Developing Standards for Interpreting Agency Performance: An Exploration of Three Models

Organizational theory has long been concerned with the question of organizational effectiveness.' That organizations have multiple and sometimes conflicting goals is well known.2 Not all these goals focus upon performance in terms of achieving the organization's mission. Building employee cohesion and morale and acquiring resources for growth are examples of goals that do not. Quinn and Cameron have recently suggested that the criteria important in evaluating an organization's effectiveness depend upon its stage in the life cycle.3 They suggest that there are four stages in an organization's life cycle-entrepreneurial, collectivity, formalization and control, and, finally, elaboration. For the formalization and control stage, which is the stage that most stable organizations seem to be in, they suggest that the effectiveness criteria emphasize planning, goal setting, efficiency, productivity, information management, communications, stability, and control.4 This paper explores statistical models that may be useful to an organization during its formalization and control stage. These models may provide useful information as a part of an organization's management control system. Two essential elements of a management control system are: (1) information about how the organization is operating, and (2) standards against which to compare this information to judge how well the organization is operating.5 While developing performance measures and collecting performance data can be a difficult, time consuming, and expensive task, this information is not of much use to managers unless it identifies areas that need corrective action. Identifying problem areas requires comparing operations information to standards or benchmarks. Possible sources of such standards for public sector organizations include an organization's goals, objectives, or targets; standards established by relevant professional associations; the performance of similar organizations; the organization's own historical performance record; and optimal or technically efficient performance levels.6 This paper discusses statistical models that managers can use to develop standards based on three of these sources: