Abstract: Abstract: Presently, there are some limitations on the verification of new effort-reward imbalance (ERI) model as follows. Firstly, covariation between variables may happen when effort, reward and overcommitment were entered at the same level in multiple linear regression analysis in some studies. Secondly, there is little research reporting the comparison of the influence on health outcomes among the three variables. Lastly, the results regarding the moderating effect of overcommitment on the relationship between ERI and health outcomes (e.g., work burnout) are inconsistent in the previous studies. The inconsistency might be because ERI is not the characteristic work stress of the investigated samples. Additionally, female nurses are typical population with high ERI and work burnout in China. This study aimed to validate the new ERI model among nurses. It hypotheses that effort, reward and overcommitment can independently predict work burnout and their prediction roles are different, and that the effort-reward ratio can positively predict work burnout, and that overcommitment can moderate the relationship between the ratio and work burnout. This study randomly recruited 456 female nurses from nine hospitals in Nanjing, Jiangsu. Effort, reward and overcommitment were measured with Chinese version of the effort-reward imbalance scale, and work burnout with Chinese version of Maslach burnout inventory-General Survey (MBI-GS) including three dimensions, emotional exhaustion, cynicism and professional inefficacy. Mplus6.11 and SPSS20.0 were utilized for the test of common method bias and multiple linear regressions that are conducted to examine the independent effects of effort, reward, overcommitment and the effort-reward ratio, and the moderation effect of overcommitment in association with the effort-reward ratio and work burnout. In this study, there was not heavy common method bias between ERI and MBI-GS scales after examining one-factor, two-factor, four-factor and six-factor models where two-factor model set ERI one factor and MBI-GS as another factor, and four-factor model set the two dimensions with bigger correlation coefficient in ERI or MBI-GS scale as one factors, and another dimension as one factor. The results revealed that effort positively predicted emotional exhaustion and cynicism and negatively predicted professional efficacy. Reward negatively predicted cynicism and positively predict professional efficacy. Overcommitment positively predicted emotional exhaustion and cynicism. Among the three factors, effort showed the strongest prediction role and reward did the weakest prediction role. The effort-reward ratio positively predicted emotional exhaustion and cynicism, and negatively predict professional efficacy. Overcommitment moderated the influences of the effort-reward ratio on emotional exhaustion and professional efficacy. Specifically, the effort-reward ratio had stronger prediction role among the nurses with lower overcommitment levels than those with higher overcommitment levels. In conclusion, these results strong support new ERI model.

Key words: effort-reward imbalance, overcommitment, moderating effect, work burnout

摘要： 摘 要 目的：在护士群体中检验付出、回报和过度投入及其交互作用对工作倦怠的影响，验证新三元付出-回报失衡模型。方法：采用问卷法测查了456名女护士的付出、回报、过度投入和工作倦怠。采用分层线性回归分析验证自变量对因变量的影响。结果：付出、过度投入和付出-回报比正向影响情绪耗竭和去个性化，付出和付出-回报比负向影响职业效能，回报负向影响去个性化，正向影响职业效能。付出对工作倦怠的影响最大，其次是过度投入，回报的影响最小。过度投入能调节付出-回报比对情绪耗竭、职业效能的影响。结论：新三元付出-回报失衡模型适用于女护士群体。

关键词: 付出-回报失衡, 过度投入, 调节作用, 工作倦怠