PREDICTING ORGANIZATIONAL EFFECTIVENESS FROM COMMUNICATION ACTIVITIES: A MULTIPLE INDICATOR MODEL

This study examines the effects of certain communication variables (member integration, boundary communication, environmental uncertainty, and system autonomy) on organizational effectiveness. While often discussed in the literature, these variables have not been precisely defined. As a result, a multiple indicator model using maximum likelihood estimates for each variable was created. Interviews with managers from 82 different organizations from a variety of industries are used to test the model. The results suggest that member integration and environmental uncertainty, when the latter is mediated by boundary communication, provide an excellent predictor of organizational effectiveness (R= .64). The results are discussed in light of the proposed theory along with implications for future research using intertypical sampling methods and LISREL analysis.     

The epistemological objection to modal primitivism

Modal primitivists hold that some modal truths are primitively true. They thus seem to face a special epistemological problem: how can primitive modal truths be known? The epistemological objection has not been adequately developed in the literature. I undertake to develop the objection, and then to argue that the best formulation of the epistemological objection targets all realists about modality, rather than the primitivist alone. Furthermore, the moves available to reductionists in response to the objection are also available to primitivists. I conclude by suggesting that extant theories of the epistemology of modality are not sensitive to the question of primitivism versus reductionism.

     

On assessing verbal stimulus hierarchies1

Typical word-association procedures result in verbal response hierarchies for a specific stimulus word. The problem of assessing corresponding stimulus hierarchies is discussed Three approaches are entertained: obtaining associations to all of the words in the dictionary, clerical cross-indexing of existing response-hierarchy data, and obtaining backward associations which are then used in forward association. A comparison of data obtainable by the last two methods is presented.