Buy this Article for $10.95

Have a coupon or promotional code? Enter it here:

When you buy this you'll get access to the ePub version, a downloadable PDF, and the ability to print the full article.


instruments, methodology, model testing, nursing practice environment, Nursing Work Index, structural equation modeling



  1. Cummings, Greta G.
  2. Hayduk, Leslie
  3. Estabrooks, Carole A.


Background: The Nursing Work Index (NWI) provided the foundation for three published instruments, each presented as a valid measure of the nursing practice environment. Aiken and Patrician [Aiken, L. H., & Patrician, P. A. (2000). Measuring organizational traits of hospitals: The Revised Nursing Work Index. Nursing Research, 49, 146-153] revised the NWI to the Revised Nursing Work Index (NWI-R), reporting four conceptually derived subscales. Lake [Lake, E. T.(2002). Development of the Practice Environment Scale of the Nursing Work Index. Research in Nursing and Health, 25,176-188] used factor analysis to empirically derive five subscales with a composite overarching practice environment factor. Estabrooks et al. [Estabrooks, C. A., Tourangeau, A. E., Humphrey,C. K., Hesketh, K. L., Giovannetti, P., Thomson, D., et al. (2002). Measuring the hospital practice environment: A Canadian context. Research in Nursing and Health, 25, 256-268] reported a single empirically derived factor solution that represented the nursing practice environment.


Objective: To examine the validity of three instruments (based on the NWI) as measures of the nursing practice environment.


Methods: The measurement models underlying the three instruments were reconstructed from the information provided by each author in published manuscripts and then were estimated using structural equation modeling (SEM), the chi-square test of model fit, and data from the 1998 Canadian Nurse Survey.


Results: Each of the three underlying measurement models was factor analytic in design (multiple indicators of each concept) and failed significantly when compared with the data (Aiken & Patrician; [chi]2 = 939.12, p <.001, df = 50; Lake, [chi]2 = 17,872.73, p <.001, df = 319; Estabrooks et al., [chi]2 = 38,590.29, p <.001). This lack of model fit with the data raises questions about the validity of these instruments as measures of the nursing practice environment.


Discussion: The nursing practice environment is complex and has been examined inadequately by factor-analytic approaches, largely because insufficient attention has been paid to implicit underlying theory. The development and testing of robust theory using powerful research methods available to examine causal relationships in complex theoretical models will advance understanding of constructs such as the nursing practice environment.