The SOMERS Index: A simple instrument designed to predict the likelihood of rural career choice
Article first published online: 25 MAR 2011
© 2011 The Authors. Australian Journal of Rural Health © National Rural Health Alliance Inc.
Australian Journal of Rural Health
Volume 19, Issue 2, pages 75–80, April 2011
How to Cite
Somers, G. T., Jolly, B. and Strasser, R. P. (2011), The SOMERS Index: A simple instrument designed to predict the likelihood of rural career choice. Australian Journal of Rural Health, 19: 75–80. doi: 10.1111/j.1440-1584.2011.01187.x
- Issue published online: 25 MAR 2011
- Article first published online: 25 MAR 2011
- Accepted for publication 19 December 2010.
- medical education;
- program evaluation;
- recruitment and retention;
- rural health services delivery
Objective: The World Health Organization has drawn up a set of strategies to encourage health workers to live and work in remote and rural areas. A comprehensive instrument designed to evaluate the effectiveness of such programs has not yet been tested. Factors such as Stated rural intention, Optional rural training, Medical sub-specialization, Ease (or self-efficacy) and Rural Status have been used individually or in limited combinations. This paper examines the development, validity, structure and reliability of the easily-administered SOMERS Index.
Design: Limited literature review and cross-sectional cohort study.
Setting: Australian medical school.
Participants: A total of 345 Australian undergraduate-entry medical students in years 1 to 4 of the 5-year course.
Main outcome measures: Validity of the factors as predictors of rural career choice was sought in the international literature. Structure of the index was investigated through Principal Components Analysis and regression modelling. Cronbach's alpha was the test for reliability.
Results: The international literature strongly supported the validity of the components of the index. Factor analysis revealed a single, strong factor (eigenvalue: 2.78) explaining 56% of the variance. Multiple regression modelling revealed that each of the other variables contributed independently and strongly to Stated Rural Intent (semi-partial correlation coefficients range: 0.20–0.25). Cronbach's alpha was high at 0.78.
Conclusions: This paper presents the reliability and validity of an index, which seeks to estimate the likelihood of rural career choice. The index might be useful in student selection, the allocation of rural undergraduate and postgraduate resources and the evaluation of programs designed to increase rural career choice.