中文摘要：

通过对从事人力资源管理工作的公务员、软件工程师、网络编辑和报纸广告销售人员四个职业的272名任职者调查数据的层次回归分析，文章发现控制了职业和人口统计学变量的影响之后，工作满意度、情感承诺和工作投入三个工作态度变量对工作技能的重要性和水平评价结果有显著影响。进一步对比发现，工作满意度对技能的重要性和水平评价结果的影响效应较大。该研究对多个职业的分析结果拓展了人们对工作分析结果影响因素的认识，并对未来工作分析研究和实践有重要的启示作用。

英文摘要：

Introduction A growing body of literature has been focusing on the factors affecting incumbents＇ job analysis ratings. The objective of the present study was to explore the effect of different facets of job attitude, including job satisfaction, affeetive commitment and job involvement, on skill ratings by incumbents＇ coming from four occupations after controlling other variables. The present study also aimed to determine which of the above three variables of job attitude would have the strongest effect on skill ratings. Method Two scales from the Occupational Information Network （O＊ NET） measuring both the importance and level domains of job skills were administered to 272 employees from four occupations. They were 50 human resources managers from public sectors, 54 computer software designers, 100 online editors, and 68 newspaper salespeople. Hierarchical regression analysis were use to analyze the survey data. Results The results indicated that, after controlling for the demographic and dummy job variables, job satisfaction, affeetive commitment and job involvement showed significant associations to both the importance and level domains of skill ratings. In particular, job satisfaction had significant effects on the ratings of organizational skills and cognitive skills in the 2 domains, as well as ratings on technical skills in the level domain. Affeetive commitment had significant effects on the ratings of cognitive skills in both the importance and level domains. Job involvement had significant effects on the ratings on organizational and cognitive skills in the 2 domains. It was also found that among the three variables of interest, job satisfaction had the strongest impact on skill ratings in both domains. Conclusion Some implications can be drawn from the present study. First, by studying the effect of job attitude on skill ratings while controlling for other variables and by collecting data from a variety of occupations and positions, such as managers, research and development professiona