1. Exploring the measurement properties of the osteopathy clinical teaching questionnaire using Rasch analysis.
- Author
-
Vaughan, Brett
- Subjects
ALLIED health personnel ,CHI-squared test ,STATISTICAL correlation ,FACTOR analysis ,OSTEOPATHIC medicine ,PROBABILITY theory ,QUESTIONNAIRES ,RESEARCH funding ,STATISTICS ,STUDENTS ,T-test (Statistics) ,CLINICAL competence ,DATA analysis ,HEALTH occupations school faculty ,MULTITRAIT multimethod techniques ,RESEARCH methodology evaluation ,DATA analysis software ,DESCRIPTIVE statistics ,EDUCATION - Abstract
Background: Clinical teaching evaluations are common in health profession education programs to ensure students are receiving a quality clinical education experience. Questionnaires students use to evaluate their clinical teachers have been developed in professions such as medicine and nursing. The development of a questionnaire that is specifically for the osteopathy on-campus, student-led clinic environment is warranted. Previous work developed the 30-item Osteopathy Clinical Teaching Questionnaire. The current study utilised Rasch analysis to investigate the construct validity of the Osteopathy Clinical Teaching Questionnaire and provide evidence for the validity argument through fit to the Rasch model. Methods: Senior osteopathy students at four institutions in Australia, New Zealand and the United Kingdom rated their clinical teachers using the Osteopathy Clinical Teaching Questionnaire. Three hundred and ninety-nine valid responses were received and the data were evaluated for fit to the Rasch model. Reliability estimations (Cronbach's alpha and McDonald's omega) were also evaluated for the final model. Results: The initial analysis demonstrated the data did not fit the Rasch model. Accordingly, modifications to the questionnaire were made including removing items, removing person responses, and rescoring one item. The final model contained 12 items and fit to the Rasch model was adequate. Support for unidimensionality was demonstrated through both the Principal Components Analysis/t-test, and the Cronbach's alpha and McDonald's omega reliability estimates. Analysis of the questionnaire using McDonald's omega hierarchical supported a general factor (quality of clinical teaching in osteopathy). Conclusion: The evidence for unidimensionality and the presence of a general factor support the calculation of a total score for the questionnaire as a sufficient statistic. Further work is now required to investigate the reliability of the 12-item Osteopathy Clinical Teaching Questionnaire to provide evidence for the validity argument. [ABSTRACT FROM AUTHOR]
- Published
- 2018
- Full Text
- View/download PDF