Confirmability in Qualitative Research

Qualitative research usually assume that every investigator provides a distinctive perspective to the study. Confirmability in Qualitative Research means the degree to which the outcomes could be confirmed or corroborated by other people.

The concept of confirmability is the qualitative investigator’s comparable concern to objectivity. There are various techniques for improving confirmability. The researcher can document the procedures for checking and rechecking the data during the entire research. Another researcher can take a “devil’s advocate” role with regards to the outcomes, and this process can be documented. The investigator can actively hunt for and describe any negative instances which contradict earlier findings. And, after the study, one can carry out a data audit which inspects the data collection and analysis procedures and makes judgements concerning the potential for bias or distortion.

Steps should be taken to help ensure as far as possible that the study outcomes are the result of the experiences and concepts of the informants, as opposed to the qualities and preferences of the researcher. A crucial criteria for confirmability is the degree to which the investigator confesses his or her own predispositions. Beliefs underpinning decisions made and techniques followed must be acknowledged in the study document, the explanations for favouring one particular method when others could have been used explained and weak points in the approaches actually employed admitted. When it comes to results, preliminary theories which finally were not borne out by the data must also be mentioned. Much of the content in regards to these areas may be derived from the ongoing “reflective commentary”.

An audit strategy is a key technique for establishing confirmability. This strategy includes an external auditor trying to follow through the natural history or progress of events in  a project in order to understand exactly how and why decisions were made. Furthermore, auditability’ implies that another investigator could come to equivalent findings given the same data and research context. The auditor considers the whole process of research along with the product, data, findings, interpretations, and recommendations. According to Lincoln and Guba, there are 6 groups of records which can be a part of the audit:

1. raw data (field records, audio and video sessions)
2. data reduction and analysis products (quantitative summaries, compacted notes, hypotheses)
3. data reconstruction  and  synthesis  products  (thematic  classifications, interpretations, inferences)
4. process notes (procedures and design techniques, trustworthiness information)
5. materials associated with intentions and dispositions (study proposal, field journal)
6. instrument development  information (pilot forms, survey format, schedules)

The audit needs to be regular during the entire research process; the limitation in this is that the auditor could possibly be co-opted into the project and therefore lose his or her objectivity. As per Guba a researcher must provide records for every claim or interpretation from a minimum of two sources to make sure that the data support the researcher’s analysis and interpretation of the results.

Various other techniques are helpful in the establishment of confirmability in qualitative research. Triangulation of multiple methods, data sources, and theoretical perspectives checks the potency of the researcher’s concepts.  A different way that can boost neutrality is to make use of a team of investigators experienced with qualitative methods as opposed to a single researcher. Reflexive analysis is also helpful to make sure that the investigator is aware of his or her influence on the data.