A multistudy validation of an instrument for nurse job satisfaction

Abstract

<p><strong>Objective:</strong> This study aims to validate an instrument of nurse job satisfaction, the Healthcare Environment Survey (HES), across settings and time through an organizational change.</p> <p><strong>Methods:</strong> This is a retrospective study that consisted of four phases of analysis using factor analysis to test the 59-item HES, which was initially developed in 2001 to assess nurse job satisfaction. A sample of 4,626 nurses was used from 10 unique studies that were conducted from 2006 to 2012 in four health care entities in the United States. Inclusion criteria were that the respondents had to provide direct care to patients and be permanent staff members of the entity under study.</p> <p><strong>Results:</strong> In phase 1, the principal component analysis (PCA) for each of the 10 studies revealed a 10-factor structure of nurse job satisfaction that was consistent across settings and time. Half of the sample from all 10 studies was randomly selected and reexamined in an exploratory factor analysis (EFA) that revealed a nine-factor solution using 50 items with a Kaiser&ndash;Meyer&ndash;Olkin of .96. All items loaded in the same respective factors as those of the previous PCA analyses. The EFA explained 74% of the variance. A confirmatory factor analysis of the other half of the sample confirmed the nine-factor solution, using 50 items, with the root mean square error of approximation of .056, comparative fit index of .919, and standardized root mean square residual of .042.</p> <p><strong>Discussion:</strong> The validation of this 50-item measure of job satisfaction, which is valid across settings and time, will assist in minimizing error while building models of research that include antecedents, correlates, and outcomes of nurse job satisfaction.</p>