臨床護理師職業倦怠雙元三因子效度研究

A Validation Study of Bi-factor Structure of Chinese Nurses Burnout with ESEM

本研究採用Moreno-Jime’nez等人編製之護理師職業倦怠量表（Nursing Burnout Scale, NBS）中文簡版為測量工具，以分層隨機抽樣方式選取江蘇省523名臨床護理師為研究對象，以Mplus8.1統計軟體進行資料分析，比較職業倦怠一因子、三因子、高階因子以及雙元因子四模式，應用CFA與ESEM進行參數估計，旨在探討華人臨床護理師職業倦怠因素結構。研究結果顯示，臨床護理師職業倦怠雙元三因子ESEM模式之適配度最佳，護理師職業倦怠除了「情緒衰竭」、「低成就感」與「對人冷漠」等三種特定倦怠因子之外，還蘊含著一個「整體倦怠」因素。唯，整體因子與「對人冷漠」特定因子能夠有效解釋對應觀察變項之變異，而「情緒衰竭」、「低成就感」特定因子只能解釋小部分觀察變項的變異。本研究結果呼應了由Mészáros Ádám、Szabó、Szigeti與Urbán（2014）提出的職業倦怠雙元因子結構，顯示職業倦怠雙元因子模式具有跨文化、跨情境、跨行業的類推性。比起CFA，在參數估計上ESEM提供了更合理精準的雙元因子參數估計值。本研究結果可提升心理計量研究對於護理師職業倦怠多維度潛在結構之認識。並可提示醫院管理者，在為處於危機中之臨床護理師提供干預時應考慮其職業倦怠總分。然而，因「情緒衰竭」與「低成就感」分量表之題項並未完全反應對應之特定因子，因而在實務或研究中需慎用此二分量表因子分。未來研究可進一步探討諸如性別、年資等背景變項在職業倦怠雙元三因子模式的測量恆等性。

This study validated the bi-factor structural construct of the Nursing Burnout Scale (NBS) by using a stratified sample of 523 nurses from Jiang Shu, China, who completed the short version of the NBS survey in Chinese. Mplus software (version 8.1) provided the statistical outputs for all analyses conducted in this study. The findings for four alternative burnout constructs-single-, three-, high-order, and bi-factor models-employing confirmatory factor analysis (CFA) and exploratory structural equation modeling (ESEM) approaches improved our understanding of work burnout among nurses in the hospitals in China. Statistical comparison among a series of nested CFA and ESEM models revealed that a bi-factor model with a global burnout factor and three content-specific factors-emotional exhaustion (EE), depersonalization (DP), and reduced personal accomplishment (PA)-involving the ESEM approach outperformed all other models in terms of the model data fit. Only global burnout factor and the content-specific DP factor explained the considerable unique proportion of variance in the observed scores. The content-specific factors EE and PA only explained a small unique proportion of variance in the observed scores. The findings support the bi-factor structure of the NBS that was originally proposed by Mészáros et al. in Hungary (2014). Moreover, the findings provide insights into understanding the cross-culture and cross-situation factorial consistency of nursing burnout. Compared with CFA, ESEM is a better approach for validating the bi-factor factorial structure. The findings improve our understanding of the complexity of the multidimensional structure of job burnout among nurses. Moreover, the proposed method provides hospital administrators with the total burnout score to ensure timely intervention for the staff who are at risk. However, the EE and PA scores should be interpreted carefully because these scores reflect a very small proportion of the burnout content. Further investigation on invariance measurement based on gender and time will provide insights into the generalizability of the bi-factor nursing burnout structure.