OCTOBER 2001, VOL 74, NO 4

R E S E A R C H C O R N E R

Critical incident technique as a qualitative research method

ritical incident technique (CIT) describes a specific c methodology and method

used in nursing research. As with all types of qualitative research, it is important to link the method with a specific methodology and its philosophical perspective. This article explores the origins of CIT, including philosophical assumptions from the time period when it was first described, the technique itself, and its current applicability.

ORIGINS In the 1950s, many scientific

researchers focused their studies on human behavior. Behaviorists sought to analyze and synthesize observations into a set of relation- ships that could be quantified, vali- dated, and generalized independent of conscious experience. Benjamin Bloom, an educator, wrote the Tmonomy of Educational Objectives in 1956, which describes cognitive categories and behavioral terms representative of learning processes. I Subsequently, many researchers in the late 1950s focused their studies on quantify- ing human behaviors in a mean- ingful, structured way.

Researchers of this t i e uni- versally revered the scientific method, and psychology served as the framework for observing and categorizing human behavior. Concurrently, rapid industrial growth influenced the science of authenticating and documenting professional roles and work requirements. For example, many

studies conducted during this time identified specific role behaviors among pilots, dentists, psychology faculty members, bookkeepers, and even nurses. The purpose of this type of research was to facilitate proce- dures for selecting, training, clas- sifying, and evaluating perform- ance standards for specific occu- pational groups. This emphasis on authenticating and document- ing professional roles supported the field of psychology known as organizational behavior. In this context, CIT emerged from assumptions that the scientific method could facilitate the obser- vation and categorization of all human behaviors.

CRITICAL INCIDENT TECHNIQUE

John C. Flanagan was first pub- lished in the Psychological Bulletin in 1954.* Flanagan was a psychologist who worked at the American Institute for Research and the University of Pittsburgh. He described CIT as a set of pro- cedures used to collect observa- tions of human behavior. These observations, in turn, are used to solve practical problems and develop psychological principles.

Flanagan identified five steps involved in CIT. The first is deter- mining the general aim of the study (ie, a brief statement of the topic under study). For example, one aim of a hypothetical study could be to determine perioperative circulating nurse job duties. According to Flanagan, this step is essential, as

The CIT approach ascribed to

planning and evaluation are impos- sible without fmt properly identi- fying the general aim.

The second step of CIT is to plan and specify how factual inci- dents regarding the general aim of the study will be collected. In the hypothetical example, to under- stand circulating nurse duties, the researcher would have to deter- mine how to identify factual inci- dents that occurred between the circulating nurse and members of the health care team, as well as patients. This determination would lead to the development of criteria used to decide which interactions to observe and record.

The third step is to collect the data. Data collection may occur via interview or may consist of an observer writing reports. In the hypothetical example, the researcher might interview circu- lating nurses or observe them dur- ing subsequent days of surgery. According to Flanagan, data should be objective to validate behavior. He defines objectivity as “the tendency for a number of independent observers to make the same report.”)

The fourth step in CIT is to analyze the data. The analysis should be summarized and the data described efficiently so they can be used for practical purposes. In the example, data analysis could occur via textual analysis or identification of themes. Flanagan did acknowledge, however, that this fourth step is less objective than the others.

The fifth and final step is to

536 AORN JOURNAL

OCTOBER 2001, VOL 74, NO 4

interpret and report the require- ments of the activity being studied. In the hypothetical study, this could include a list of circulating nurse job duties that emerged from reporting critical incidents described by circulating nurses and by observing them during surgery.

A common attribute of CIT is that it elicits aspects of best and worst practices. For example, in the hypothetical study, the researcher might ask participants to discuss incidents when they believed their nursing practice was of high quality. The researcher also might ask partici- pants to discuss incidents when they believed their nursing prac- tice was ineffective, problematic, or resulted in negative outcomes.

One group of researchers iden- tified three assumptions common- ly associated with CIT.4 First, the term critical incident refers to a clearly demarcated scene. Second, if a detailed account of what actu- ally happened cannot be obtained, that incident is not valid. Third, the critical incident itself is the basic unit of analysis.

CURRENT APPLlCABlLlN OF CRITICAL INCIDENT TECHNIQUE

In 198 1, one group of researchers conducted a study evaluating the clinical practicum setting using CIT? The purpose of this study was to identify the fac- tors, conditions, and opportunities that significantly influence nursing students’ learning experience. Researchers recorded 332 critical incidents, and more than 75% were positive in that they facilitat- ed learning in the clinical situation. Researchers believe CIT was use- ful as it provided a broad picture of clinical activities, and the criti- cal incidents helped faculty mem- bers evaluate their teaching styles

and techniques. Applicability of Flanagan’s model, however, has been limited by changes in society and modem work environments. Although he is considered the father of CIT, not all researchers cite Flanagan or subscribe to his

A common attribute of critical incident technique is that it elicits aspects of best and worst

assumptions in their application of his technique:

In the early 199Os, one group of researchers found that their par- ticipants’ descriptions and embed- ded meanings emerged from a combination of experiences rather than from a single or clearly demarcated incident. They pro- vide a thorough critique of Flanagan and how his assump- tions of objectivity were problem- atic in their nursing study.’ These researchers suggest that Flanagan’s technique is too rigid and narrowly prescribed to cap- ture the complexity of human behavior. They state that incidents do not always measure up to Flanagan’s criteria, yet they are valid incidents. According to Flanagan, these incidents should not be included in a study. Yet many researchers include these types of incidents without disclos- ing the fact that they do not meet Flanagan’s criteria.

One confusion identified in

the literature is the association of CIT to phenomenology. This confusion ofien occurs because people do not understand Flanagan’s definition of critical incident, which is an incident in which the purpose or intent of the act is clear to the observer and the consequences are defi- nite.* Researchers are encouraged to compare and contrast Flanagan’s assumptions with those of phenomenology.’ One researcher who uses the phenom- enological method has been asso- ciated mistakenly with CIT.’O Although this researcher denies that her phenomenological meth- ods are based on CIT, the confu- sion may have arisen from her early work in which the data col- lection tool she used asked par- ticipants to describe a critical incident.“ She no longer uses this terminology.

Researchers, should realize that there is no single universal definition or interpretation of CIT. They must carefully validate their assumptions regarding CIT and are encouraged to revisit Flanagan’s original work before undertaking research using the CIT approach.

SUMMARY Critical incident technique is a

helphl methodology used to examine a number of issues perti- nent to nursing science. It allows nurses to understand the dimen- sions of their role in clinical set- tings and their interactions with patients and other clinicians. It also helps nurses understand nurs- ing practice in a variety of roles (eg, clinical educator, nurse infor- matician, faculty member). Researchers using this methodolo- gy should review Flanagan’s orig- inal assumptions and compare

538 AORN JOURNAL

OCTOBER 2001, VOL 74, NO 4

them to their proposed methodol- ogy to determine whether CIT is the best fit. Additionally, with the research question. and analysis. researchers should compare this Critical incident technique can MICHELLE BYRNE method to phenomenology, provide insight into nursing issues RN, MS, PnD, CNOR grounded theory, or hermeneutics when used with rigor and when NURSING RESEARCH COMMITTEE

to determine whether these approaches are more congruent

appropriate rationale is provided for data collection, interpretation,

NOTES

The Ciassif;cation of Educational Goals. Handbook I: Cognitive Domain (New York: McKay, 1956).

2. J C Flanagan, “The critical incident technique,” Psychological Bulletin 51 (1954) 327-358.

3. Ibid, 335. 4. I J Norman et al, “Developing Flanagan’s critical

1. B S Bloom, Taonomy ofEducationa1 Objectives:

incident technique to elicit indicators of high and low quality nursing care from patients and their nurses,’’ Journal ofAdvanced Nursing 17 (May 1992) 590-600.

applied to the evaluation of the clinical practicum setting,” Journal of Nursing Education 20 (October 1981) 15-3 1.

6. L Rosenthal, “Exploring the learner’s world: Critical incident methodology,” The Journal of Continuing Education in Nursing 26 (May/June 1995) I 15-1 18; V Wilde, “Controversial hypotheses on the rela- tionship between researcher and informant in qualitative

5. C Z Dachelet et al, “The critical incident technique

research,” Journal of Advanced Nursing 17 (February 1992) 234-242.

7. Norman et al, “Developing Flanagan’s critical inci- dent technique to elicit indicators of high and low quality nursing care from patients and their nurses,” 590-600.

8. Flanagan, “The critical incident technique,” 327. 9. M Byme, “Understanding life experiences through

a phenomenological approach to research,” AORN Journal 73 (April 2001) 830-832.

the components of best nursing practice in the care of acutely ill hospitalized older patients with coincidental dementia: A muIti-method design,” Journal ofAdvanced Nursing 30 (November 1999) 1127-1136.

11. P Benner, personal communication with the author, San Francisco, 12 June 2001; P E Benner, From Novice to Expert: Excellence and Power in Clinical Nursing Practice (Menlo Park, Calif: Addison-Wesley Publishing

10. D Tolson, M Smith, P Knight, “An investigation of

CO, 1984) 300-302.

Silicon Chip May Be Used To House Normal Liver Cells Researchers may have discovered a way to keep liver cells alive outside the human body, according to an April 3,200 1, news release from the University of California, San Diego. This would be done by placing cells on silicon chips with miniature wells, similar to those in muffing tins, that allow the maintenance of fully functioning cells. This achieve- ment could lead to new methods for treating liver disease and testing medication toxicity, according to the release.

Although cancerous liver cells easily can be grown in culture dishes, growing healthy liver cells is not so easy. The porous silicon “liver bioreactor,” however, mimics liver conditions. It contains indi- vidual cells in well-like structures that can be modi- fied chemically to mimic the extracellular matrix distribution found in the liver. The fact that silicon is porous allows it to promote the flow of nutrients and chemicals through the cell culture and filter larger particles (eg, bacteria, viruses). Researchers are

striving to create the conditions necessary for numerous cells on the chip to behave collectively, according to the release.

The ultimate goal of researchers is to use the liver bioreactor for development of an artificial liver. Currently, five companies have artificial livers in clinical trials. These are external devices that are intended for patients with end-stage liver disease and that house pig liver cells or cancerous cell lines that act as a bridge to keep patients alive until a donor liver is available. Maintaining live, function- ing cells, however, has been a challenge with these devices. The silicon chips may help researchers dis- cover new techniques for maintaining these cells.

UCSD Researchers Create a Home for Liver Cells on a Silicon Chip (news release, Son Diego: Universiv of Colifomia, Son Diego, April 3, 200 1) http:/ucsdnew . ucsd. eddnewreYsciencdmcbioreactor. htm (accessed 19 April 200 I ) .

539 AORN JOURNAL