| Learning Objective #1: n/a | |||
| Learning Objective #2: n/a |
Objective: The study purpose was to test existing and new assertions about the measurement of the nursing practice environment.
Design: We examined the models underlying three published perspectives on the nursing work index using structural equation modeling, and then developed and tested an alternative model based on theoretical assertions about the healthy nursing workplace, using structural equation modeling.
Concepts: The Revised Nursing Work Index (NWI-R) is a survey scale used to measure the nursing work environment. Aiken and Patrician (2000) reported four conceptually derived subscales (autonomy, control over practice setting, nurse/physician relationships and organizational support). Lake (2002) reported five empirically derived subscales (nurse participation in hospital affairs, nursing foundations for quality of care, nurse manager ability, leadership and support, staffing and resource adequacy, and collegial nurse-physician relations) using a factor analytic method. Estabrooks (2002) reported a single factor solution representing the practice environment.
Methods: The Aiken 4-factor, Lake 5-factor and Estabrooks 1-factor models were estimated using the Canadian Hospital Outcomes Survey database. A theoretical model was then developed based on assertions about healthy hospital nursing workplaces and tested as a structural equation model against the same data. This model was analyzed and compared to the three factor models, with emphasis on measurement and the integration of theory and statistical analysis.
Findings: The published factor analytic models failed significantly. Although the structural equation model also failed, its diagnostics showed a much closer fit of its theoretical assertions to the data.
Conclusions: The failure of the published models stands in stark contrast to the reported claims that these scales and subscales are adequate and appropriate measures of the nursing practice environment.
Implications: The most useful advances in developing the concept of the practice environment will result from testing clearly specified causal relationships using powerful methods such as structural equation modeling.
Back to Accepted Posters
Back to 37th Biennial Convention - Scientific Session
Sigma Theta Tau International