BEHAVIORAL EVENT INTERVIEW: SOUND METHOD FOR IN-DEPTH INTERVIEWS

1Dr.Murillo de Oliveira Dias

Coordinator of the DBA Program at Fundação Getulio Vargas, Brazil
Corresponding Email [email protected] // [email protected]
2Roberto Aylmer

Rennes School of Business

Corresponding Email [email protected]

Abstract
The present article investigated the previous body of research on Behavioral Event Interview (BEI), based on early
Critical Incident Technique (CIT), providing a concise literature review on the subject, discussing its technique
application. This study is helpful to students, interviewers, academics, managers, and other professionals in general,
stimulating the reflection on interpretive and naturalistic approaches, as well as deepening the underlying process
related to in-depth interviews. It is a reliable tool for multiple and mixed methods usage. This article investigated the
following theoretical approaches: (a) Critical Incident Technique and (b) Behavioral Event Interview. This literature
review is part of my doctoral thesis and aims at providing consolidated information on the subject BEI for academics,
interviewers, and doctoral students in general, as well as presenting an example of its application. Discussion and
future research recommendations complete the present study.

KEYWORDS: Behavioral Event Interview; Critical Incident Technique

1. Introduction

Behavioral Event Interview (BEI) is a technique designed to collect primary data from qualitative in-depth
interviews, in turn an evolution of a previous method called the Critical Incident Technique (CIT), devised by
Flanagan (1954).

The Critical Incident Technique is an investigative approach, used for data collection and further analysis of
the human behavior related to the situation involved. The Critical Incident Technique was created in the 1940s
by the North American psychologist John Clemans Flanagan (1906-1996) and first published in 1954.
Flanagan was a pioneer in the armed air force’s psychology and a Harvard doctorate since 1934. He tested
pilots’ aptitude for the air combat missions in the World War II and provided a consistent approach that
involved some real case scenario experiences (Hughes, 2012). The Critical Incident Technique involved
critical incidents, turning points, crisis or significant instances on a given activity, in which its constituents
had participated. Similarities, differences, patterns can emerge from the Critical Incident Technique usage, to
discover the reason or how the person engaged in such activity (Hughes, 2012).

The Critical Incident Technique has been revised and adapted to different circumstances. CIT inspired later
Behavioral Event Interview, which conserves, a similar qualitative research structure (Delorme, 2007;
McLelland, 1961, 1973; Dailey, 1971; Spencer & Spencer, 1993). Therefore, the Behavioral Event Interview
is the primary data collection method investigated in this study.

CIT is used when the event is to be recalled. BEI is useful when the interviewees' behaviors are to be registered.
Both are important to the following aspects of an interview: (a) recency and (b) importance. In such cases,
BEI and CIT, when combined with the naturalistic approach, i.e., when the interviewees do not know the
subjects under research. BEI can be used in both semi-structured, structured, unstructured interviews (Dias,
2016; Myers & Newman, 2007). BEI has also proved to be an useful method to investigate workplace
socialization processes (Aylmer, 2018).

2. Qualitative In-depth Interviews

In-depth interviews are critical in qualitative research (Myers & Newman, 2007). They are supported by
Goffman's Dramaturgical theory (1959), that the whole interview process is seen as a drama. Goffman uses
theater analogies to investigate social interactions, defined as
The reciprocal influence of individuals upon one another’s actions when in one another’s immediate physical
presence’. Performance is defined as ‘all the activity of a given interviewee on a given occasion which serves to
influence in any way any of the other interviewees (Goffman, 1959, p. 26).

In the dramaturgical theory, like in a drama, there are: (a) actors (individuals or groups) who perform on a (b)
stage with all sort of settings and social situations arranged for the interview, and who perform on a (c) script
(structured, semi-structured or even unstructured). There are also (d) audiences, such as interviewer,
interviewee, facilitators, monitors, observers and the interviewees. (e) Entries, like the ways to start the
interview (for instance, with warm-up questions); (f) exists, the end of the interview, and (g) performances,
during the interview process. As mentioned in chapter two, interviews, however, are artificial situations,
according to Hermanns (2004), because they involve parlance between two strangers. The disadvantages are,
among others: (a) to rely on memory can bring imprecise statements to the research; (b) the interviewees may
feel embarrassed when remembering facts or due to the interviewer’s presence; (c) one incident may not be
very representative of the everyday situation (Hermanns, 2004). We followed Yin (2009) regarding in-depth
interviews:

You can ask key interviewees about the facts of a matter as well as their opinions about events. In some situations, you
may even ask the interviewees to propose her or his own insights into certain occurrences and may use such propositions
as the basis for further inquiry. The “interview” may therefore take place over an extended period of time, not just a single
sitting (p.107).

Myers & Newman (2007) were followed in adopting the semi-structured interview because it has provided
some freedom to make questions regarding the situational context. Guidelines for conducting the interviews
were adopted, as suggested by Myers & Newman (2007). It is essential to state the:

(a) the researcher as an actor, “because the interview is a social encounter and the data gathered from
interviews are idiographic, the interviewer should situate themselves as well as the interviewee” (Myers &
Newman, 2007, p.16).

(b) The social dissonance was minimized through preparing the interview to let the interviewee as comfortable
as possible, generally managing “first impressions, dressing appropriately, and using the suitable
language/jargon” (Myers & Newman, 2007, p.16).

(c) The interviewees were invited to different rank positions to avoid elite bias.

(d) everyone is considered as an interpreter.

(e) questions mirrored answers.

(f) The availability of interviewees was taken into consideration, and

(g) the strict confidentiality was assured to all the interviewees.

Qualitative interview, despite its investigative value, is not perfect. There are some cognitive biases and pitfalls
to be overcome, according to Myers & Newman (2007); Marlowe (1960); Miles, Huberman, & Saldaña
(2014).
Strategies were devised to attenuate some cognitive biases, as follows: (a) the Elite bias. Following Myers &
Newman (2007) interviewing only high-position employees was avoided. The sampling criteria included the
top, high, middle and low-level employees; (b) the Hawthorne effect, or a phenomenon where persons change
their behavior in response to a fact, regarding the workplace, instead of responding according to their nature
(Myers & Newman, 2007). A naturalistic approach was employed, in which interviewees did not have any
information regarding the purposes of the research. (c) The Social Desirability Bias (SDB) that the
interviewees may answer questions in a way morally or socially acceptable by others (Marlowe, 1960).
In this case, the ways to avoid Social Desirability Bias were the following: (c.1) keeping the confidentiality
through formal commitment and signature of informed consent forms; (c.2) considering data only after
interviewee’s final approval; (c.3) allowing interview interruption at any time. Finally, we built (c.4) a healthy
environment, in which the interviewee felt comfortable. Last, we tried to avoid (d) the ambiguity of language,
recognizing that in extreme cases, the “dramaturgical model could lead to unethical behavior” (Myers &
Newman, 2007, p.23).

Jargons supposed to be avoided; should be restated, rephrased through posing new questions, as many times
as necessary, to make them clear to the interviewees.

Lewicki, Tomlinson & Gillespie, (2006) also posited that longitudinal qualitative studies on trust should be
investigated, such as (a) qualitative in-depth interviews; (b) the Critical Incident Technique (consequently the
Behavioral Event Interview), (c) case studies and communication analysis. In sum, Lewicki, Tomlinson &
Gillespie (2006) pointed out qualitative research as paramount:

Therefore, we encourage the researchers to consider employing complementary methods. Promising qualitative
methods that are particularly suitable for tracking relationships over time and assessing the dynamics and
development of trust (…). Such qualitative methods have high external validity and allow insight into the way that
trust is socially and subjectively constructed (Lewicki, Tomlinson & Gillespie, 2006, p. 1015).

Nevertheless, the disadvantages of the methods used to reduce biases are: (a) the interviewer may forget some
details, even in good faith. (b) The interviewee may omit some details because of personal convenience. (c)
The interviewer may describe the facts as they should happen not the way they occurred. (c) In the worst case
scenario, the interviewee may deliberately lie about the events described (Miles, Huberman & Saldaña, 2014;
Myers & Newman, 2007).

In the Behavioral Event Interview, it is up to the researcher to capture essential details about the investigation
conducted through cases narrated.

3.Behavioral Event Interview (BEI): background

The Behavioral Event Interview is a technique developed by David McLelland (1917-1998), Psychologist at
Harvard University and also the creator of Need Theory and Thematic Apperception Test (McLelland, 1973,
1976). The Behavioral Event Interview is a technique rooted in Flanagan’s Critical Incident Technique
(Flanagan, 1954), which is focused on the incident or event, whereas Behavioral Event Interview is concerned
with the person who did it. The Behavioral Event Interview was designed to identify the competencies needed
to perform a job in a proper manner (McLelland, 1973, 1976; Spencer & Spencer, 1993).

Some activities, particularly the behaviors, are noticeable. For example, two people negotiating at the
negotiation table: both are talkative, speak aloud, both use their arms to reinforce non-verbal language and so
on. This behavior (extraversion) is easily observable, therefore, denominated the overt behavior. Some
behaviors are not observable because they occur within the individual. As an illustration, the impressions and
the feelings of the two persons mentioned earlier. The covert behaviors are non-observable behaviors (Kaplan
& Saccuzzo, 2009). Therefore, in this research, we focused on overt or observable behaviors.
The Behavioral Event Interview compared two different perspectives of data, positive and negative outcomes
per interview. Therefore, Contrast Analysis is a strategy that fits with this research.

According to Schreier (2012), the strategy consists of three steps: (a) to identify patterns within the positive
outcomes; (b) to determine patterns within the negative outcomes, and (c) to determine differences between
positive and negative outcomes (p.124). The positive and negative outcomes were compared separately and
then compared both in three moments: before, during and after the negotiation (event).
Throughout time, BEI has been used for assessments related to successful entrepreneurs (McClelland, 1987),
on BEI quantitative analysis (Buchhorn, 1991), identification of core business competencies (McClelland,
1998), determination of competency models for primary and Secondary School Teaching (Jianping, &
Houcan, 2006), assessments of in-company socialization process (Aylmer, 2018), as a complementary method
for case studies representation (Ledford & Gast, 2018), to investigate dispositional factors, such as trust and
social value orientation (Dias, 2016), even to assess overall practitioner performance (Rothwell, Hohne &
King, 2018).

4. BEI combined with Constrast analysis

BEI is a helpful tool when combined with Contrast analysisMiles, Huberman & Saldaña, 2014), which is a
strategy used to analyze data-based or the content-based codes when comparing two different sources,
recommended when the researcher has already an idea of which main codes will be coded (Schreier, 2012).

The Behavioral Event Interview compared two different perspectives of data, positive and negative outcomes
per interview. Therefore, Contrast Analysis is a strategy that fits with the qualitative research objectives.

According to Schreier (2012), the strategy consists of three steps: (a) to identify patterns within the positive
outcomes; (b) to determine patterns within the negative outcomes, and (c) to determine differences between
positive and negative outcomes (p.124). The positive and negative outcomes were compared separately and
then compared both in three moments: before, during and after the negotiation (event).

Since in BEI, the interviewees tell two different stories, one with positive outcomes and other with negative
outcomes, following naturalistic approach (i.e., the interviewee does not have a clue on the research subject
under investigation), it is useful to gather the results of such interviews in a graphic design form (see Table
1), helpful for further analysis. Table 1 presents an example of contrast analysis report:

Table 1

Contrast analysis: positive vs. negative outcomes

Positive Outcomes Negative Outcomes

Code type Code Before During After Before During After
Event Event Event Event Event Event
TR-COBT -- - - - - --
TR-DTR -- - -- 0 + ++
TR-DTBT -- - -- -- -- --
TRUST: TR
TR-CABT - -- -- - -- --
TR-KNBT - - 0 - -- --
TR-IDBT -- -- - -- -- --
SVO-PSELF 0 -- -- + ++ ++
SVO
SVO-PSOCIAL + ++ ++ 0 0 0
LEGEND: -- = lowest - =low 0= intermediate + = high ++ = highest

Source: Dias, 2016. Reprinted under permision

5. Discussion

BEI was an evolution of CIT throughout time and is more suitable for interviews when the behavior of the
interviewees are more important than the fact, or event itself.
More recent studies (Aylmer, 2018; Rothwell, Hohne & King, 2018; Ledford & Gast, 2018, Dias, 2016)
indicate that BEI should become more efficient, if combined with other methods, such as toy models (see
Figure 1), and the constructivist approach (Aylmer, 2018), case studies (Ledford & Gast, 2018), naturalistic
approach (Dias, 2016), or to assess practitioner performance (Rothwell, Hohne & King, 2018).

LEGO Serious Play (LSP) for instance, is suitable for this purpose since the interviewees use a constructivist
approach to represent their ideas through sets of information, depicted in Figure 1, to relate positive and
negative experiences in the socialization process within a state-owned Brazilian company (Aylmer, 2018) as
follows:

Figure 1 BEI using Toy model with Negative (left) and Positive (right) socialization experiences. Source: Aylmer, 2018.

The advantages of the usage of Behavioral Event Interviews (BEI) are: (a) empirical identification of
competencies beyond or different from those generated by the other data collection methods. (b) Precision
about competencies; (c) identification of algorithms; (d) freedom from racial, gender, and cultural bias; (e)
generation of data for assessment, training, and career pathing (Spencer & Spencer, 1993, pp. 98-99). The
disadvantages of the Behavioral Event Interview method are (f) time and expense; (g) expertise requirements
(h) missed job tasks; (i) impractical for analysis of many jobs (Spencer & Spencer, 1993, pp. 98-99).

Therefore, the Behavioral Event Interview technique was suitable for any qualitative research, because it is
concerned with the behavior and the person than the fact and the critical incident itself since we are
investigating a social interaction process.

Finally, the present research consolidated complementary but separate past body of knowledge, providing a
concise and broad vision on BEI, here under investigation.

6. Limitations and Recommendations for future research

In qualitative research, reliability is concerned with sound alternatives that researchers would disclose when
dealing with similar data (Benson & Brown, 2011). The concern about reliability in these kinds of studies
regards the issues of bias.

Despite the usefulness of BEI for qualitative, in-depth interviews, research methods are not infallible (Dias,
2016). Some limitations should be taken into consideration, not restricted to: (i) elite bias (Myers & Newman,
2007), which occurs when the sample contains only top-level managers. The interviews may be biased by
opinions that represent only one company segment, disregarding others. The antidote is to interview
respondents from different sectors within a company, such as low, medium, high, and top-level managers, to
provide a broader view of the subject under investigation.

A second limiting factor is the (ii) Hawthorne effect or a phenomenon where persons change their behavior in
response to a fact, regarding the workplace, instead of responding according to their nature (Myers &
Newman, 2007). To avoid this limitation, we recommend the usage of the naturalistic approach, in which
interviewees ddo not have any information regarding the purposes of the research.

A third limiting factor is the (iii) social desirability bias (SDB) that the interviewees may answer questions in
a way morally or socially acceptable by others (Marlowe, 1960). To avoid SDB, we recommend the following
actions: (iii.1) keeping the confidentiality through formal commitment and signature of informed consent
forms; (iii.2) considering data only after interviewee’s final approval; (iii.3) allowing interview interruption
at any time. It is also essential to foster (iii.4) a healthy environment, in which the interviewees feel
comfortable. Then, avoid (iii.5) the ambiguity of language, recognizing that in extreme cases, the
“dramaturgical model could lead to unethical behavior” (Myers & Newman, 2007, p.23).

Then, structured interviews (Myers & Newman, 2007) limit the questions and the scope of the answers,
making the interview more rigid and less fluid. A more practical recommendation is to use semi-structured
interviews because it provides the possibility of insertion of new and relevant questions that eventually appear
in the middle of the interview.

Also, the environment could be a strong challenging and therefore, limiting factor. It is recommended to avoid
noisy places. Noise is a reliable source of distraction and failures in audio recordings.

Finally, toy models (Aylmer, 2018) are limited to the number of toys provided, local infrastructure and
availability of proper instruments. For instance, the research would be biased if only toys representing males
were employed. Another example is related to characters: in a socialization process, for instance, regarding
the relationship newcomer-boss, the interview would be biased if the only demon, monster toys were
employed to represent this relationship. The recommendation is to provide the broadest type of different action
figures, to avoid any further biases (Aylmer, 2018). When using toy models, if an action figure of a good cop
is used, a bad cop should be available too, in the same proportion, for the interviewees to build their
experiences.

Future research is encouraged to investigate the BEIs applied to other scenarios, such as workplace
socialization process, in which the experiences, behaviors of the interviewees are essential for the research
than the facts itself. It is also recommended to study BEI combined with other methods and to asses the impact
of them in different scenarios, such as the aforementioned.

REFERENCES

Aylmer, R. (2018) The Socialization Process in a Brazilian Public Company. (unpublished doctoral thesis).
ESC Rennes School of Business, France.
Benson, J., & Brown, M. (2011). Generations at work: are there differences and do they matter?. The
International Journal of Human Resource Management, 22(9), 1843-1865.
Buchhorn, D. (1991). Behavioral Event Interview Quantitative Results. L'Oreal Corporation, New York.
Delorme, M. (2007). Les compétences stratégiques des dirigeants: application au cas des entreprises
canadiennes de biotechnologie. Thèse de doctorat, Université Robert Schuman (Strasbourg 3).
Dailey, C. (1971). Assessment of lives. San Francisco, CA: Jossey-Bass.
Dias, Murillo de Oliveira (2016). Factors Influencing the Success of Business Negotiations in the Brazilian
Culture (Doctoral Thesis). ESC Rennes School of Business, France.
Lewicki, R.J., and Litterer, J.A. (1985). Negotiation. Homewood, IL: R.D. Irwin.
Flanagan, J.C. (1954) 'The Critical Incident Technique', Psychological Bulletin 51(4): 327-358.
Goffman, E (1959). The Presentation of Self in Everyday Life. New York: Doubleday.
Goffman, E. (1961). Encounters: Two Studies in the sociology of interaction. Indianapolis: The Bobbs-Merrill
Company.
Hughes, H.; (2012). An expanded critical incident approach for exploring information use and learning.
Library and Information Research, 36 (112), 72-95.
Jianping, X., & Houcan, Z. (2006). Competency Model of Primary and Secondary School Teachers: A Study
of Behavioral Event Interviews [J]. Educational Research, 1, 011.
Kaplan, Robert M., Saccuzzo, Dennis P. (2009) Psychological Testing: principles, applications, and issues.
Wadsworth, CA: Wadsworth Cengage Learning, 7th edition.
Kassin, S. (2003). Psychology. Upper Saddle River, NJ: Prentice-Hall, Inc.
Ledford, J. R., & Gast, D. L. (2018). Single case research methodology: Applications in special education and
behavioral sciences. Routledge.
Lewicki, R.J., and Bunker, B.B. (1995). Trust in relationships: A model of trust development and decline. In
B.B. Bunker and J.Z. Rubin (eds), Conflict, cooperation, and justice (pp. 133–174). San Francisco:
Jossey- Bass.
Lewicki, R.J., and Bunker, B.B. (1996). Developing and maintaining trust in work relationships. Trust in
Organizations: Frontiers of Theory and Research (pp. 114–139). Thousand Oaks, CA: Sage
Publications.
Lewicki, R.J., McAllister, D.J. & Bies, R.J. (1998). Trust and distrust: New relationships and realities.
Academy of Management Review, 23(3), 438-458.
Lewicki, R.J., and Stevenson, M. (1998). Trust development in negotiation: Proposed actions and a research
agenda. Journal of Business and Professional Ethics, 99-132.
Lewicki, R.J., Tomlinson, E.C., and Gillespie, N. (2006). Models of Interpersonal Trust Development:
Theoretical Approaches, Empirical Evidence, and Future Directions. In: Journal of Management, 32,
991-1022.
Lewicki, R.J., Barry, B., and Saunders, D. (2010). Negotiation (6th edition). Burr Ridge, IL: McGraw- Hill
Irwin.
Lewicki, R.J. & Brinsfield, C. (2012). Trust as a heuristic. In: W.A. Donohue, R.R. Rogan and S.Kaufman
(Eds.). Framing in Negotiation: State of the Art. New York, Peter Lang.
Lewicki, R.J. & Hanke, R. (2012). Once fooled, shame on you! Twice fooled, shame on me! What deception
does to deceivers and victims: Implications for negotiators in situations where ethicality is unclear. In
B. Goldman and D. Shapiro (eds), The Psychology of Negotiations for the 21st Century (pp. 211-244).
Oxford, UK: Routledge Press.
Lewicki, R.J., and Polin, B. (2013). Trust and negotiation. In: Olekalns, M, and Adair, W.L. (eds.) Handbook
of Research in Negotiation. UK: Edward Elgar Publishing Limited, pp. 161-190.
Marlowe, D. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting
Psychology, 24, 349-354.
McLelland, D.C. (1961). The Achieving Society. Princeton, NJ: Van Nostrand.
McLelland, D.C. (1973). Testing for Competence rather that ‘Intelligence’. American Psychologist, 28, 1-14.
McLelland, D.C. (1976). A guide to job competency assessment. Boston, MA: McBer.
McClelland, D. C. (1987). Characteristics of successful entrepreneurs. The journal of creative behavior, 21(3),
219-233.
McClelland, D. C. (1998). Identifying competencies with behavioral-event interviews. Psychological science,
9(5), 331-339.
Miles, M. & Huberman, A. (1994). Qualitative Data Analysis. 2nd edition. London: Sage.
Miles, M.; Huberman, A. and Saldaña, J. (2014). Qualitative Data Analysis: a Methods Sourcebook. 3rd
edition. London: Sage.
Myers, M. D. and Newman, M. (2007). The Qualitative Interview in IS Research: Examining the Craft.
Information and Organization, 17, 2-26.
Rothwell, W. J., Hohne, C. K., & King, S. B. (2018). Human performance improvement: Building practitioner
performance. Routledge.
Schreier, M. (2012). Qualitative content analysis in practice. London: Sage.
Spencer, L.M. Jr. & Spencer, S.M. (1993), Competence at Work: Models for Superior Performance, John
Wiley, and Sons, Inc.
Yin, R. K. (2009). Case Study Research: Design and Methods. Thousand Oaks, CA: The SAGE Publications,
4th. Edition.