The 2nd International Conference on Inovations

in Social Sciences Education and Engineering ( ICoISSEE )

August 07th, 2021

CHARACTERIZING INTERVIEW-BASED STUDIES IN

CONSTRUCTION MANAGEMENT RESEARCH: ANALYSIS OF
EMPIRICAL LITERATURE EVIDENCES

Seng Hansen1*
1
Construction Engineering and Management Program, Universitas Agung Podomoro,
Jakarta

*Corresponding author: [Link]@[Link]

Abstract. It has become more common in construction management research to

use interview technique to gather qualitative data on practitioner opinions, attitudes,
skills and best practices. This data is then analyzed using various qualitative
analyses to produce findings and implications. However, one of the main problems
raised in this study is the lack of standards and guidelines in conducting interview
technique. Some of the issues are the lack of interview size justifications, the
absence of criteria in selecting interviewees, and the lack of explanation regarding
the analysis type conducted. These may cause weak interview data and lack of
reliability of interview results. An empirical survey of 223 relevant papers was done
to investigate this problem. This study reviews the gathered literature and presents
descriptive statistics to characterize interview technique in construction
management field. The discussions offer identification of interview characteristics
in construction management research, explanation of factors influencing the data
quality and saturation, and elucidation of factors influencing the analysis quality.
Finally, a straightforward guideline to assist researchers in determining interview
size in their studies is proposed.

Keywords: Construction management research, Data saturation, Empirical review,

Interviews.

1. INTRODUCTION
Interview technique has often been used as one of the data collection methods in
construction management research. It is considered as a qualitative method since it is
characterized by an in-depth and holistic observation of a unique phenomenon. In
practice, it involves at least two people who are the interviewer and the interviewee
(Hofisi, Hofisi & Mago, 2014). The data obtained tends to be qualitative rather than
quantitative (Galvin, 2015).
The popularity of interview as a data collection technique has substantially
increased in recent years. The main value of interview technique is the richness in detail
of the collected data. Construction management research can benefit from using
interview technique to obtain a more realistic view of the phenomenon under
investigation that cannot be understood using numerical data. It is descriptive in nature
which allows the researcher to describe existing phenomena and situations. However,
there have been several criticisms given in previous studies. These criticisms are
mainly related to the lack of standards and guidelines in interview technique, data
collection and analysis are often cost and time consuming, highly dependent on the
researcher’s ability to collect and analyze the data, lack of consistency and reliability,
 The 2nd International Conference on Inovations
in Social Sciences Education and Engineering ( ICoISSEE )
August 07th, 2021

and ethical issues since it involves human subjects (Alshenqeeti, 2014; Bogner & Menz,
2009; Hofisi, Hofisi & Mago, 2014).
The lack of standards and guidelines is also related to determining sample size
and expert justifications in interview technique. Generally, samples in qualitative
research tend to be small in order to support the depth of case-oriented analysis.
Additionally, interview samples are purposively selected by virtue of their capacity to
provide relevant information (Vasilelou et al. 2018). Some researchers have tried to
establish the recommended number of interviews. For instance, Guest, Bunce &
Johnson (2006) recommend at least 12 interviews while Kwok & Lau (2018) and Yang
& Shen (2015) suggest a range from 10 to 20 interviews. Therefore, it is quite
problematic to determine the interview size prior data collection. On the other hand,
analysis of interview data is very dependent on the skills of the researchers involved
which may result in the inconsistent findings of similar studies (lack of reliability).
This paper aims to investigate these problems from a statistical point of view
through a systematic analysis of empirical literature evidence. An extensive literature
over a 20-year period from Journal of Construction Engineering and Management
(JCEM) and Media Komunikasi Teknik Sipil (MKTS) was used as the scope of this
study. JCEM was chosen because it is a top global journal in the construction
management field which has been publishing empirical studies online since 1983.
Meanwhile, MKTS was chosen as a comparison from Indonesia. There has been no
previous study that empirically discusses interview technique of both publications. This
study contributes for future construction management research considering the
increased popularity of interview technique.

2. RESEARCH METHODS/METHODOLOGY
This paper adopts empirical literature study to gain insights from literature
evidences. According to Bettis et al. (2014), empirical studies can be purely
exploratory using graphical and statistical approaches. In this paper, a systematic
literature review was used to analyze the empirical literature evidences focusing on
identifying and characterizing interview technique in construction management
research. Table 1 presents the systematic literature review strategies as prescribed
by Chan & Owusu (2017) and Hansen, Rostiyanti & Rif’at (2020).
Table 1. Systematic literature review strategies
No Strategies
1 Targeting literature sources:
JCEM and MKTS are used to represent international and national academic journal in
the construction management field. Both are top referred journals and have frequently
published online high-quality construction management papers since 1983 (JCEM) and
2003 (MKTS). In doing so, both journals use a transparent peer-reviewed procedure.
2 Searching for related literature:
The search engine in JCEM was located and an advanced search was performed to
narrow down the search so that only related articles will be retrieved. All papers
containing the word “interview” related to the topic of “construction management”
published within the last 20 years (from 2000 to July 23, 2019) were extracted from the
JCEM virtual library. A total of 299 articles were retrieved. Similarly, the search engine
in MKTS was used to find all papers containing the word “wawancara” or “interview”.
Six articles were retrieved.
3 Selecting relevant publications:
Once the related papers were available, a visual examination through reading the
abstracts and skimming the contents was performed to identify relevant publications in
this study. For that, the following selection criteria are used to include or exclude the
retrieved articles: (1) interview technique is used as a method of data collection and/or
validation, (2) interview technique which is referred to only as a reference (as in the
literature review section) was excluded, (3) all types of data was included (i.e. studies
 The 2nd International Conference on Inovations
in Social Sciences Education and Engineering ( ICoISSEE )
August 07th, 2021

that quantified their qualitative interview data was included), and (4) studies that did not
provide the exact number of interview size was included as well. As the result, a total of
223 articles were selected for analysis.
4 Analyzing the content:
Thematic content analysis was performed to analyze the content of selected
publications. Some themes observed in this study include: year of publication, number
of interviews, number of interviewees, interview type, interview format, interview
technique, average interview duration, and research approach. After each article was
analyzed and coded according to the theme, descriptive statistics are used to explain
certain characteristics of interview technique.
5 Discussing and presenting the findings:
As part of data presentation, figures and tables are used to illustrate the characteristics
of interview technique. All findings and implications were discussed and presented in
the ‘Results and Discussion’ section.

3. RESULTS AND DISCUSSION

There are two types of findings in this study, namely descriptive and statistical.
Descriptive findings describe qualitatively various issues related to interview technique
as a method of data collection and/or validation. This includes some key descriptions of
interview technique, data saturation statements, expert justifications, and other
characteristics. Basic statistics are then used to illustrate relationships between the
existing characteristics which at the same time strengthened the descriptive findings.
The results are grouped and discussed in the following four subsections.

3.1 Characteristics of the Interview Technique

The result of this study shows a positive trend in the use of interview technique in
construction management research. This study reviewed the number of publications of
JCEM for 20-year period which can be seen in Figure 1 (with dotted line indicating the
increasing trend). On the other hand, the use of interview technique in MKTS has not
been intensified over the years. There are only six articles that employed this technique.
25

20
No. of studies

15

10

Year of publication

Figure 1. Number of studies over 20-year period

The number of interviews and interviewees can also be grouped to identify which
groups are most often found in this study. Number of interviews and number of
interviewees are distinguished because at the time of analysis, several situations were
found where the number of interviews was not always the same as the number of
interviewees. For example, Zhou, Fang & Mohamed (2011) interviewed four experts but
in two groups. Kale (2009) interviewed three experts but in four series of interviews.
Thus, it can be concluded that there are four types of interviews:
(1) single interview. The interview is conducted in a one-to-one basis, meaning one
interview reflects one interviewee. This is the most common type found in this
study.
(2) group interview. An interview is conducted with many interviewees at the same
time. Examples can be found in Zhou, Fang & Mohamed (2011) and Chen &
 The 2nd International Conference on Inovations
in Social Sciences Education and Engineering ( ICoISSEE )
August 07th, 2021
Fong (2013).
(3) interview rounds. An interviewee can be involved in more than one interview.
This usually happens when the involvement of an interviewee is needed more
than once. An example of this type can be found in Shapira & Goldenberg
(2007).
(4) unclear number of interviews/interviewees. In this case, the number of
interviews and/or interviewees are not clearly stated in the publication.
Researchers only explain that they have conducted interviews with several
experts as representatives of groups or populations. Examples can be found in
Arriagada & Alarcón (2014) and Ozorhon & Oral (2017).
The groupings of number of studies to number of interviews is illustrated in Figure
2. From 217 JCEM papers, as many as 67 papers (30.9%) did not clearly provide
number of interviews. It shows that the majority of researchers use small to medium
interview size in their studies. While there is no agreement regarding the category of
sample size in interview technique, this study considers number of interviews less than
10 as small size, between 10 to 20 as medium size, and above 20 as large size based
on the distribution of number of studies to the interview range.
70

60

50
No. of studies

40

30

20

10

1-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-100 >100
No. of interviews

Figure 2. The grouping of number of interviews

Basic descriptive statistics for JCEM papers are displayed in Table 2. It was found
that the mean value of interview number is 28.1. The smallest sample of interviews in
the studied papers was 1 and the largest was 860. Based on the analysis, small size
interviews are commonly found in in-depth interviews while large size interviews are
found in shallow interviews. Interestingly, the median of interview and interviewee
numbers is 12. This corresponds to the opinion of Guest, Bunce & Johnson (2006) who
concluded that data saturation was reached within the number of 12 interviews.
Table 2. Descriptive statistics of JCEM papers
Sample size of studies No. of interviews (n=150)
Mean 28.1
SD 74.9
Min number 1
Max number 860
Median 12
This study also succeeded in identifying two purposes of conducting interview
technique in JCEM: as primary method and as secondary method. Interview as primary
method occurs when interview is used as the main approach in collecting data by
researchers. Usually this can be easily identified if an interview procedure is provided
in the ‘Methods’ section of the publication. Meanwhile, interview as secondary method
occurs when it is not used as the main method of data collection but rather as a
complementary method. There are two situations: (1) interview is used as an initial
technique to assist researchers in survey design (e.g. Tabish & Jha, 2018), and (2)
interview is used after the main technique as a way to validate the findings (e.g. Aljassmi
& Han, 2013). It was found that the majority of studied papers (59.9%) used interviews
as a secondary method. On the other hand, this study cannot provide descriptive
statistics for MKTS papers due to lack of details available in the publications.

3.2 Factors Influencing Data Quality and Data Saturation

 The 2nd International Conference on Inovations
in Social Sciences Education and Engineering ( ICoISSEE )
August 07th, 2021
Other characteristics can be grouped based on their similarities, i.e. as factors
influencing data quality and data saturation. Both data quality and data saturation are
very closely related. In fact, data saturation has become a central issue when
conducting interview technique since it affects the quality of data produced. It is defined
as ‘the point at which no new relevant information is forthcoming, even if more people
are interviewed’ (Galvin, 2015). While this issue is crucial, most of the studied articles
however did not provide justifications on how they achieved data saturation or whether
they had reached it. Only few papers provided certain statements which may lead to the
conclusion of them having reach data saturation (as shown in Table 3).
Table 3. Papers containing data saturation statement
References Quotes
Orozco et al. “Given the similar answers of these individuals, no additional responses
(2014) were necessary.”
Dehghan & “The interviews continued until no new information could be extracted
Ruwnapura (2014) and information saturation was reached.”
Lim, Schultmann “Therefore, samples may be small as long as “saturation” occurs: this is
& Ofori (2010) when no new information emerges that is important for the study.”
Kim et al. (2019) “In particular, the research team performed expert interviews one-by-
one until information saturation was observed in the same manner as in
previous interview-driven studies.”
The results of extensive empirical literature analysis provide several factors
influencing the data saturation. In general, data saturation can be quickly reached in
studies with a narrower scope than those with a broader research scope. For instance,
a study on BIM education for CEM students (Sacks & Pikas, 2013) is narrower in scope
than a project wanting to formulate fuzzy enterprise risk management maturity model
for construction firms (Zhao, Hwang & Low, 2013).
On the other hand, interview size greatly affects data saturation. Therefore, it is
important to justify the size of interviews in the study, especially considering that almost
all research has limitations regarding time and cost to invest in data gathering. In
general, the larger interview size conducted, the more data collected and can be coded
which will improve the data quality. While it is important, most of the JCEM studied
papers did not provide an interview size justification in their study (94.9%). The
remaining provides three types of justifications as shown in Table 4. Similarly, all MKTS
papers did not provide an interview size justification.
Table 4. Interview size justifications in JCEM
Type of justification Freq. %A %B
Data saturation 4 1.8% 36.4%
Rules of thumb 4 1.8% 36.4%
Pragmatic reasons 3 1.4% 27.3%
Unjustified studies 206 94.9% -
Total 217 100% 100%
%A: percentage from total number of studies,
%B: percentage excluding unjustified studies
Achieving data saturation is one of the two most common interview size
justifications besides the rule of thumb. Here, the researchers explained that their
interviews were carried out until data saturation is reached. Examples of papers use
data saturation as interview size justification include Deghan & Ruwnapura (2014),
Orozco et al. (2014), and Kim et al. (2019).
On the other hand, rules of thumb have been commonly used for interview size
justification. It is based on previous studies recommendations which reflected the
achievement of data saturation from similar past research experience. For example, a
recommendation range from 10 to 20 interviews given by Kwok & Lau (2018) and Yang
& Shen (2015) has been followed by Bahadorestani, Karlsen & Farimani (2019), and
recommendation exceeds 12 interviews as suggested by Guest, Bunce & Johnson
(2006) has been followed by Nguyen (2019).
Three studies have used pragmatic reasons to justify their interview size. Menches
& Hanna (2006) for example, argue that the interview size will largely depend on the
available budget since they were conducting face-to-face interviews and cost was a
 The 2nd International Conference on Inovations
in Social Sciences Education and Engineering ( ICoISSEE )
August 07th, 2021
significant factor in the sampling methodology. Guo et al. (2017) reason that interview
size in their study was determined based on the consideration that any information
provided by an interviewee must be supported or verified by another. Meanwhile,
determination of interview size based on representation of interviewee for each of
different population in the study was adopted by Raoufi & Fayek (2018).
The interview duration can also affect data quality and saturation. In general, the
interview duration indicates the richness of data can be collected. It depends on many
other factors such as the type of interview, the number of interview questions being
asked, and the availability of interviewee’s time. It is generally expressed in terms of
average duration (e.g. the average interview duration was approximately 1.5 h in
Deghan & Ruwnapura’s 2014 study) or duration range (e.g. interviews lasted between
20 and 45 minutes in O’Connor & Mock’s 2019 study). Although it is important to
describe the quality of the obtained data, not many researchers clearly state the
interview duration in their papers. The analysis of this study showed that 88.5% of JCEM
papers did not express the interview duration while the remaining 25 papers that
expressed interview duration can be grouped into three categories: those with duration
less than 1 hour (11 papers or 44%), duration between 1 to 2 hours (9 papers or 36%),
and duration more than 2 hours (5 papers or 20%). On the other, all MKTS papers did
not express the interview duration.
Interview format which can be grouped into three types (unstructured, semi-
structured, and structured interviews) also influences data quality and saturation. The
difference between these three types lies in the flexibility of interviewers in asking
questions (flexible/prompt) and the freedom of interviewees in giving responses. For
instance, in the case of semi-structured interviews, the interviewer may have a list of
interview questions but he can act promptly by asking probe questions adjusting the
interviewee’s response. Likewise, interviewee can provide responses freely, on and
around the topic area, which may allow new ideas and insights. In this study, it was
found that as many as 65% of JCEM papers did not express their interview format while
the remaining 76 papers can be divided into unstructured, semi-structured and
structured interviews with 3.9%, 65.8% and 30.3% respectively. Thus, semi-structured
interviews are the most common interview format in construction management
research.
The diversity of interviewees may affect the data quality as well. It can be divided
into two: homogenous and heterogenous. Homogenous means that the interviewees
have similar characteristics whereas interviewees with dissimilar characteristics are
grouped as heterogenous. In general, the more heterogenous the more diverse the
given responses, meaning that achieving data saturation may take longer period.
Similarly, the more homogenous the interviewee, the sooner data saturation will occur
(Bonde, 2013). Figure 3 displays typical data saturation patterns from interviewees
diversity perspective. In case of homogenous interviewees, the data saturation is
reached earlier since the majority of information comes from initial interviews with less
and less new information being revealed by later interviews. Meanwhile, in case of
heterogenous interviewees, the wide diversity will provide more new insights and hence
making the data saturation is reached later than homogenous interviewees.

Heterogenous
Number of codes

Homogenous

Number of interviews

Figure 3. Data saturation pattern

Interviewee type is one of the important factors that influence data quality and
 The 2nd International Conference on Inovations
in Social Sciences Education and Engineering ( ICoISSEE )
August 07th, 2021
saturation. For instance, interview sampling is generally done purposively where
interviewees are chosen because of their expertise. Thus, it is crucial to have expert
justifications as criteria in selecting interviewees. This study has successfully identified
eleven expert criteria as shown in Table 5. While providing expert justification is
essential to ensure data quality, it was found that very few researchers (only 11 of JCEM
papers) provided justifications in the selection of expert interviewees.
Table 5. Expert justifications
Expert criteria n %
Practical experience 11 28.2%
Educational background 5 12.8%
Certification/accreditation 4 10.3%
Other competencies/skills 3 7.7%
Trainings and activities 2 5.1%
Membership 1 2.6%
Publications 1 2.6%
Perceived knowledge 4 10.3%
Position within organization 3 7.7%
Ability to communicate knowledge 2 5.1%
Practical considerations 3 7.7%
It appears that practical experience is the main criterion in justifying one’s expertise.
This can be assessed from two aspects: how far the direct involvement of the
interviewee on the research topic and how long the interviewee has related experience
in the field. For instance, Zhang & El-Gohary (2016) mentioned that an expert must be
experienced in the AEC (architectural, engineering and construction) domain, while
Patel & Jha (2017) stated that an expert must have at least 10 years of professional
experience in construction safety. This study found that majority of these eleven papers
requires at least 10 years of practical experience.
Educational background relates to the required degrees in relevant fields. For
instance, an expert must have a Master or a Ph.D. degree in the field of project
management (Bahadorestani, Karlsen & Farimani, 2019). A person can also be called
an expert if he already has a certificate of expertise or accredited in a particular field.
For instance, an expert must be a practitioner who is accredited mediator/arbitrator of
the recognized professional bodies (Chan, Suen & Chan, 2006). Other competencies
relate to other personal skills required in the relevant studies, such as having awareness
of knowledge modeling (Zhang & El-Gohary, 2016) or hold BIM-related roles within their
organizations (Akintola, Venkatachalam & Root, 2017).
Some studies also required experts to have certain trainings or activities, such as
at least 2 weeks of OHS related training, invited to present at a conference, etc. (Patel
& Jha, 2017). Expertise can also be judged by the membership of the interviewees in a
recognized professional body as well as the number of publications produced by the
interviewees (Patel & Jha, 2017). Some studies may not require relevant educational
background of interviewees as long as they have relevant knowledge. For instance,
Chan, Suen & Chan (2006) considered experts are those who exhibit a good
understanding of the alternative dispute resolution.
Most importantly, experts must be able to communicate their knowledge while
voluntarily being interviewed (Brockman, 2014; Akintola, Venkatachalam & Root, 2017).
Thus, the quality of interview data obtained also depends on enthusiasm and
willingness of interviewees to share their knowledge and opinions. Some studies may
require experts to be those with certain job positions within their organizations (Le et al.
2014; Akintola, Venkatachalam & Root, 2017). This is closely related to interviewees’
roles which makes it easy for them to access certain information needed for the
research. Meanwhile, practical considerations may include a selection of interviewees
which considers the diversity of expertise and geographical location of the experts (Le
et al. 2014).
Judging from the readiness of the researchers/interviewers, data quality and
saturation depend on the researcher’s skills and expertise when conducting interviews.
 The 2nd International Conference on Inovations
in Social Sciences Education and Engineering ( ICoISSEE )
August 07th, 2021
This includes having adequate knowledge and technical background on the research
topic, readiness in conducting interviews (such as having interview protocol, listing
appropriate interview questions, etc.) as well as the quality of interview dialogue
(whether interview dialogue runs smoothly/not and strong/weak). In short, the more
experienced researchers are with the subject of investigation, the fewer interviews are
required to reach data saturation. In addition, research limitations such as time and
budget allowance to conduct the studies can also affect the number of interviews that
can be conducted which ultimately influences the achievement of data saturation.
Finally, external factors such as requirements by the research funding board and
interventions by supervisors may also influence the quality of data and achievement of
data saturation.

3.3 Factors Influencing Analysis Quality

Another important aspect to consider when using interview technique is the quality
of interview analysis. As a technique in qualitative research, interview data analysis
tends to be subjective and relies heavily on the expertise of the researchers conducting
the analysis. Here, the more expert a researcher, the better the analysis quality
produced. If the researchers are familiar with the subject of the phenomena under study
or have had previous similar studies experience, it is easier for them to carry out
comprehensive qualitative analysis and reached data saturation.
The data type also influence the analysis quality conducted. While interview is a
qualitative technique, the obtained data can be analyzed and produced quantitative
findings. In this study, it should be noted that some of the studied papers have provided
interview results combining both qualitative and quantitative findings. These quantitative
findings may be measurement of codes frequency or descriptive statistics of particular
issues (e.g. interviewees’ profiles as in Yang et al. 2010 and Zhao et al. 2015).
Furthermore, analysis quality can also be influenced by the type of analysis
conducted. There are three types of qualitative analysis commonly used: content
analysis, coding analysis and discourse analysis. Content analysis is a technique for
making replicable and valid inferences to the contexts of their use (Krippendorff, 2012).
It enables a systematic process to examine the content of interview transcripts.
Qualitative data is reduced to concepts describing the research phenomena by creating
categories, conceptual systems, or conceptual maps (Elo et al. 2014). Examples of the
studied papers that use content analysis are Taylor (2007) and Zhao, Hwang & Low
(2014).
Similar to it, coding analysis also known as thematic coding analysis is a commonly
data analysis approach which emphasizes the codification aspect of the data to find
explicit and implicit patterns and relationships of the studied phenomena. Examples of
the studied papers using coding analysis are Blacud et al. (2009) and Jeelani, Albert &
Gambatese (2017). Meanwhile, discourse analysis involves detailed investigation of
statements and arguments that are regularly uttered by interviewees. None of the
studied papers used this type of analysis.
Furthermore, the analysis quality is also determined by the ability of its results to
be triangulated with other findings. This is commonly found in mixed method studies.
For instance, Tabish & Jha (2018) used face-to-face interviews as an initial method to
develop questionnaire. The results of interviews were used to modify the questionnaire
before it was distributed to the main respondents. Similarly, Tao et al. (2017) used
interviews to solicit the experts’ opinions before focus groups and questionnaire surveys
were conducted.
Currently, there are many qualitative analysis software available to support
interview data analysis such as NVIVO, MAXQDA, [Link], etc. These software helps
researchers to process interview transcripts by providing a wide range of features to
conduct content analysis, discourse analysis or coding analysis. However, only few of
the studied papers have expressed the use of these support tools in their studies. These
includes Lee et al. (2011), Brockman (2014), Chan et al. (2016), and Jeelani, Albert &
Gambatese (2017) who used various versions of NVIVO software package.
Both factors influencing data quality and saturation as well as factors influencing
analysis quality will ultimately affect the reliability of interview results. Therefore, it is
crucial for researchers who use interview technique as a qualitative method to
 The 2nd International Conference on Inovations
in Social Sciences Education and Engineering ( ICoISSEE )
August 07th, 2021
understand the factors above.

3.4 A Structured Guideline in Determining Interview Size

In this study, the author develops a structured guideline in determining interview
size. Interview size determination is still a major and challenging issue in construction
management research. Therefore, based on the characteristics and factors that have
been identified above, this study proposes a structured matrix as shown in Table 6. It
divides interviews based on their sizes into small, medium and large interviews. There
are eight key qualifiers as benchmarks in determining the recommended interview size.
Researchers can use this matrix to determine the recommended interview size based
on the dominant group of identified key qualifiers.
Table 6. Interview size guideline
Size
Key Qualifiers
Small Medium Large
Number of interviews less than 10 10 ≤ x < 20 more than 20
Research scope recommended for in between recommended for
narrow/focus broad/complex
research scope research scope
Research limitations research research may be no limitation
resources are limited by one
constrained by factor
time, cost and
other factors
Interview format recommended for recommended for recommended for
unstructured and unstructured and semi-structured
semi-structured semi-structured and structured
interview formats interview formats interview formats
Diversity of interviewees likely for in between likely for
homogenous heterogenous
target interviewees target interviewees
Interviewee’s expertise target interviewees in between target interviewees
possess high level possess low level
of expertise of expertise
Researcher’s expertise more experienced in between less experienced
researchers researchers
require fewer require more
interviews interviews
Research characteristics the phenomenon in between the phenomenon
under investigation under investigation
is unique is comparable to
others
An illustration to use the above matrix is provided as follows. A Ph.D. student raises
the topic of construction project risk assessment from various perspectives (in between
narrow and broad research scope) as his dissertation. He plans to interview several
professionals from contractors, owners, and consultants (heterogenous and experts).
He decides to conduct semi-structured and face-to-face interviews (cost becomes a
significant factor in the data collection process). Considering these conditions, a
medium size interview is recommended.
Another example of the compatibility of this matrix with previous research can be
found in Vaughan et al. (2013) study. They investigated the cost-benefit analysis of
construction information management system implementation (narrow) from data
collected over a 6-month period (limited). The target interviewees were members of a
project team (homogenous) with average experience of 13.8 years (high level of
expertise). The study is unique by developing a framework to assess costs and benefits
of innovative construction information management system. Thus, they have employed
five expert interviewees which is considered as a small interview size.
Finally, the author realizes that determining the sample size in interview technique
will depend on many factors other than those identified as key qualifiers above.
Interview as one of data collection technique is investigative and explorative towards a
phenomenon so that the reliability of the results is very dependent on the quality of the
data collected and the analysis conducted.
 The 2nd International Conference on Inovations
in Social Sciences Education and Engineering ( ICoISSEE )
August 07th, 2021

CONCLUSION
This empirical literature study focused on publications that used interview
technique as part of their data collection and/or validation method. A total of 223 papers
published since 2000 was used as sample in this study. Through a comprehensive
analysis, the characteristics of interview technique was evaluated. Results show that
the use of interview technique in construction management research globally has
increased substantially over the past two decades. Meanwhile, its use in Indonesian
context must still be promoted.
In summary, this study recommended future studies using interview technique to
consider the following: (1) provide clear expert justifications, including the required
practical experience, educational background, certifications, and other competencies,
(2) provide clear interview size justifications, (3) provide interview purpose, type, format,
duration, and limitations (if any), (4) provide the profiles of interviewees, (5) provide the
type of analysis conducted and support tools used for the analysis (if any), and (6)
provide a summary of interview key responses and results.
This study has three main contributions. Firstly, through an empirical literature
study it presents the key characteristics of interview technique as found in JCEM &
MKTS publications. The identification of these characteristics is crucial to better
understand the nature of interview technique in construction management field.
Secondly, it has identified several factors influencing data quality and saturation as well
as analysis quality. Understanding these factors is crucial to produce reliable interview
results. Lastly, this study presents a structured guideline in determining interview size.
This structured guideline is derived from an assessment of 223 papers and can be used
to assist further research using interview technique as part of their data collection and/or
validation.

REFERENCES
Akintola, A., Venkatachalam, S., & Root, D. (2017). “New BIM roles’ legitimacy and changing
power dynamics on BIM-enabled projects.” Journal of Construction Engineering and
Management, 143(9), 04017066.
Aljassmi, H., & Han, S. (2013). “Analysis of causes of construction defects using fault trees and
risk importance measures.” Journal of Construction Engineering and Management, 139(7),
870-880.
Alshenqeeti, H. (2014). “Interviewing as a data collection method: a critical review.” English
Linguistics Research, 3(1), 39-45.
Arriagada, R. E., & Alarcón, L. F. (2014). “Knowledge management and maturation model in
construction companies.” Journal of Construction Engineering and Management, 140(4),
B4013006.
Bahadorestani, A., Karlsen, J. T., & Farimani, N. M. (2019). “A comprehensive stakeholder-
typology model based on salience attributes in construction projects.” Journal of
Construction Engineering and Management, 145(9), 04019048.
Bettis, R., Gambardella, A., Helfat, C., & Mitchell, W. (2014). “Quantitative empirical analysis in
strategic management.” Strategic Management Journal, 35, 949-953.
Blacud, N. A., Bogus, S. M., Diekmann, J. E., & Molenaar, K. R. (2009). “Sensitivity of construction
activities under design uncertainty.” Journal of Construction Engineering and Management,
135(3), 199-206.
Bogner, A., & Menz, W. (2009). “The theory-generating expert interview: epistemological interest,
forms of knowledge, interaction.” In: Bogner, A., Litlig, B., & Menz, W. (editors). Interviewing
experts. Plagrave Macmillan, New York.
Bonde, D. (2013). Qualitative interviews: when enough is enough. Research by Design, Australia.
(online) available at: [Link] (accessed 30.03.20).
Brockman, J. L. (2014). “Interpersonal conflict in construction: cost, cause, and consequence.”
Journal of Construction Engineering and Management, 140(2), 04013050.
Chan, A. P. C., Javed, A. A., Lyu, S., Hon, C. K. H., & Wong, F. K. W. (2016). “Strategies for
improving safety and health of ethnic minority construction workers.” Journal of Construction
Engineering and Management, 142(9), 05016007.
Chan, A. P. C., & Owusu, E. K. (2017). “Corruption forms in the construction industry: literature
review.” Journal of Construction Engineering and Management, 143(8), 04017057.
Chan, E. H. W., Suen, H. C. H., & Chan, C. K. L. (2006). “MAUT-based dispute resolution selection
model prototype for international construction projects.” Journal of Construction Engineering
 The 2nd International Conference on Inovations
in Social Sciences Education and Engineering ( ICoISSEE )
August 07th, 2021
and Management, 132(5), 444-451.
Chen, L., & Fong, P. S. W. (2013). “Visualizing evolution of knowledge management capability in
construction firms.” Journal of Construction Engineering and Management, 139(7), 839-851.
Dehghan, R., & Ruwnapura, J. Y. (2014). “Model of trade-off between overlapping and rework of
design activities.” Journal of Construction Engineering and Management, 140(2), 04013043.
Elo, S., Kääriäinen, M., Kanste, O., Pölkki, T., Ultriainen, K., & Kyngäs, H. (2014). “Qualitative
content analysis: a focus on trustworthiness.” SAGE Open, January-March 2014, 1-10.
Galvin, R. (2015). “How many interviews are enough? Do qualitative interviews in building energy
consumption research produce reliable knowledge?” Journal of Building Engineering, 1, 2-
12.
Guest, G., Bunce, A., & Johnson, L. (2006). “How many interviews are enough? An experiment
with data saturation and variability.” Field Methods, 18(1), 59-82.
Guo, B. H. W., Yiu, T. W., González, V. A., & Goh, Y. M. (2017). “Using a pressure-state-practice
model to develop safety leading indicators for construction projects.” Journal of Construction
Engineering and Management, 143(2), 04016092.
Hansen, S., Rostiyanti, S. F., & Rif’at, A. (2020). “Causes, effects, and mitigations framework of
contract change orders: lessons learned from GBK Aquatic Stadium Project.” Journal of
Legal Affairs and Dispute Resolution in Engineering and Construction, 12(1), 05019008.
Hofisi, C., Hofisi, M., & Mago, S. (2014). “Critiquing interviewing as a data collection method.”
Mediterranean Journal of Social Sciences, 5(16), 60-64.
Jeelani, I., Albert, A., & Gambatese, J. A. (2017). “Why do construction hazards remain
unrecognized at the work interface?” Journal of Construction Engineering and Management,
143(5), 04016128.
Jung, W., Han, S. H., Park, H., & Kim, D. Y. (2010). “Empirical assessment of internationalization
strategies for small and medium construction companies.” Journal of Construction
Engineering and Management, 136(12), 1306-1316.
Kale, S. (2009). “Fuzzy intellectual capital index for construction firms.” Journal of Construction
Engineering and Management, 135(6), 508-517.
Kim, J., Ham, Y., Chung, Y., & Chi, S. (2019). “Systematic camera placement framework for
operation-level visual monitoring on construction jobsites.” Journal of Construction
Engineering and Management, 145(4), 04019019.
Krippendorff, K. (2012). Content analysis: an introduction to its methodology (3rd edition). SAGE,
Beverly Hills.
Kwok, P. K., & Lau, H. Y. (2018). “A modified consensus-building methodology for reaching a
group decision using minimum costs.” IEEE Access, 6, 3509-3523.
Le, Y., Shan, M., Chan, A. P. C., & Hu, Y. (2014). “Investigating the causal relationships between
causes of and vulnerabilities to corruption in the Chinese public construction sector.” Journal
of Construction Engineering and Management, 140(9), 05014007.
Lee, S. H., Jeon, R. K., Kim, J. H., & Kim, J. J. (2011). “Strategies for developing countries to
expand their shares in the global construction market: phase-based SWOT and AAA
analyses of Korea.” Journal of Construction Engineering and Management, 137(6), 460-470.
Lim, J. N., Schultmann, F., & Ofori, G. (2010). “Tailoring competitive advantages derived from
innovation to the needs of construction firms.” Journal of Construction Engineering and
Management, 136(5), 568-580.
Menches, C. L., & Hanna, A. S. (2006). “Conceptual planning process for electrical construction.”
Journal of Construction Engineering and Management, 132(12), 1306-1313.
Namian, M., Albert, A., Zuluaga, C. M., & Jaselskis, E. J. (2016). “Improving hazard-recognition
performance and safety training outcomes: integrating strategies for training transfer.”
Journal of Construction Engineering and Management, 142(10), 04016048.
Nguyen, L. H. (2019). “Relationships between critical factors related to team behaviors and client
satisfaction in construction project organizations.” Journal of Construction Engineering and
Management, 145(3), 04019002.
O’Connor, J. T., & Mock, B. D. (2019). “Construction, commissioning, and startup execution:
problematic activities on capital projects.” Journal of Construction Engineering and
Management, 145(4), 04019009.
Orozco, F. A., Serpell, A. F., Molenaar, K. R., & Forcael, E. (2014). “Modeling competitiveness
factors and indexes for construction companies: findings of Chile.” Journal of Construction
Engineering and Management, 140(4), B4013002.
Ozorhon, B., & Oral, K. (2017). “Drivers of innovation in construction projects.” Journal of
Construction Engineering and Management, 143(4), 04016118.
Patel, D. A., & Jha, K. N. (2017). “Developing a process to evaluate construction project safety
hazard index using the possibility approach in India.” Journal of Construction Engineering
and Management, 143(1), 04016081.
 The 2nd International Conference on Inovations
in Social Sciences Education and Engineering ( ICoISSEE )
August 07th, 2021
Pellicer, E., Yepes, V., Correa, C. L., & Alarcón, L. F. (2014). “Model for systematic innovation in
construction companies.” Journal of Construction Engineering and Management, 140(4),
B4014001.
Raoufi, M., & Fayek, A. R. (2018). “Framework for identification of factors affecting construction
crew motivation and performance.” Journal of Construction Engineering and Management,
144(9), 04018080.
Sacks, R., & Pikas, E. (2013). “Building information modeling education for construction
engineering and management. I: Industry requirements, state of the art, and gap analysis.”
Journal of Construction Engineering and Management, 139(11), 04013016.
Shapira, A., & Goldenberg, M. (2007). “Soft considerations in equipment selection for building
construction projects.” Journal of Construction Engineering and Management, 133(10), 749-
760.
Tabish, S. Z. S., & Jha, K. N. (2018). “Beyond the iron triangle in public construction projects.”
Journal of Construction Engineering and Management, 144(8), 04018067.
Tao, L., Wu, C., Chiang, Y. H., Wong, F. K. W., & Liang, S. (2017). “Generational perceptions of
freedom-related work values: Hong Kong’s implementation of a no-Saturday-site-work policy
in construction.” Journal of Construction Engineering and Management, 143(7), 06017002.
Taylor, J. E. (2007). “Antecedents of successful three-dimensional computer-aided design
implementation in design and construction networks.” Journal of Construction Engineering
and Management, 133(12), 993-1002.
Vasileiou, K., Barnett, J., Thorpe, S., & Young, T. (2018). “Characterising and justifying sample
size sufficiency in interview-based studies: systematic analysis of qualitative health research
over a 15-year period.” BMC Medical Research Methodology, 18(148), 1-18.
Vaughan, J. L., Leming, M. L., Liu, M., & Jaselskis, E. (2013). “Cost-benefit analysis of
construction information management system implementation: case study.” Journal of
Construction Engineering and Management, 139(4), 445-455.
Yang, J., Shen, G. Q., Drew, D. S., & Ho, M. (2010). “Critical success factors for stakeholder
management: construction practitioners’ perspectives.” Journal of Construction Engineering
and Management, 136(7), 778-786.
Yang, R. J., & Shen, G. Q. (2015). “Framework for stakeholder management in construction
projects.” Journal of Management in Engineering, 31(4), 04014064.
Zhang, L., & El-Gohary, N. M. (2016). “Epistemology-based context-aware semantic model for
sustainable construction practices.” Journal of Construction Engineering and Management,
142(3), 04015084.
Zhao, X., Hwang, B. G., & Low, S. P. (2013). “Developing fuzzy enterprise risk management
maturity model for construction firms.” Journal of Construction Engineering and
Management, 139(9), 1179-1189.
Zhao, X., Hwang, B. G., & Low, S. P. (2014). “Investigating enterprise risk management maturity
in construction firms.” Journal of Construction Engineering and Management, 140(8),
05014006.
Zhao, X., Hwang, B. G., Low, S. P., & Wu, P. (2015). “Reducing hindrances to enterprise risk
management implementation in construction firms.” Journal of Construction Engineering and
Management, 141(3), 04014083.
Zhou, Q., Fang, D., & Mohamed, S. (2011). “Safety climate improvement: case study in a Chinese
construction company.” Journal of Construction Engineering and Management, 137(1), 86-
95.