The De-Escalating Aggressive Behaviour Scale: development and psychometric testing
Article first published online: 7 AUG 2009
© 2009 The Authors. Journal compilation © 2009 Blackwell Publishing Ltd
Journal of Advanced Nursing
Volume 65, Issue 9, pages 1956–1964, September 2009
How to Cite
Nau, J., Halfens, R., Needham, I. and Dassen, T. (2009), The De-Escalating Aggressive Behaviour Scale: development and psychometric testing. Journal of Advanced Nursing, 65: 1956–1964. doi: 10.1111/j.1365-2648.2009.05087.x
- Issue published online: 7 AUG 2009
- Article first published online: 7 AUG 2009
- Accepted for publication 22 May 2009
- De-Escalating Aggressive Behaviour Scale;
- instrument development;
- nurse education;
- psychometric testing
Title. The De-Escalating Aggressive Behaviour Scale: development and psychometric testing.
Aim. This paper is a report of a study to develop and test the psychometric properties of a scale measuring nursing students’ performance in de-escalation of aggressive behaviour.
Background. Successful training should lead not merely to more knowledge and amended attitudes but also to improved performance. However, the quality of de-escalation performance is difficult to assess.
Method. Based on a qualitative investigation, seven topics pertaining to de-escalating behaviour were identified and the wording of items tested. The properties of the items and the scale were investigated quantitatively. A total of 1748 performance evaluations by students (rater group 1) from a skills laboratory were used to check distribution and conduct a factor analysis. Likewise, 456 completed evaluations by de-escalation experts (rater group 2) of videotaped performances at pre- and posttest were used to investigate internal consistency, interrater reliability, test–retest reliability, effect size and factor structure. Data were collected in 2007–2008 in German.
Findings. Factor analysis showed a unidimensional 7-item scale with factor loadings ranging from 0·55 to 0·81 (rater group 1) and 0·48 to 0·88 (rater group 2). Cronbach’s alphas of 0·87 and 0·88 indicated good internal consistency irrespective of rater group. A Pearson’s r of 0·80 confirmed acceptable test–retest reliability, and interrater reliability Intraclass Correlation 3 ranging from 0·77 to 0·93 also showed acceptable results. The effect size r of 0·53 plus Cohen’s d of 1·25 indicates the capacity of the scale to detect changes in performance.
Conclusion. Further research is needed to test the English version of the scale and its validity.