no chance of upward mobility’ (2005: 663). They observe that workers who do these jobs typically
earn only the minimum wage and there tends to be a high level of annual turnover, between 50 and 100 per cent.

The data came from a sample of 15 community hospitals across the USA, focusing on the most representative
group, hospitals with between 200 and 400 beds. The researchers chose telephone interviewing because of the
‘inherent instability in the lives of this sector of the workforce’ (2005: 669). This method was seen as being one
way of reliably reaching the workers. The researchers carried out telephone interviews with 589 workers, asking
them about all aspects of their jobs and careers. Telephone numbers were obtained from employers. However,
there were a number of difficulties associated with the telephone interview method that stemmed from the nature
of the population in which the research was interested. ‘Many of the phone numbers we secured from employers
were simply no good: the phone service had been disconnected; the person no longer lived at that phone number;
or the respondent would not answer the telephone’ (2005: 669). One of the reasons a phone call was not
answered was because respondents had call display. If they did not recognize the number, they would not pick up
the phone because they were trying to avoid debt collection agents. These difficulties adversely affected the
response rate, which ended up at 45 per cent. The researchers conclude that the people they were able to survey
probably represent the most stable part of this population, so the results are likely to overstate the positive aspects
associated with these jobs: ‘those whose work lives keep them living in a precarious fashion are likely those not
responding to our telephone survey’ (2005: 669).

this potential source of bias to a significant extent. The personal interviews (Frey and Oishi 1995; Shuy 2002;
interviewer’s personal characteristics cannot be seen, Frey 2004).
and the fact that he or she is not physically present may
• There is some evidence to suggest that telephone
offset the likelihood of respondents’ answers being interviews are less effective for the asking of questions
affected by the interviewer. In the case of web-based about sensitive issues, such as workplace bullying or
applications such as Skype, interviewers may choose drug and alcohol use. However, the evidence is not
not to use the video capability precisely because it entirely consistent on this point, though it is probably
introduces the possibility of this kind of effect. sufficient to suggest that, when many questions of this
Telephone interviewing suffers from certain limita- kind are to be used, a personal interview may be supe-
tions when compared to the personal interview. rior (Shuy 2002).

• Developments in telephone communications, such as

• The length of a telephone interview is unlikely to be
answerphones, other forms of call screening, and
sustainable beyond 20–25 minutes, whereas personal
mobile phones, have almost certainly had an adverse
interviews can be much longer than this (Frey 2004).
effect on telephone surveys in terms of response rates
• The question of whether response rates (see Key con- and the general difficulty of getting access to respond-
cept 9.3) are lower with surveys by telephone inter- ents through conventional landlines. Households that
view than with surveys by personal interview is rely exclusively on mobile phones represent a particu-
unclear, in that there is little consistent evidence on lar difficulty. On the other hand, the increased spread
this question. However, there is a general belief that of internet access and the availability of web-based
telephone interviews achieve slightly lower rates than communication may limit this problem.
214 10 Structured interviewing

formats for computer-assisted interviewing: computer-

• Telephone interviewers have less opportunity to
engage in observation. This means that they are not assisted personal interviewing (CAPI) and computer-
able to respond to signs of puzzlement or unease on assisted telephone interviewing (CATI). A very large
the faces of respondents when they are asked a ques- percentage of telephone interviews are conducted with
tion. In a personal interview, the interviewer may the aid of computers. Among commercial survey organi-
respond to such signs by restating the question or zations, almost all telephone interviewing is of the CATI
attempting to clarify the meaning of the question, kind, and this kind of interview has become one of the
though this must be handled in a standardized way as most popular formats for such firms. The main reasons
far as possible. A further issue relating to the inability for the growing use of CAPI have been the increased por-
of the interviewer to observe is that, sometimes, inter- tability and affordability of laptop computers and tablets,
viewers may be asked to collect subsidiary information and the growth in the number and quality of software
about their visits (for example, whether health and packages that provide a platform for devising interview
safety procedures are made evident at a business schedules. If the interviewer is out in an organization all
premises). Such information cannot be directly col- day, he or she may take a removable storage device with
lected when telephone interviews are employed. the saved data to the research office or upload the data.
With computer-assisted interviewing, the questions
• It is frequently the case that specific individuals in
that comprise an interview schedule appear on the screen.
households or firms are the targets of an interview. In
As interviewers ask each question and the respondent
other words, simply anybody will not do. This require-
replies, the interviewer inputs the appropriate reply and
ment is likely to arise from the specifications of the
proceed to the next question. This process has the great
population to be sampled, which means that people in
advantage that, when filter questions are asked, so that
a certain role or position or with particular character-
certain answers may be skipped as a result of a person’s
istics are to be interviewed. It is probably more difficult
reply, the computer ‘jumps’ to the next relevant question.
to ascertain by telephone interview whether or not the
This removes the possibility of interviewers inadvertently
correct person is replying.
asking inappropriate questions or failing to ask ones that
• There is some evidence to suggest that the quality of should be asked. In these ways, computer-assisted inter-
data derived from telephone interviews is inferior to viewing enhances the degree of control over the inter-
that of comparable face-to-face interviews. A series of view process and can therefore improve standardization
experiments reported by Holbrook et al. (2003) on the of the asking and recording of questions. However, there
mode of survey administration in the USA using long is very little evidence to suggest that the quality of data
questionnaires found that respondents interviewed by deriving from computer-assisted interviews is demon-
telephone were more likely to express no opinion or strably superior to that provided by comparable paper-
‘don’t know’ (see Chapter 12 for more on this issue); to and-pencil interviews (Couper and Hansen 2002).
answer in the same way to a series of linked questions; It is possible that technophobic respondents may be
to provide socially desirable answers; to be apprehen- a bit alarmed by the use of a laptop or tablet during an
sive about the interview; and to be more likely to be dis- interview. One of us has had personal experience of
satisfied with the time taken by the interview (even this technique as a respondent in a market research sur-
though interviews were typically shorter than in the vey: in this instance the laptop started to beep part of
face-to-face mode). Also, telephone interviewees the way through the interview because the battery was
tended to be less engaged with the interview process. about to expire and needed to be replaced with a back-
While these results should be viewed with caution, since up. An incident such as this could be disruptive to the
studies like these are bound to be affected by such fac- flow of an interview and be alarming for technophobic
tors as the use of a large questionnaire on a national respondents.
sample, they do provide interesting food for thought. CAPI and CATI have not influenced academic survey
research to anything like the same degree that they have
in commercial survey research, although that picture is
Computer-assisted interviewing likely to change considerably because of the many advan-
In recent years, increasing use has been made of comput- tages they possess. In any case, many of the large datas-
ers in the interviewing process, especially in commercial ets that are used for secondary analysis (see Chapter 14
survey research of the kind conducted by market research for examples) derive from computer-assisted interview-
and opinion polling organizations. There are two main ing studies undertaken by commercial or large social
 Conducting interviews 215
research organizations. One further point to register in reason for this is that such surveys are self-completion
connection with computer-assisted interviewing is that questionnaires rather than interviews. We cover them in
so far we have avoided discussion of online surveys. The Chapter 11.

Conducting interviews
Issues concerning the conduct of interviews are exam- spoken form in such situations as when interviewers make
ined here in a very general way. In addition to the mat- contact with respondents on the street or when they ‘cold
ters considered here, there is clearly the important issue call’ respondents in their homes or at their place of work,
of how to word the interview questions themselves. This whether in person or by telephone. A written rationale
area will be explored in Chapter 12; many of the rules will be required to alert respondents that someone will be
of question-asking relate to self-completion question- contacting them in person or on the telephone to request
naire techniques such as postal or online questionnaires an interview. Respondents will frequently encounter both
as well as to structured interviews. One further general forms—for example, when they are sent a letter and then
point to make here is that the advice concerning the ask the interviewer who turns up to interview them what
conduct of interviews provided in this chapter relates to the research is all about. It is important for the rationale
structured interviews. The framework for conducting the given by telephone to be consistent with the one given by
kinds of interviewing used in qualitative research (such letter, as if respondents pick up inconsistencies they may
as unstructured and semi-structured interviewing and well be less likely to participate in the survey.
focus groups) will be handled in Part Three of the book. Introductions to research should typically contain
the bits of information outlined in Tips and skills ‘Top-
ics and issues to include in an introductory statement’.
Know the schedule Since interviewers represent the interface between the
Before interviewing anybody, an interviewer should be research and the respondent, they have an important
fully conversant with the schedule. Even if you are the role in maximizing the response rate for the survey. In
only person conducting interviews, make sure you know addition, the following points should be borne in mind:
it inside out. Interviewing can be stressful for interview-
ers; in such cases, standard interview procedures such • Interviewers seeking respondents by phone should be
prepared to keep calling back if interviewees are out or
as filter questions (described under ‘Clear instructions’
unavailable. This will require taking into account peo-
below) can cause interviewers to get flustered and miss
ple’s likely work and leisure habits—for example, there
questions out or ask the wrong questions. If two or more
is no point in calling at home on people who work during
interviewers are involved, they need to be fully trained
the day. In addition, first thing in the morning may not
to know what is required of them and to know their way
be the best time to contact a busy manager who is likely
around the schedule. Training is especially important
to be briefing colleagues and responding to queries.
to reduce the likelihood of interviewer variability in the
asking of questions, which is a source of error. • Be self-assured; you may get a better response if you
presume that people will agree to be interviewed
rather than that they will refuse.
Introducing the research
• Reassure people that you are not a salesperson. Because
The researcher should give prospective respondents a of the tactics of certain organizations whose represent-
credible rationale for the research in which they are being atives say they are doing market or business research,
asked to participate and for giving up their valuable time. many people have become very suspicious of people
This aspect of conducting interview research is of particu- saying they would just like to ask you a few questions.
lar significance at a time when response rates to social sur-
• If you will be seen by respondents, dress in a way that
vey research appear to be declining (though, as noted in will be acceptable to a wide spectrum of people.
Chapter 9, the evidence on this issue is the focus of some
disagreement). The introductory rationale may be either • Make it clear that you will be happy to find a time to
suit the respondent.
spoken by the interviewer or written down. It comes in
216 10 Structured interviewing

TIPS AND SKILLS

Topics and issues to include in an introductory
­statement
There are several issues to include in an introductory statement to a prospective interviewee. The following list
comprises the principal considerations.

• Make clear the identity of the person who is contacting the respondent.
• Identify the auspices under which the research is being conducted—for example, a university, a market
research agency.
• Mention any research funder, or, if you are a student doing an undergraduate or postgraduate dissertation,
make this clear.
• Indicate what the research is about in broad terms and why it is important, and give an indication of the kind of
information to be collected.
• Indicate why the respondent has been selected—for example, selected by a random process.
• Provide reassurance about the confidentiality of any information provided.
• Make it clear that participation is voluntary.
• Reassure the respondent that he or she will not be identified or be identifiable in any way. This can usually be
achieved by pointing out that data are anonymized when they are entered into the computer and that analysis
will be conducted at an aggregate level.
• Provide the respondent with the opportunity to ask any questions—for example, provide a contact telephone
number if the introduction is in the form of a written statement, or, if in person, simply ask if the respondent has
any questions.

These suggestions are also relevant to the covering letter that accompanies questionnaires (except that
researchers wishing to receive responses on paper through the post need to remember to include a stamped
addressed envelope!).

Rapport questions in a way that is designed to please the inter-

viewer. The achievement of rapport between interviewer
It is frequently suggested that it is important for the and respondent is therefore a delicate balancing act.
interviewer to achieve rapport with the respondent. This Moreover, it is probably somewhat easier to achieve in
means that very quickly a relationship must be estab- the context of the face-to-face interview rather than the
lished that encourages the respondent to want (or at least telephone interview, since in the latter the interviewer
be prepared) to participate in and persist with the inter- is unable to offer obvious visual cues of friendliness like
view. Unless an element of rapport can be established, smiling or maintaining good eye contact, which are fre-
some respondents may initially agree to be interviewed quently regarded as conducive to gaining and maintain-
but then decide to terminate their participation because ing rapport.
of the length of time the interview is taking or perhaps
because of the nature of the questions being asked.
While this injunction essentially invites the interviewer
Asking questions
to be friendly with respondents and to put them at ease, It was suggested earlier in the chapter that one of the
it is important that this quality is not stretched too far. aims of the structured interview is to ensure that each
Too much rapport may result in the interview going on respondent is asked exactly the same questions. It was
too long and the respondent suddenly deciding that too also pointed out that variation in the ways a ques-
much time is being spent on the activity. Also, the mood tion is asked is a potential source of error in survey
of friendliness may result in the respondent answering research. The structured interview is meant to reduce
 Conducting interviews 217
the likelihood of this occurring, but it cannot guaran- replies to a category, as in a closed question. This process
tee that this will not occur, because there is always the can require a certain amount of interpretation on the
possibility that interviewers will embellish or otherwise part of the interviewer, but the error that is introduced
change a question when it is asked. There is consider- is far less than when answers to open questions are being
able evidence that this occurs, even among centres of written down (Fowler and Mangione 1990).
social research that have a solid reputation for being
rigorous in following established methodological pro-
Clear instructions
tocol (Bradburn and Sudman 1979). The problem with
such variation in the asking of questions was outlined In addition to instructions about the asking of questions
above: it is likely to engender variation in replies that and the recording of answers, interviewers need instruc-
does not reflect ‘true’ variation—in other words, error. tions about their progress through an interview sched-
Consequently, it is important for interviewers to appreci- ule. An example of the kind of context in which this is
ate the importance of keeping exactly to the wording of likely to occur is in relation to filter questions. Filter ques-
the questions they are charged with asking. tions require the interviewer to ask questions of some
You might say: ‘Does it really matter?’ In other words, respondents but not others. For example, the question
surely small variations to wording cannot make a signifi-
How many days of on-the-job training have you
cant difference to people’s replies? While the impact of
received in the past 12 months?
variation in wording obviously differs from context to
context and is in any case difficult to quantify exactly, presumes that the respondent is in employment. This
experiments in question wording suggest that even small option can be reflected in the fixed-choice answers that
variations in wording can exert an impact on replies are provided, so that one of these is a ‘not-in-employ-
(Schuman and Presser 1981). Three experiments in ment’ alternative. However, a better solution is not to
England conducted by Social and Community Planning presume anything about respondents’ work behaviour
Research concluded that a considerable number of inter- but to ask them if they are currently in employment and
view questions are affected by interviewer variability. then to filter out those respondents who are not. Tips and
The researchers estimated that, for about two-thirds of skills ‘Instructions for interviewers in the use of a filter
the questions that were considered, interviewers con- question’ provides a simple example in connection with
tributed to less than 2 per cent of the total variation in an imaginary study of feedback and job performance.
each question (M. Collins 1997). On the face of it, this is The chief point to register about this example is that it
a small amount of error, but the researchers regarded it requires clear instructions for the interviewer. If such
as a cause for concern. instructions are not provided, there is the risk either
The key point to emerge, then, is the importance of that respondents will be asked inappropriate questions
getting across to interviewers the importance of asking (which can be irritating for them) or that the interviewer
questions as they are written. There are many reasons will inadvertently fail to ask a question (which results in
why interviewers may vary question wording, such as missing information).
reluctance to ask certain questions, perhaps because
of embarrassment (M. Collins 1997), but the general
Question order
admonition to keep to the wording of the question needs
to be constantly reinforced when interviewers are being In addition to warning interviewers about the importance
trained. It also needs to be borne in mind for your own of not varying the asking of questions and the recording
research. of answers, they should be alerted to the importance of
keeping to the order of asking questions. For one thing,
varying the question order can result in certain questions
Recording answers being accidentally omitted, because the interviewer may
An identical warning for identical reasons can be reg- forget to ask those that have been leapfrogged during the
istered in connection with the recording of answers interview. Also, variation in question order may have an
by interviewers, who should write down respondents’ impact on replies: if some respondents have been previ-
replies as exactly as possible. Not to do so can result in ously asked a question that they should have been asked
interviewers distorting respondents’ answers and hence whereas others have not, a source of variability in the
introducing error. Such errors are less likely to occur asking of questions will have been introduced and there-
when the interviewer has merely to allocate respondents’ fore a potential source of error.
218 10 Structured interviewing

TIPS AND SKILLS

Instructions for interviewers in the use of a filter
question
1. Have you received any feedback concerning your job performance during the last twelve months?
Yes ____
No ____
(if No proceed to question 4)
2. (To be asked if respondent replied Yes to question 1)
Who provided you with this feedback?
(Ask respondent to choose the category that represents the person who most often gives them feedback and to
choose one category only.)
Line manager ____
HR manager ____
Other ______________________ (specify)
3. How frequently do you receive feedback concerning your job performance?
(Ask respondent to choose the category that comes closest to his or her current experience.)
Once or twice a week ____
Once or twice a month ____
A few times a year ____
Once or twice a year ____
4. (To be asked if respondent replied No to question 1)
Have you received feedback concerning your job performance at any time during your employment by this
organization?
Yes ____
No ____

Quite a lot of research has been carried out on the

• Early questions should be directly related to the topic of
general issue of question order, but this has shown few, the research, about which the respondent has been
if any, consistent effects on people’s responses as a result informed. This removes the possibility that the
of asking questions at different points in a questionnaire respondent will be wondering at an early stage in the
or interview schedule. Different effects have been dem- interview why he or she is being asked apparently irrel-
onstrated on various occasions. There are two general evant questions. This injunction means that personal
lessons. questions about age, social background, and so on
should not be asked at the beginning of an interview.
• Within a survey, question order should not be varied
(unless, of course, question order is the subject of the • As far as possible, questions that are more likely to be
study!). salient to respondents should be asked early in the
interview schedule, so that their interest and attention
• Researchers should be sensitive to the possible impli-
cations of the effect of early questions on answers to are more likely to be secured. This suggestion may con-
subsequent questions. flict with the previous one, in that questions specifi-
cally on the research topic may not be obviously salient
The following rules about question order are sometimes to respondents, but it implies that as far as possible
proposed: questions relating to the research topic that are more
 Conducting interviews 219
likely to grab their attention should be asked at, or
• It is sometimes recommended that questions dealing
close to, the start of the interview. with opinions and attitudes should precede questions
• Potentially embarrassing questions or ones that may to do with behaviour and knowledge. This is because it
be a source of anxiety should be left till later. In fact, is felt that behaviour and knowledge questions are less
research should be designed to ensure that as far as affected by question order than questions that tap
possible respondents are not discomfited, but it has to opinions and attitudes.
be acknowledged that with certain topics this effect
• During the course of an interview, it sometimes hap-
may be unavoidable. pens that a respondent provides an answer to a ques-
• With a long schedule or questionnaire, questions tion that is to be asked later in the interview. Because
should be grouped into sections, since this allows a of the possibility of a question order effect, when the
better flow than skipping from one topic to another. interviewer arrives at the question that appears
already to have been answered, it should be repeated.
• Within each group of questions, general questions
should precede specific ones. Research in focus 10.4 However, question order effects remain one of the more
provides an illustration of such a sequence. frustrating areas of structured interview and question-
• A further aspect of the rule that general questions naire design, because of the inconsistent evidence that
should precede specific ones is that it has been argued has beeen found and because it is difficult to formulate
that, when a specific question comes before a general generalizations or rules from the evidence that does point
one, the aspect of the general question that is covered to the existence of question order effects.
by the specific one is discounted in the minds of
respondents because they feel they have already
Probing
covered it. Thus, if a question about how people feel
about the amount they are paid precedes a general Probing is a highly problematic area for researchers
question about job satisfaction, there are grounds for employing a structured interview method. It frequently
thinking that respondents will discount the issue of happens in interviews that respondents need help with
pay when responding about job satisfaction. their answers. One obvious case is where it is evident that

10.4 RESEARCH IN FOCUS

A question sequence
Here is hypothetical set of questions for an interview on the privatization of public authorities, the process by which
government agencies are sold to commercial interests. It follows the recommendations of Foddy (1993: 61–2)
about question order. The question order sequence is designed with a number of features in mind. It is designed
to establish people’s levels of knowledge of privatization before asking questions about it and to distinguish those
who feel strongly about it from those who do not. The second question is open-ended, so respondents’ frames of
reference can be established with respect to the topic at hand. Subsequent questions are aimed at gathering data
on people’s views, and the strengths of those views.

1. Have you heard about the possible sale of public authorities to commercial interests?
Yes ____ No ____
2. What do you think about the sale of public authorities?
3. Do you favour or not favour the sale of public authorities?
Favour ____ Not favour ____
4. Why do you favour (not favour) the sale of public authorities?
5. How strongly do you feel about this?
Very strongly ____
Fairly strongly ____
Not at all strongly ____
220 10 Structured interviewing

they do not understand the question—they may ask for question and to suggest possible answers only to some
further information, or it may be clear from what they respondents, such as those who appear to be struggling
say that they are struggling to understand the question to think of an appropriate reply.
or to provide an adequate answer. The second kind of During the course of a face-to-face interview, there are
situation the interviewer faces is when the respondent several circumstances in which it will be better for the
does not provide a sufficiently complete answer and must interviewer to use ‘show cards’ rather than rely on read-
be probed for more information. The problem in either ing out a series of fixed-choice alternatives. Show cards
situation is obvious: the interviewer’s intervention may (sometimes called ‘flash cards’) display all the answers
influence the respondent and the nature of interviewers’ from which the respondent is to choose and are handed
interventions may differ. A potential source of variability to the respondent at different points of the interview.
in respondents’ replies that does not reflect ‘true’ varia- Three kinds of context in which it might be preferable to
tion is introduced—that is, error. employ show cards, rather than to read out the entire set
Some general tactics regarding probes are as follows: of possible answers are as follows:

• If further information is required, usually in the con- • There may be a very long list of possible answers. For
text of an open-ended question, standardized probes example, respondents may be asked which daily news-
can be employed, such as ‘Could you say a little more paper they each read most frequently. To read out a list
about that?’ or ‘Are there any other reasons why you of newspapers would be tedious and it is probably bet-
think that?’ or simply ‘Mmmm … ?’ Probes have to be ter to hand the respondent a list of newspapers from
handled carefully. If they are not introduced in a con- which to choose.
sistent way or if they suggest a particular kind of
• Sometimes, during the course of interviews, respond-
answer to the interviewee, error will increase. ents are presented with a group of questions to which
• If the problem is that when presented with a closed the same possible answers are attached. An example of
question the respondent replies in a way that does not this approach is Likert scaling, which is commonly
allow the interviewee to select one of the pre-designed used to measure attitudes. A typical strategy entails
answers, the interviewer should repeat the fixed-choice providing respondents with a series of statements and
alternatives and make it apparent that the answer needs asking them how far they agree or disagree with the
to be chosen from the ones that have been provided. statements (see Chapter 8). These are often referred to
as items rather than as questions, since, strictly speak-
• If the interviewer needs to know about something that
requires quantification, such as the number of visits to ing, the respondent is not being asked a question. An
banks in the last four weeks or the number of banks example was provided in Research in focus 8.3. It
with which the respondent has accounts, but the would be excruciatingly dull to read out all five possi-
respondent resists this by answering in general terms ble answers 12 times. Also, it may be expecting too
(‘quite often’ or ‘I usually go to the bank every week’), much of respondents if the interviewer reads out the
the interviewer needs to persist with securing a num- answers once and then requires respondents to keep
ber from the respondent. This will usually entail the possible answers in their heads for the entire batch
repeating the question. The interviewer should not try of questions to which they apply. A show card that can
to second guess a figure on the basis of the respond- be used for the entire batch and to which respondents
ent’s reply and then suggest that figure to him or her, can constantly refer is an obvious solution. As was
since the latter may be unwilling to demur from the mentioned in Research in focus 8.3, most Likert scales
interviewer’s suggested figure. of this kind comprise five levels of agreement/disa-
greement and it is this conventional approach that is
illustrated in Tips and skills ‘A show card’.
Prompting
• Some people are not keen to divulge personal details
Prompting occurs when the interviewer suggests a pos- such as their age or their income. One way of neutraliz-
sible answer to a question to the respondent. The key ing the impact of such questioning is to present respond-
prerequisite here is that all respondents receive the ents with age or income bands with a letter or number
same prompts. All closed questions entail standardized attached to each band. They can then be asked to say
prompting, because the respondent is provided with a which letter applies to them (see Tips and skills ‘Another
list of possible answers from which to choose. An unac- show card’). This procedure will obviously not be appro-
ceptable approach to prompting would be to ask an open priate if the research requires exact ages or incomes.
 Conducting interviews 221

TIPS AND SKILLS

A show card
Card 6
Strongly agree
Agree
Undecided
Disagree
Strongly disagree

TIPS AND SKILLS

Another show card
Card 11

(a) Below 20
(b) 20–29
(c) 30–39
(d) 40–49
(e) 50–59
(f) 60–69
(g) 70 and over

Leaving the interview researcher hires an interviewer to conduct many or even

all the interviews. It also has considerable importance
Do not forget common courtesies such as thanking in research in which several interviewers (who may be
respondents for giving up their time. However, the period either collaborators or hired interviewers) are involved
immediately after the interview is one in which some in a study, since the risk of interviewer variability in the
care is necessary, in that sometimes respondents try to asking of questions needs to be avoided.
engage the interviewer in a discussion about the purpose For many readers of this book who are planning to
of the interview. Interviewers should resist elaboration do research, such situations are unlikely to be relevant,
beyond their standard statement, because respondents because they will be ‘lone’ researchers. You may be
may communicate what they are told to others, which doing an undergraduate dissertation, or an exercise for
may bias the findings. a research methods course, or you may be a postgradu-
ate conducting research for a Master’s dissertation or
for a PhD thesis. Most people in such a situation will not
Training and supervision
have the luxury of being able to hire a researcher to do
On several occasions, reference has been made to the any interviewing (though you may be able to find some-
need for interviewers to be trained. The standard texts one to help you a little). When interviewing on your
on survey research and on interviewing practice tend to own, you must train yourself to follow the procedures
be replete with advice on how best to train interviewers. and advice provided above. This is a very different situ-
Such advice is typically directed at contexts in which a ation from a large research institute or market research
222 10 Structured interviewing

agency, which relies on an army of hired interviewers Fowler (1993) cites evidence that suggests that training
who carry out the interviews. Whenever people other of less than one full day rarely creates good interviewers.
than the lead researcher are involved in interviewing, Supervision of interviewers in relation to these issues
they will need training and supervision in the following can be achieved in the following ways:
areas:
• checking individual interviewers’ response rates;
• contacting prospective respondents and providing an
• making audio recordings of at least a sample of inter-
introduction to the study; views;
• reading out questions as written and following
• examining completed schedules to determine whether
instructions in the interview schedule (for example, in any questions are being left out or if they are not being
connection with filter questions); completed properly;
• using appropriate styles of probing;
• call-backs on a sample of respondents (usually around
• recording exactly what is said; 10 per cent) to determine whether or not they were
interviewed and to ask about the interviewers’
• maintaining an interview style that does not bias
respondents’ answers. ­conduct.

Other approaches to structured interviewing

A number of other methods or techniques are used in activity where the consequences are sufficiently clear as
business research as part of either the structured or the to leave the observer with a definite idea as to their likely
semi-structured interview. Three main types will be dis- effects. The term is derived from the analysis of near-
cussed in this section: disaster situations, where a version of the technique can
be used to build up a picture of the events that contribute
• the critical incident method;
to a potential disaster and to develop a plan of action for
• projective methods, pictorial methods, and photo- dealing with them. The most common use of the criti-
elicitation; cal incident method involves interviewing respondents
• the verbal protocol approach. about particular types of event or behaviour in order to
develop an understanding of their sequence and their
We have grouped these three methods together here significance to the individual.
because they can form part of a structured interview. One of the earliest and most well-known illustrations
However, they can also form part of a semi-structured of this method in business research is the study by
interview (see Chapter 20) in a qualitative investiga- Herzberg et al. (1959), which was mentioned in
tion, and so to an extent they cut across the quantitative/ Chapter 9. The authors explain: ‘We decided to ask peo-
qualitative divide (see Chapter 26). They are sometimes ple to tell us stories about times when they felt excep-
used as one part of an interview, in combination with tionally good or bad about their jobs. We decided that
other questions that form part of a more conventional from these stories we could discover the kinds of situ-
interview schedule; in other research designs they form ations leading to negative or positive attitudes toward
the basis for the entire interview. A further use of these the job and the effects of these attitudes’ (1959: 17).
methods is to check findings from more conventional Their initial interview strategy was followed up by a
quantitative approaches such as structured interviews or series of probe questions that filled in missing informa-
questionnaire surveys. tion in the spontaneously told accounts. Content analysis
(see Chapter 13) was then used to focus on exploring
the essential features of the critical incident in order
The critical incident method
to reveal the values that they reflected. A more recent
The critical incident method involves asking respondents example of the use of the critical incident method is
to describe critical incidents, which are defined very found in the study of managerial effectiveness described
broadly by Flanagan (1954) as any observable human in Research in focus 10.5.
 Other approaches to structured interviewing 223

10.5 RESEARCH IN FOCUS

An example of the critical incident method
Chai et al. (2016) used the critical incident technique in interviews with Korean managerial and non-managerial
employees. They interviewed 45 respondents across a number of large Korean organizations. Their aim was to
develop an understanding of what was regarded as effective, and ineffective, management and leadership in the
Korean context. Respondents were asked to identify four or more critical incidents which involved effective
management and leadership and four or more which involved ineffective management and leadership. The time
period was limited to six months before the interview. Respondents were asked three questions about each
incident:

1. What was the background situation, circumstance, or context that led up to the critical incident you have in
mind?
2. What exactly did the manager you observed do/say or not do/say, and/or in what way did s/he do/not do it or
say/not say it that was either effective or ineffective?
3. What was the specific outcome of the critical incident that you have described, and on reflection why do you
perceive this to be an example of effective or ineffective managerial behavior?
(Chai et al. 2016: 796)

The kinds of incidents which were recalled as indicating effective management included managers publicly
praising staff and seeking to resolve problems for staff, while ineffective management was illustrated by examples
including trying to force employees to do as they were told and setting overly ambitious targets for teams.

Finally, although we have introduced the critical inci- how they had coped with it. Blackburn and Stokes’s anal-
dent method here in Part Two of the book, which deals ysis of the data was primarily qualitative, relying on the
with quantitative research, we should point out that this use of themes illustrated by the inclusion of direct quotes
method is often used as part of a qualitative research from respondents.
investigation. An example of this is the study of small
business owner-managers by Blackburn and Stokes
Projective methods, pictorial methods,
(2000) (see Research in focus 21.3). In this instance,
respondents were asked to recall a situation that had and photo-elicitation
arisen in the previous two years in which they had lost a Projective methods classically involve the presenta-
major customer and to explain what had happened and tion of ambiguous stimuli to individuals, and responses

STUDENT EXPERIENCE
Using the critical incident method in a student
research project
Tom followed the advice of his supervisor and used the critical incident technique when he was interviewing call
centre workers about wellbeing and the experience of working in a call centre. He explained:

I did use critical incident technique in the interview to try and get people to give an example of a time when
things have gone particularly well at work or particularly badly. It was quite a useful tool to get people to talk in
an interesting way really about their work experience. And sometimes it wasn’t really necessary because they
would give lots of examples anyway, but it was quite a useful prompt to get people to be more specific and give
examples of the sort of stuff that their work life was like.
224 10 Structured interviewing

are interpreted by the researcher to reveal underlying 40). The stories were compared with those that had been
characteristics of the individual concerned. A common written by a control group under normal conditions. The
example is the Rorschach inkblot test, where respon- experimental group was found to refer more often in
dents are asked to describe random inkblots. Analysis their stories to ideas related to achievement. From this,
relies on expert psychological interpretation of the way McClelland concluded that, if someone ‘in writing his
that respondents have described the inkblots, and this stories consistently uses achievement-related ideas of the
is suggested to be indicative of their dominant channels same kind as those elicited in everyone under achieve-
of thinking. Another form of projective analysis involves ment “pressure,” then he would appear to be someone
the ‘sentence completion test’, where the individual is with a “bias,” a “concern,” or a “need” for achievement’
asked to complete a number of unfinished sentences; (1961: 43). This led him to develop a score for the need
this technique has been used in the context of recruit- for achievement, defined as the number of achievement-
ment and selection, often as an assessment centre related ideas in stories written by an individual under
exercise. normal conditions.
One of the best-known examples of the use of projective A more recent example of projective methods can also
techniques in management research involves the study by be found in marketing research (see Research in focus
McClelland (1961) of leadership and the need for individ- 10.6). The authors of this study sought to investigate the
ual achievement. Informed by experimental psychology influence of companies’ corporate social responsibility
and the psychoanalytic insights of Freud, McClelland’s on consumers’ views of those companies and purchas-
study first involved stimulating the achievement motive ing intentions. However, the use of projective methods
in a group of subjects. He then sought to elicit their ‘spon- is relatively uncommon in business research. They have
taneous thoughts’ and fantasies in order to determine the largely been superseded by the use of visual techniques
effect of achievement motivation. The subjects were male to stimulate creative thinking and problem solving, and
college students who were told that they were going to to explore feelings, emotions, and values. For example,
be tested to determine their intelligence and leadership Stiles (2004) asked members of UK and North American
ability; it was assumed that this would arouse a desire business schools to express how they saw their organi-
in the subjects to do well. After the ‘tests’ had been com- zation’s identity by drawing pictures (see Research in
pleted, subjects were asked to write short five-minute focus 10.7). The use of photo-elicitation can also be seen
stories suggested by pictures that flashed onto a screen as an adaptation of projective methods (see Key concept
for a few seconds. ‘The pictures represented a variety of 10.8 for an explanation and Research in focus 10.9 for
life situations centering particularly around work’ (1961: an example).

10.6 RESEARCH IN FOCUS

Using projective methods in consumer research
McEachern (2015) was interested in the extent to which corporate citizenship behaviour by private companies (for
example, voluntary philanthropic activity) influenced consumer views of those companies. More specifically, he
was interested in levels of awareness by consumers and the nature and extent of their reactions to corporate
citizenship behaviour. His study focused on the chocolate manufacturing industry and the production of ‘Fairtrade’
chocolate (using cocoa and other products which have been grown, processed, and transported ethically) and
involved data collection from UK consumers. The study collected data using a number of techniques; the
projective techniques were used as part of focus group interviews. Members of the groups were presented with
magazines and chocolate wrappers and asked to construct collages in response to one of two questions: ‘What
does Fairtrade mean to you?’ and ‘Illustrate how you feel about leading, conventional chocolate brands becoming
Fairtrade’. The intention was to provide participants with the opportunity to express their feelings on these issues
without being influenced by the researcher. After the collages were constructed, interviews were conducted about
them. McEachern’s findings indicated that while consumers were aware of Fairtrade initiatives in the chocolate
industry and while they approved of them, such initiatives had limited impact on decisions about which products
to purchase.
 Other approaches to structured interviewing 225

10.7 RESEARCH IN FOCUS

Using pictorial exercises in a study of business
school identity
Stiles (2004) used pictorial methods in a study of strategy in UK and North American business schools. The first
stage of the research involved asking individual interviewees to imagine their organization as having a personality
and then asking them to draw a picture of what that personality looks like. The second stage of the research
involved showing these drawings to members of a focus group (see Chapter 21), who were invited to reach a
consensus in choosing five pictures, ranging from an unfavourable depiction of the organization, to neutral,
through to a favourable one. ‘The group then produces a composite free-drawn personality image of its own’
(2004: 130). The focus group discussion was video-recorded and transcribed. The importance of the pictures
stems from the discussion that respondents had around their selection decisions, which revealed insights into the
way academics perceived the management styles associated with their organizations and leaders. Stiles notes
that, although this study was conducted in a business school setting, it could equally be applied in relation to a
variety of other organizational settings. Stiles concludes that the pictorial exercises revealed constructs that were
not identified using verbal research instruments, thus introducing the possibility that images are useful in revealing
more latent perceptions.

10.8 KEY CONCEPT

What is photo-elicitation?
This method involves integrating photographs into the interview by asking the respondent questions about
photographs that the researcher or the respondent has taken of the research setting. Respondents are asked to
reflect, explain, and comment on the meaning of the objects in the photograph, the events that are taking place, or
the emotions they associate with them. Photographs can provide a point of reference for the discussion and can
help to move the interview from ‘the concrete (a cataloguing of the objects in the photograph) to the socially
abstract (what the objects in the photograph mean to the individual being interviewed)’ (Harper 1986: 25). Harper
suggests that the most useful photographs tend to be those that are visually arresting, because they are more likely
to get the respondent’s attention and provoke a response.

10.9 RESEARCH IN FOCUS

Using photo-elicitation to study tourist behaviour
Tonge et al. (2013) used photo-elicitation to study the reasons that tourists become attached to particular
locations and visit them repeatedly, even when they are often distant and difficult to reach. Their study focused
on visitors to Ningaloo Reef, in remote north-western Australia, and involved providing respondents (all of whom
were repeat visitors to the reef) with digital cameras and asking them to take photographs which they felt
summed up what the location meant to them and why they returned to it. The researchers uploaded the
photographs to laptops and then conducted interviews with the participants while they looked at the photos they
had taken. This method was chosen not just because it is an effective way of eliciting useful responses during
interviews, but also because it involved respondents producing data while engaging in a familiar holiday
activity—taking photographs.
226 10 Structured interviewing

The verbal protocol approach subject’s account of what he or she is doing and why is
usually audio-recorded and transcribed and then content
The verbal protocol approach builds on the work of analysed (a process discussed in Chapter 13 of this book)
Newell and Simon (1972) in the area of human problem using a coding scheme to discern different categories
solving and has since been used in relation to a number of thinking. An interesting example of the use of verbal
of topics that are relevant to business researchers. The protocol analysis can be found in a study by Ramiah and
approach involves asking respondents to ‘think aloud’ Banks (2015), who wanted to explore how decisions
while they are performing a task. The idea is to elicit the were made in the resolution of labour disputes (see
respondent’s thought processes while he or she is mak- Research in focus 10.10).
ing a decision or judgement or solving a problem. The

10.10 RESEARCH IN FOCUS

A study using the verbal protocol method
In their study of decision-making in labour disputes, Ramiah and Banks (2015) wished to understand processes
of decision-making by actors involved in dispute resolution. In particular, they sought to make sense of the
analytic processes involved in making decisions in real time while working to resolve disputes. For the study, 50
civil servants who worked in the Department of Labour in Malaysia were presented with realistic scenarios
presenting labour disputes (all were based on actual disputes) which were characterized by complexity and
uncertainty. These civil servants were responsible in their jobs for resolving disputes in ways which were intended
to balance the rights and responsibilities of workers and employers. For the purposes of the study, they were
divided into expert and novice groups based on their length of service and the extent of experience which they
had in dealing with disputes. Participants had to analyse the scenarios and decide how to deal with them, while
thinking aloud. The study found that experienced participants tended to construct arguments as they worked
through the decision, rather than framing their decision as a narrative. The more experienced participants also
tended to construct clear links between the evidence presented in the case and specific decisions at each stage
of the process, which appeared to lead to better decisions. The use of the verbal protocol method allowed
Ramiah and Banks to develop an understanding of the processes of decision-making which was more
naturalistic than if they had simply asked respondents a series of questions about how they performed
their roles at work.

Problems with structured interviewing

While the structured interview is a commonly used Characteristics of interviewers
method of business research, certain problems associ-
There is evidence that interviewers’ attributes can have
ated with it have been identified over the years. These
an impact on respondents’ replies, but, unfortunately,
problems are not necessarily unique to the structured
the literature on this issue does not lend itself to defini-
interview, in that they can sometimes be attributed to
tive generalizations. In large part, this ambiguity in the
kindred methods, such as the self-completion question-
broader implications of experiments relating to the effects
naire in survey research or even semi-structured inter-
of interviewer characteristics is due to several prob-
viewing in qualitative research. However, it is common
lems, such as: the problem of disentangling the effects
for the structured interview to be a focus for the identi-
of interviewers’ different attributes from each other
fication of certain limitations that are briefly examined
(ethnicity, gender, socio-economic status); the interac-
below.
tion between the characteristics of interviewers and
 Problems with structured interviewing 227
the characteristics of respondents; and the interaction Acquiescence
between any effects observed and the topic of the inter- Acquiescence refers to a tendency for some people con-
view. Nonetheless, there is undoubtedly some evidence sistently to agree or disagree with a set of questions or
that effects due to characteristics of interviewers can be items. Imagine respondents who replied to all the items in
discerned. Research in focus 8.3 stating that they believed they were
The ethnicity of interviewers is one area that has all unethical (scale = 5) and judging that they and their
attracted some attention. Schuman and Presser (1981) peers infrequently acted in the way implied by the state-
cite a study that asked respondents to nominate two or ment (scale = 1). The problem with this multiple-item
three of their favourite actors or entertainers. Respon- measure is that none of the item measure statements is
dents were much more likely to mention African- written in a way that implies an opposite stance. In other
American actors or entertainers when interviewed by words, there are no items that are ethical or likely to be
African-American interviewers. Schuman and Converse engaged in frequently by many ethically responsible peo-
(1971) interviewed 619 African-American Detroiters ple. This could be seen as a potential source of bias in this
shortly after Martin Luther King’s assassination in 1968. multiple-item measure. A wording that would imply an
The researchers found significant response differences opposite stance might be ‘being prepared to take respon-
depending on whether the interviewers were also Afri- sibility for errors’ or ‘refusing to accept gifts/favours in
can-American in around one-quarter of the questions exchange for preferential treatment’. The inclusion of
asked. some items in this form would help to weed out those
Although this proportion is quite striking, the fact that respondents who were replying within the framework of
the majority of questions appear to have been largely an acquiescence response set.
unaffected does not give rise to a great deal of confidence
that a consistent biasing factor is being uncovered. Simi-
larly inconclusive findings tend to occur in relation to Social desirability bias
experiments with other sets of characteristics of inter- The social desirability effect refers to evidence that some
viewers. These remarks are not meant to play down the respondents’ answers to questions are related to their
potential significance of interviewers’ characteristics for perception of the social desirability of those answers. An
measurement error, but to draw attention to the limi- answer that is perceived to be socially desirable is more
tations of drawing conclusive inferences about the evi- likely to be endorsed than one that is not. This phenom-
dence. All that needs to be registered at this juncture is enon has been demonstrated in studies of ethical behav-
that almost certainly the characteristics of interviewers iour, managerial decision-making, and attitudes towards
do have an impact on respondents’ replies, but that the entrepreneurs (see Research in focus 10.11). In order
extent and nature of the impact are not clear and are to try to prevent social desirability bias, Terence Jackson
likely to vary from context to context. (2001) framed questions in a way that was intended to
enable the respondents to distance themselves from their
responses, by imagining what a peer might do rather
Response sets than having to state what they would do. It was expected
Some writers have suggested that the structured inter- that this would reduce the likelihood that individuals
view is particularly prone to the operation among would respond in a way that they anticipated would
respondents of what Webb et al. (1966) call ‘response be more acceptable. However, Steenkamp et al. (2010)
sets’, which they define as ‘irrelevant but lawful sources have proposed that it is crucial to distinguish between
of variance’ (1966: 19). This form of response bias is social desirability bias that is conscious and that which is
especially relevant to multiple-indicator measure (see unconscious. The underlying motives and psychological
Chapter 8), where respondents reply to a battery of processes associated with each are likely to differ, so that
related questions or items, of the kind found in a Likert it is not obvious whether a procedure like that employed
scale (see Research in focus 8.3). The idea of a response in Research in focus 10.11 applies to both or just one of
set implies that people respond to the series of questions the two forms.
in a consistent way but one that is irrelevant to the con- In so far as these forms of response error go undetected,
cept being measured. Two of the most prominent types of they represent sources of error in the measurement of
response set are known as the ‘acquiescence’ (also known concepts. However, while some writers have proposed
as the ‘yeasaying’ and ‘naysaying’) effect, and the ‘social outright condemnation of social research on the basis
desirability’ effect. of evidence of response sets (e.g. Phillips 1973), it is
228 10 Structured interviewing

10.11 RESEARCH IN FOCUS

A study of the effects of social desirability bias
Ogbulu and Singh (2013) conducted a study on entrepreneurship by African Americans. As part of the study,
they sought to measure social desirability bias and its impact on results from interviews on consumer
perceptions of ‘black’ and ‘white’ (we note that this terminology is potentially problematic, but we are using the
terms which the authors of the article use) entrepreneurs in the USA. The authors summarize their
agenda as follows:

The overarching hypothesis that guided the present research was that respondent data would differ based on
both the race of the respondent and the race of the data collector when asked about racially different
entrepreneurs and their firms. We focused on three areas of inquiry: attitudes toward entrepreneurs,
perceptions of legitimacy and patronage intentions.
(Ogbulu and Singh 2013: 4)

The study involved the administration of a structured questionnaire to a sample of 843 respondents, both black
and white. The questionnaires were administered by both black and white data collectors.

The questionnaire included a series of questions about a fictional venture (a restaurant) which was being
established in the city where data were collected. All the details in the questionnaires were the same except
that one set of questionnaires included a photograph of a black male as the fictional entrepreneur while the
other included a photograph of a white male. The biographical details of the fictional entrepreneur were
identical and no mention of ethnic background was provided. Likert scale items were used to capture
respondent attitudes to the entrepreneur, their opinion of the legitimacy of the venture, and their likelihood of
patronizing the restaurant.

The results showed that black respondents did not rate the black entrepreneur more favourably or as more
legitimate, but reported a greater likelihood of patronizing his restaurant. White respondents reported more
favourable perceptions, higher legitimacy, and more likelihood of patronizing the restaurant if it were being
established by a black entrepreneur. Significantly for our understanding of social desirability bias, white
respondents responded more favourably to the black entrepreneur and indicated higher likelihood to
patronize the restaurant when the questionnaire was administered by a black data collector than if it was
administered by a white data collector. The researchers conclude that this provides evidence that social
desirability bias is likely to influence how people respond to questions about issues of race and ethnicity and
that this may generate misleading results in research on African-American entrepreneurs. This highlights the
need to design studies and analyse data in ways which take account of this possible effect.

important not to get carried away with such findings. We interpretivist ideas of the kinds touched on in Chapter
cannot be sure how prevalent these effects are, and to some 2 (Cicourel 1964, 1982; Filmer et al. 1972; Briggs 1986;
extent awareness of them has led to measures to limit their Mishler 1986). This critique revolves around what is
impact on data (for example, by weeding out cases obvi- often referred to in a shorthand way as the ‘problem
ously affected by them) or by instructing interviewers to of meaning’. The kernel of the argument is that when
limit the possible impact of the social desirability effect by humans communicate they do so in a way that not only
not becoming overly friendly with respondents and by not draws on commonly held meanings but also simulta-
being judgemental about their replies. neously creates meanings. ‘Meaning’ in this sense is
something that is worked at and achieved—it is not
simply pre-given. Allusions to the problem of meaning
The problem of meaning in structured interviewing draw attention to the notion
A critique of survey interview data and findings gleaned that survey researchers presume that interviewer and
from similar techniques was developed by social sci- respondent share the same meanings of terms employed
entists influenced by phenomenological and other in the interview questions and answers. In fact, the