Karl Weick, "Educational organizations as loosely coupled systems", Administrative Science Quarterly, 21 (1976), 1-9 (part).
In this paper Weick uses the US educational system as an example of how loosely coupled systems are both prevalent and important for organizational function. Understanding an organization as a loose coupling of actors, rewards, and technology may help better explain how organizations adapt to their environments and survive amidst uncertainties.
Weick observes that manifests of loosely coupled systems often are:
While these manifests appear negative, they actually may help the organization by:
In general, loosely coupled systems probably are cheaper to coordinate, but are very difficult to systematically change.
Karl comments that most reseachers make the assumption that one can "understand" an organization by examining the formal structures, goals, and activities of an organization. Yet clearly there is another part of organizations that is informal, chaotic, yet somehow productive, adaptable, and crudely organizing.
He proposes to create a new language and an alternative way to look at organizations that is sensitive to the "soft" side of organizations.
Loose coupling "conveys the image that coupled events are responsive, but that each event has it's individual identity and the coupling can vary over time'. It also suggests that you can break many organizations into largely self-functioning subsystems, and loose coupling is really the "glue" that holds them together.
Two coupling elements often discussed are technical couplings (between technology, task, role) that are task-induced and authority couplings (positions, offices, rewards, sanctions) that somehow hold the organization together. However, in educational systems these couplings are prominent, so what holds the educational system together?
It's clear that many couplings could be considered in schools ( teachers-materials, voters-schoolboard, process-outcome, teacher-teacher, etc.). Any model to explain these couplings would also have to acknowledge the temporality and variability in these couplings, which really complicates things.
In educational systems, often people consider systems "loosely coupled" in areas like the following:
* excess amount of resouces relative to demands "slack time"
* situations where several means will produce the same ends
* highly connected networks and yet influence is slow or weakens quickly on spreading
* lack of coordination, or dampened coordination through system
* absence of regulations
* planned unresponsiveness
* actual causal independence
and so forth.
The Functions (Advantages) of A Loosely Coupled System
Weick believes that loosely coupled systems, though messy, have valid functions within organizations:
1. May allow parts of the organization to persist (despite changes in environment ). For example, terms of office allow people to continue ruling even when the populace is against particular actions.
2. May provide a sensitive sensing mechanism. Like sand as a better wind sensor than rocks, a loosely coupled system may detect changes in their environment better than a large, tightly run organization. Of course, a system can be too sensitive and respond to "fads" or minor changes.
3. Good system for localized adaptation. Local groups can adapt to their part of the environment without changing the entire system. The opposite, standardization, may be too restrictive.
4. May allow more novel solutions and mutations to occur than a tightly coupled system. LCS provide more diversity to adapt to changing environmental situations (those the loose coupling may also be a barrier for diffusion of good ideas).
5. LCS are less affected by a breakdown in part of the system (islolate trouble spots and keep from spreading). However, a LCS may make repair of a subsystem more difficult.
6. More room available for self determination by actors (like teachers, classes, etc.)
7. A LCS should be less expensive to run because it takes time and money to coordinate people. However, it's hard to systematically change a LCS.
These seven points could be used a starting point of hypothesis to study LCS.