Premium
Swedish Translation, Adaptation and Psychometric Evaluation of the Context Assessment Index (CAI)
Author(s) -
Nilsson Kajermo Kerstin,
Böe Heidi,
Johansson Eva,
Henriksen Eva,
McCormack Brendan,
Gustavsson J. Petter,
Wallin Lars
Publication year - 2013
Publication title -
worldviews on evidence‐based nursing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.052
H-Index - 49
eISSN - 1741-6787
pISSN - 1545-102X
DOI - 10.1111/j.1741-6787.2012.00252.x
Subject(s) - cronbach's alpha , context (archaeology) , health care , think aloud protocol , applied psychology , psychology , ceiling effect , psychometrics , scale (ratio) , medical education , computer science , medicine , clinical psychology , usability , paleontology , alternative medicine , pathology , human–computer interaction , economics , physics , quantum mechanics , biology , economic growth
Background The strength of and relationship between the fundamental elements context, evidence and facilitation of the PARIHS framework are proposed to be key for successful implementation of evidence into healthcare practice. A better understanding of the presence and strength of contextual factors is assumed to enhance the opportunities of adequately developing an implementation strategy for a specific setting. A tool for assessing context—The Context Assessment Index (CAI)—was developed and published 2009. A Swedish version of the instrument was developed and evaluated among registered nurses. This work forms the focus of this paper. Purpose The purpose of this study was to translate the CAI into Swedish, adapt the instrument for use in Swedish healthcare practice and assess its psychometric properties. Methods The instrument was translated and back‐translated to English. The feasibility of items and response scales were evaluated through think aloud interviews with clinically active nurses. Psychometric properties were evaluated in a sample of registered nurses ( n = 373) working in a variety of healthcare organisations in the Stockholm area. Item and factor analyses and Cronbach's alpha were computed to evaluate internal structure and internal consistency. Result Sixteen items were modified based on the think aloud interviews and to adapt the instrument for use in acute care. A ceiling effect was observed for many items and the originally identified 37 item five‐factor model was not confirmed. Item analyses showed an overlap between factors and indicated a one‐dimensional scale. Discussion The Swedish version of the CAI has a wider application than the original instrument. This might have contributed to the differences in factor structure. Different opportunities for further development of the scale are discussed. Conclusions Further evaluation of the psychometric properties of the CAI is required.