The use of Delphi and Nominal Group Technique in nursing education: A review

Abstract

ObjectivesConsensus methods are used by healthcare professionals and educators within nursing education because of their presumed capacity to extract the profession's’ “collective knowledge” which is often considered tacit knowledge that is difficult to verbalize and to formalize. Since their emergence, consensus methods have been criticized and their rigour has been questioned. Our study focuses on the use of consensus methods in nursing education and seeks to explore how extensively consensus methods are used, the types of consensus methods employed, the purpose of the research and how standardized the application of the methods is. Design and data sourcesA systematic approach was employed to identify articles reporting the use of consensus methods in nursing education. The search strategy included keyword search in five electronic databases [Medline (Ovid), Embase (Ovid), AMED (Ovid), ERIC (Ovid) and CINAHL (EBSCO)] for the period 2004–2014. We included articles published in English, French, German and Greek discussing the use of consensus methods in nursing education or in the context of identifying competencies. Review methodA standardized extraction form was developed using an iterative process with results from the search. General descriptors such as type of journal, nursing speciality, type of educational issue addressed, method used, geographic scope were recorded. Features reflecting methodology such as number, selection and composition of panel participants, number of rounds, response rates, definition of consensus, and feedback were recorded. Results1230 articles were screened resulting in 101 included studies. The Delphi was used in 88.2% of studies. Most were reported in nursing journals (63.4%). The most common purpose to use these methods was defining competencies, curriculum development and renewal, and assessment. Remarkably, both standardization and reporting of consensus methods was noted to be generally poor. Areas where the methodology appeared weak included: preparation of the initial questionnaire; the selection and description of participants; number of rounds and number of participants remaining after each round; formal feedback of group ratings; definitions of consensus and a priori definition of numbers of rounds; and modifications to the methodology. ConclusionsThe findings of this study are concerning if interpreted within the context of the structural critiques because our findings lend support to these critiques. If consensus methods should continue being used to inform best practices in nursing education, they must be rigorous in design.