Interaction and decision making rely heavily on group processes. - Rensis Likert

The management system of an organization must have compatible component parts if it is to function effectively. This conclusion has a very important implication; experiments in organizations must involve internally consistent changes. The traditional atomistic research design is not appropriate for experiments involving organizational theory or management systems. Every aspect of a management system is related to every other part and interacts with it.
The results obtained by altering a single variable or procedure while keeping all others the same usually will yield quite different results from those obtained when that variable is changed along with simultaneous and compatible changes in all other aspects of the management system. The true influence of altering one aspect of the system cannot be determined by varying it and it alone... In experiments involving organizational theory and management systems, therefore, a systems approach must be used. The organic integrity of each system must be maintained while experimental variations are being made.

Conflict is viewed as the active striving for one's own preferred outcome which, if attained, precludes the attainment by others of their own preferred outcome, thereby producing hostility.

A number of statistical assumptions are made in the application of his (Thurstone's) attitude scale e.g. that the scale values of the statements are independent of the attitude distribution of the readers who sort the statements assumptions which as Thurstone points out, cannot always be verified. The method is more over laborious. It seems legitimate to enquire whether it actually does its work better than the simple scales which may be employed and the same breath to ask also whether it is not possible to construct equally reliable scales without making unnecessary statistical assumptions.