Identification of Critical Variables and Critical Gap Variables in Hospital Nurses' Job Satisfaction During the Dynamic Adjustment Phase of COVID-19 Prevention in China: A Hybrid Machine Learning and Decision Analysis Approach.

PURPOSE: This study aims to analyze the critical variables and gap variables affecting hospital nurses' job satisfaction and propose improvement strategies based on the knowledge domains of nursing decision-makers.

METHODS

This study, conducted between September and October 2022 during the dynamic adjustment phase of COVID-19 prevention and control in China, was based on the McCloskey/Mueller Satisfaction Scale (MMSS) and developed a hybrid machine learning and decision analysis tool model. The random forest (RF) method was used to estimate the importance of each variable in the data, and the importance-performance analysis (IPA) was used to identify critical gap variables and propose improvement strategies.

RESULTS

The RF analysis (OOB error rate = 17.93%) identified "Decision-making" (C30, importance score = 0.053) and "Control-work conditions" (C29, importance score = 0.067) as the most influential factors (critical variables) in determining nurses' job satisfaction. The IPA analysis identified C30 as the most critical gap variable, indicating a significant need to improve nurses' involvement in hospital decision-making processes.

CONCLUSIONS

To improve nurse job satisfaction and retention, hospital decision-makers and nursing departments should implement policies that enhance nurses' involvement in decision-making, particularly those with experience in pandemic-related healthcare challenges. Addressing these factors could foster a more supportive and resilient nursing work environment.