positions are often eliminated, and their tasks (such as housekeeping) added to those of existing jobs. The supervisory positions are also reduced, and their tasks are dispersed to the group as a whole. When employees are cross-trained to handle many different tasks, their skill sets shift to include a much broader set of knowledge, skills, and abilities. The hierarchy of jobs based on increasing skill within an area gives way to greater commonalities in status and more informal authority structures.
In the context of such teams, exactly how new employees learn the skills for this cross-functional team is less obvious than in the old workplace, where unskilled workers could be assigned to unskilled jobs and then slowly learn new skills and positions. There may be less mobility in the new team model, particularly when jobs are organized according to principles of scientific management, which has broadened individual jobs by eliminating narrow, hierarchical occupational structures. Employees in these new cross-functional teams may be much less likely to identify with a skill and the job defined around it, such as crafts (e.g., "machinist") and more likely to see their identity as a member of a specific group of employees. Variations in the nature of teams can lead to very different work structures and, in turn, different job and occupation descriptions. These variations further increase the heterogeneity in the workplace.
Much of what we know about teams comes from experience with production workers in manufacturing, on which the majority of the research has been done (Osterman, 1996; Hunter, 1998a; Batt, 1999a). Many of the benefits of teamwork appear to be applicable elsewhere (e.g., Cohen and Bailey, 1997; Batt, 1999a), and the use of teams may well continue to expand. The reasons for this include: the increasing complexity of certain types of work, which makes it difficult for individuals to perform it on their own, the tendency for organizations to attempt to flatten their structures by reducing their levels of management, and the related apparent trend to provide employees greater opportunity for decision making at lower organization levels. Support for the latter observation can be found in the WorkTrends™ data summarized in Table 2.3. Respondents in all occupational categories except sales reported somewhat greater participation in decision making in 1996 than in 1990 ("In my company, employees are