A Confirmatory Factor Analysis to Support the Construct Validity of the Direct Care Staff Survey Scale: A Measure of Work Environment and Work Effectiveness in Long-Term Care.

Abstract

Background: Efforts are limited in investigating psychometric properties of the Direct Care Staff Survey (DCSS) that measures the nurse practice environment in long-term care. Purpose: This study aimed to determine if the DCSS is consistent with its underlying five-factor model (Impact of Work Environment on Organizational Performance) and evaluate its validity in a long-term care nurse sample. Methods: A national United States sample of long-term care nurses completed the DCSS. SPSS Amos was used to perform a confirmatory factor analysis (CFA). Results: CFA found modest support for a five-factor structure for the Impact of Work Environment on Organizational Performance model, whereas χ²/df exceeded the commonly accepted criteria (χ²/df = 2.47, <3), fit indices that accounted for model complexity did not (comparative fit index = 0.70, not >0.8; Tucker-Lewis index = 0.67, not ≥ 0.95; root-mean-square error of approximation = .08, not <.08). Further analyses indicated that some items may benefit from revision. Conclusion: The CFA provided modest support for the currently accepted five-factor structure of the DCSS. This study suggests that there is still a need to revise the DCSS questionnaire to address the directionality of the relationship between the factors. This revision could facilitate the use of the DCSS to understand and guide improvements in long-term care nurses' work environments and is vital for enhancing staff well-being and patient outcomes.