Table 4 Critical Incident Technique mapped against Realist Methodology for the purpose of the case study
From: Application of the critical incident technique in refining a realist initial programme theory
Establishing the general aim of the activity Introductory statement explaining the purpose of the study. Request for general aim What would you say is the primary purpose? Request for summary In a few words, how would you summarise the general aim of a specific activity | Establishing the general aim of the activity “We are making a study of multi-disciplinary team interventions to improve patient care in acute hospital contexts”. “The primary purpose is to help understand enablers and barriers to success of these interventions” In a few words, how would you summarise the general aim of the team intervention? What were the objectives? |
Purpose and specifications | Purpose and Specifications |
Situation Relevance to aim Persons to collect the data need to be familiar with activity. | What was the structure of the team Tell me about what happened and please be as detailed with the facts as possible? Keep in mind the relevance of the team intervention described by the key informant to building programme theory. How data being collected could relate and contribute to programme theory development UC is the primary researcher. Co-author – ADB who is familiar with the research question and purpose of the research was the second interviewer. |
Collecting the data | Collecting the data |
Specifications regarding observations | Specifications as follows: |
• Knowledge concerning the activity • Relation to those observed • Training requirements | • UC trained ADB with regard to the purpose of the critical incident technique and how to unpick the relevance of the team intervention that the key informant is describing to the purpose of the study. ADB is an experienced psychologist and qualitative researcher and is aware of the purpose of the research as co-author. UC drafted the interview and it was reviewed by ADB and EMcA. Both are already familiar with UCs foundational programme theory synthesised from the literature. • UC will complete two trial interviews and send the audio-files for review by ADB and EMcA |
Groups to be observed: | Key Informants to be interviewed |
Location Persons Times Conditions | Location Persons Times Conditions |
Behaviours to be observed | Detail to be extrapolated Information relating to contextual conditions and outcomes of the team intervention and how and why participants as individuals and as a collective in the team behaved the way they did in these circumstances. |
Rationale for asking for incidents to be recalled as opposed to direct observation- if suitable precautions are taken, recalled incidents can be relied upon to provide adequate data for a fairly satisfactory for a first approximation to a statement of requirements for the activity. Direct observations are to be preferred but the efficiency, immediacy and minimum demands on co-operating personnel which are achieved by using recalled incident data frequently make their use the more practical procedure. | Rationale Hospital workers are extremely busy and the idea of observing in live conditions over the prolonged period of a team intervention will not be practical … a lot of observation could be wasted time as there may be only a couple of critical incidents during a long period of time relating to the research question … this way the participants can be asked for detail of the intervention it relates to the research question and building of programme theory- |
Someone known and respected by the observer has suggested the interview | Purposeful sampling by CEO/ General Managers in 4 hospitals They selected candidates that they think will be able to contribute to the research question i.e., those who had some experience of team interventions either leading or being involved in the team intervention process. |
Questions should be trialled | Questions to be trialled. |
Interviewer remarks should be neutral and permissive and should demonstrate that he accepts the observer as the expert. Important to get unbiased events | Interviewer remarks should be neutral and permissive and should demonstrate that (s)he accepts the observer as the expert. Important to get unbiased incidents |
If only giving part of story he should be encouraged by restating the essence of his remarks. This will encourage and help him to bring out many of the details of the incident that the interviewer did not know details of the situation to ask for | During interviews UC and ADB will recall their understanding of what key informants said where necessary requesting clarification or expansion or a response in the form of more detail for example: ProbeWhat was the outcome for patient care in this event… ..The outcome for the team in this event… .. How did you react to this? How did you feel as a result? How did the team react to this? / how did the team feel as a result? |
Recorded electrically and transcribed | Recorded electronically, transcribed and imported into NViVo software |
Behaviour reports observed by the reporter Were all relevant factors in the situation given? Has the observer made a definite critical judgment about the relevance of the incident? Has the observer made it clear just why he believes the behaviour was critical? | Behaviour reports observed by the reporter Were all relevant factors in the situation given? Has the interviewer made a definite critical judgment about the relevance of the incident? Has the interviewer made it clear how and why he/she believes the contextual conditions generated the outcome and what mechanisms were enacted in doing so. |
Analysing the data and Interpreting and reporting the data. Imperative reporting is objective. | Analysing the data and Interpreting and reporting the data. RAMESES guidelines will be used- inductive, deductive and retroductive logic. Co-authors and realist support group will be consulted to help make judgment calls and challenge thinking. |