These lists display a degree of discontinuity, even inconsistency. Hemphill's 'Position elements' mix both 'managerial' and 'specialist' elements and the discontinuity between these and Pheysey's list is notable given that Pheysey's research was based upon Hemphill's original study. Certainly here are early grounds for suspecting that the content of managerial work is not common across levels of management or cultures.
Mintzberg (1973) and Sayles (1964) show greater agreement of substance beneath superficial differences of terminology, although this agreement cuts through their different categories. For example, Sayles' 'Leadership' category subsumes Mintzberg's 'Figurehead', 'Leader' and 'Spokesman' roles and his emphasis upon 'Participation in external work flows' via different types of relationships expands upon Mintzberg's 'Liaison' role, as well as indicating the external character of the 'Entrepreneur', 'Disseminator', 'Disturbance handler' and 'Negotiator' roles (Andrews, 2007, 92).
The major difference between Sayles and Mintzberg is that the latter views managerial work as more self-contained, whereas the former is more concerned to locate managerial work within the context of organizational processes.
Discussion and Analysis
Amid the diversity of evidence, some common findings recur. First, managers perform both specialist/technical and general/administrative work. Second, the latter is sufficiently ill-defined that part of managerial work is determining its own boundaries. Finally, within these fluid boundaries, the following strands are common, if not universal:
Acting as figurehead and leader of an organizational unit
Liaison: the formation and maintenance of contacts
Monitoring, filtering and disseminating information
Allocating resources
Handling disturbances and maintaining work flows
Negotiating
Innovating
Planning
Controlling and directing subordinates
Commanding
Whilst exhibiting striking parallels with the supposedly outdated 'classical principles of management', this evidence takes us further for two reasons. Firstly, it includes certain elements which could not, without stretching a point, be subsumed under any of the 'classical' principles. Secondly, the research studies do offer detailed indications of what these principles may involve. Sayles, Mintzberg, Stewart and Kotter all provide fresh insights and subtleties to the tasks of 'planning', 'co-ordinating' and 'commanding'.
Stewart (1976) also sheds light on the chronological patterns and sources of managerial work by distinguishing the duration of work, time span, recurrence, unexpectedness and, finally, source of initiation. These give additional dimensions to the constituents of managerial work: what managers do has different durations, rhythms, degrees of uncertainty and origins.
It is evident from table II that the different studies and, indeed, sometimes the same study, point to different ways of conceptualizing the constituent features of managerial work - in particular the difference between observable activities which constitute the performance of the job, and implied or reported tasks which represent expected or intended outcomes. Hemphill (1959), Pheysey (1972) and, to a lesser degree, Sayles (1964) suggest that activities and tasks are empirically intertwined or, at least, do not attempt to separate them. Mintzberg (1975) does distinguish between activities and managerial roles - which constitute 'tasks' as defined above, given that they are developed by asking why a manager undertook a particular activity. However, as Mintzberg describes them, the 'Interpersonal' and 'Decisional' ...