Construction And Demonstration Of An Index System Of Knowledge Amount Of Position

For over a century, researches on labor workers' productivity have been basically completed. With the gradual increase of knowledge amount in most positions, it is certain that the value of knowledge in work outputs will be worth more considering. For this reason, a group named " knowledge worker" appears, representing more importance and influence than those labor workers. This trend has marked the knowledge worker productivity problem the biggest challenge in the 21st century s management and also key elements to national competitive forces.To improve the productivity of knowledge works, the knowledge amount of positions must not be avoided. Some index systems for macro information measurement have been established. However, researches in position measurement are just beginning, faced with some fundamental problems. The one to be firstly solved is how to measure the relatively exact amount of knowledge in positions and then separate high knowledge positions out from those lower ones . The reason why different methods should be adopted to raise knowledge worker productivities lies in the position characteristics, which also urges a different method to perform exploration into them. At this moment, there still exists a wide gap between research methods and its great value.Starting from job analysis, this dissertation refers to the available qualitative research findings and summarized the specialties of knowledge work. The highly reliable, commonly used position analysis questionnaire (PAQ) is expanded into an "Employee Productivity Questionnaire" which was used to investigate various positions in different work units.Next an index system including 8 indexes to measure position knowledge amount is constructed, by which we performed classification on those positions by the statistical methods of data clustering and factor analysis. The consistency between demonstration results and common knowledge proves the validity of the index system.For those kinds of positions lies in two extremes of knowledge amount, different zones are covered in job-characteristics charts andmain factor grids. In this way we demonstrated the index dimensional difference of the position characteristics and provided a foundation for further research on knowledge work productivity.The research findings in this dissertation includes:First, it demonstrates the research method of using job analysis to conclude the position characteristics and perform position classification is available and believable.Second, the differences and versatility among positions can be clearly shown in an 8-elements-based index system.Third, in demonstration, the results of data classification and factor analysis coincide with the judgments of common knowledge. Hypothesis by Helton and Strassman can also been proved when exploring the position characteristics. The 8 basic indexes can be further condensed into 2 main factors, which leads to a 2 mensioned grid that categorizes all positions into 4 different groups.In research, decision tree analysis is also assumed as an innovation in future data processing and knowledge discovery work, in order to make out a suitable way to large-scope questionnaire investigation and productivity research.