The least understood of complex systems is that which produces educated people. Superficially, many key elements of a manufacturing organization are part of the system: a product, product tolerance limits, methods for assessing product deviation, and a feed-back system for making compensatory adjustments. In layman's terms these elements are: educated youth, educational goals or objectives, performance tests, and a reporting system. Their existence has led to an attempt at applying a wholesale systems framework to education with mixed results. Reasons for the lack of success in this attempt is analysed in this study. It is suggested that system analysts, rather than modeling the total system, should focus on those aspects of education which can be systematized. Such aspects are: resource allocation, scheduling, prediction systems (research and evaluation), and data bank systems. It is theorized that two conditions must be met to enable a systems approach to be successful in an educational application: (1) The designers must exhibit a fine sensitivity to the fact that schools are political entities, and (2) the system must be focused at the level of the individual student. The implications of these assumptions are discussed. (Author)