The relationship between adoption of administrative and technical innovations over time and its impact on organizational performance was studied. A confirmatory analysis of the data from 85 public libraries showed that, over consecutive time periods, changes in the social structure, portrayed by the adoption of administrative innovations, lead to changes in the technical system, portrayed by the adoption of technical innovations. Empirical support was also provided for Daft's (1982) framework for organizational innovation that was found to be effective in separating organizations based on their performance levels.

No abstract

No abstract

It is proposed that shared leadership environments have an effect on six sigma team decision-making as well as the methods used to make decisions. It is also proposed that the consensus decision-making method is the technique that is the most effective and related to the fundamental conditions of a shared leadership environment. Furthermore, it is believed that a model of shared leadership decision-making can be developed from the results of these hypotheses. Research has found that a strong relationship exists between shared leadership and team decision-making principles and approaches. As a six sigma team becomes more involved in the shared leadership model, the methods utilized to make decisions change in such a way that the power is more distributed among members; the team moves from a single individual decision method, towards a shared decision method. Review of the relationship between shared leadership and team decision-making techniques has shown that the consensus approach is the most effective in satisfying the fundamental conditions and requirements of the shared leadership model. A model of decision-making in shared leadership environments is proposed for use by teams to determine the type of decision-making method that should be employed as a team's level of shared leadership increases.

Prediction analysis is an approach to the analysis of cross‐classified, cross‐sectional or longitudinal data. It is a method for predicting events based on specification of predicted relations among qualitative variables and includes techniques for both stating and evaluating those event predictions. This article presents the basic methodology employed in prediction analysis and highlights approaches developed by Hildebrand, Laing, and Rosenthal (6), and by von Eye, Brandtstadter & Rovine, (12, 13). The Hildebrand et al. approach involves measuring prediction success of a set of predictions via a proportionate reduction in error (PRE) measure, known as del . The von Eye et al. approach models the prediction hypothesis via nonstandard log‐linear modeling and evaluates the extent to which the model adequately describes the data. Both approaches to prediction analysis provide a viable, valuable data analysis tool that researchers in the social and behavioral sciences can use in a wide range of research settings.