Toxicologic findings in the USS Iowa disaster.

The toxicologic results from the 47 victims of the explosion on the USS Iowa are presented. Good correlation between carboxyhemoglobin saturations and cause of death was found. There were no correlations between blood cyanide concentrations and causes of death. Volatile analysis suggested postmortem ethanol production rather than antemortem ethanol ingestion. No drugs except nicotine were detected in any of the victims.     

The referral decision scale with mentally disordered inmates

Correctional facilities are increasingly burdened with mentally disordered offenders and are faced with the formidable task of effectively screening inmates for major mental disorders and suicide potential. Teplin and Swartz (1989) developed the Referral Decision Scale (RDS) as a rapid screen for schizophrenia and major mood disorders in jail populations. However, more recent research by Hart, Roesch, Corrado, and Cox (1993) has called into question the effectiveness of the RDS. In this study, we examined the convergent and discriminant validity of the RDS with the Schedule of Affective Disorders and Schizophrenia—Change Version (SADS-C) and the Personality Assessment Inventory (PAI) on a sample of 108 mentally disordered inmates. The RDS provided moderate evidence of convergent validity but failed to demonstrate acceptable discriminant validity, because of high intercorrelations on RDS subscales (heterotrait-monomethod coefficients). In comparison, the SADS-C yielded similar evidence of convergent validity and excellent discriminant validity.The authors would like to thank the Research Opportunities Program, University of North Texas for its grant support and Captain James Moore for his active cooperation.Tarrant County Mental Health and Mental Retardation Services.     

Systems Changes Associated with Criminal Justice Treatment Networks

This article examines the ability of service networks to improve substance abuse treatment access and service delivery to adult, substance-abusing female offenders. In fiscal year 1995, the Center for Substance Abuse Treatment (part of the U.S. Department of Health and Human Services's Substance Abuse and Mental Health Services Administration) funded four demonstration projects to establish Criminal Justice Treatment Networks to improve substance abuse treatment access and service delivery to adult female offenders while they were involved in the criminal justice system. This study seeks to determine whether the networks promoted systemic change in the community agencies and organizations where they are housed. This article is based on a series of annual site visits, reports, and regular communication with network staff and local evaluators. The findings indicate that networks can improve service delivery in criminal justice and substance abuse treatment systems.     

The Limits of Ecological Inference: The Case of Split-Ticket Voting

We examine the limits of ecological inference methods by focusing on the case of split-ticket voting. Burden and Kimball (1998) report that, by using the King estimation procedure for inferring individual-level behavior from aggregate data, they are the first to produce accurate estimates of split-ticket voting rates in congressional districts. However, a closer examination of their data reveals that a satisfactory analysis of this problem is more complex than may initially appear. We show that the estimation technique is highly suspect in general and especially unhelpful with their particular data.