Group sequential designs using both type I and type II error probability spending functions

We propose flexible group sequential designs using type I and type II error probability spending functions. The proposed designs preserve the overall significance level and power and allow the repeated testing to be perloimed at a flexible schedule. Computational methods are described. An example on a mega clinical trial is provided.     

The influence of the organization on clinical practice

Almost all, if not all clinical decisions represent agency policy and organizational imperatives in action. The organization influences and shapes services, problem definition, assessment and intervention, and careers of clients. This paper explores ideas about the organizational context of practice, organizational problems and related issues, and offers suggestions for increasing the clinician's influence and effectiveness.     

Equivalent decision trees and their associated strategy sets

The purpose of this paper is to examine the structure of individual decisions under uncertainty in extensive (i.e., decision-tree) form at a sufficient level of generality to encompass many traditional as well as novel criteria of general choice under uncertainty. Considerations both of structure and of strategic effect arise. In § 2 we define weak and strong forms of strategic equivalence of decision trees and describe partial-normalization procedures by which all economically defined complete pure strategies may be identified for the decider and for chance. In § 3 we examine characteristics of locally randomized, behavior strategies for the decider as the degree of normalization is varied, and we deduce useful properties of the function which specifies an outcome lottery for each behavior strategy of the decider in conjuction with a deterministic or stochastic selection by Chance. In § 4 we list a wide variety of choice criteria which may be invoked within the framework developed here.     

Laud Humphreys: A Pioneer in the Practice of Social Science

The life and work of Laud Humphreys is reviewed in an attempt to assess their effect on American sociology. We focus on the intersections between biography and professional contributions as they emerged from Humphreys'graduate work at Washington University, St. Louis through his years at Pitzer College. The commentary is couched in the theoretical traditions of C. Wright Mills, the sociology of knowledge, and the sociology of sociology.     

Minority Composition,Inequality and the Growth of Municipal Police Forces, 1960–71

Abstract

Recent studies, in finding that community variables such as percentage Black and income inequality have significant effects on police size and expenditures, call into question the findings of earlier research. However, these recent studies are themselves subject to serious methodological limitations and leave open the question of whether it is class or racial conflict which affects the deployment of police resources. In this study we estimate several simultaneous equation models of police size per capita in 1971 for 88 non-Southern cities. Bach of these models holds constant the effects of per capita police size in 1960. Racial composition and racial inequality are found to be the only significant predictors of the growth in the number of police per capita.     

The Social Power of Health-Care Practitioners as Agents of Change

This paper examines dominant sources of social power and influence that affect the degree to which patients adhere to recommended regimens for prevention or treatment and their rate of recovery. Referent power of health-care practitioners, as contrasted with their expert, coercive, reward and legitimate power, is expected to be most effective when internalization of medical recommendations by the patients is essential. The paper explores specific ways that a health-care professional can gain referent power, using weight control treatment as an example. It presents a conceptual framework that provides a rationale for a number of relationship-building techniques and offers evidence that referent power is effective in promoting adherence to treatment and more healthful behavior, and in influencing critical outcome measures such as weight loss. Analysis of the relative efficacy of various methods of building and using referent power and the possible limiting conditions on its effectiveness are also discussed.