The Impact of Confidence on the Accuracy of Structured Professional and Actuarial Violence Risk Judgments in a Sample of Forensic Psychiatric Patients

Some previous research indicates that confidence affects the accuracy of probabilistic clinical ratings of risk for violence among civil psychiatric inpatients. The current study investigated the impact of confidence on actuarial and structured professional risk assessments, in a forensic psychiatric population, using community violence as the outcome criteria. Raters completed the HCR-20 violence risk assessment scheme for a sample of 100 forensic psychiatric patients. Results showed that accuracy of both actuarial judgments (HCR-20 total scores) and structured professional judgments (of low, moderate, and high risk) were substantially more accurate when raters were more confident about their judgments. Findings suggest that confidence of ratings should be studied as a potentially important mediator of structured professional and actuarial risk judgments.     

Book reviews

Journal of Youth and Adolescence -     

Image analysis of gunshot residue on entry wounds. II--A statistical estimation of firing range.

A statistical investigation of the relationship between firing range and the amount and distribution of gunshot residue (GSR), used automated image analysis (IA) to quantify GSR deposit resulting from firings into pig skin, from distances ranging between contact and 45 cm. Overall, for a Ruger .22 semi-automatic rifle using CCI solid point, high velocity ammunition, the total area of GSR deposit on the skin sections decreased in a non-linear fashion with firing range. More specifically there were significant differences in the amount of GSR deposited from shots fired at contact compared with shots fired from distances between 2.5 and 45 cm; and between shots fired from a distance of 20 cm or less, with shots fired at a distance of 30 cm or more. In addition, GSR particles were heavily concentrated in the wound tract only for contact and close range shots at 2.5 cm, while the particle distribution was more uniform between the wound tract and the skin surfaces for shots fired from distances greater than 2.5 cm. Consequently, for future scientific investigations of gunshot fatalities, once standards have been established for the weapon and ammunition type in question, image analysis quantification of GSR deposited in and around the gunshot wound may be capable of providing a reliable, statistical basis for estimating firing range.     

Prescriptive and Elicitive Approaches to Conflict Resolution: Examples from Papua New Guinea

Negotiation Journal -     

Correlations on radioimmunoassay, fluorescence polarization immunoassay, and enzyme immunoassay of cannabis metabolites with gas chromatography/mass spectrometry analysis of 11-nor-delta 9-tetrahydrocannabinol-9-carboxylic acid in urine specimens.

Results obtained from three commercial immunoassay kits, Abuscreen, TDx, and EMIT, commonly used for the initial test of urine cannabinoids (and metabolites) were correlated with the 11-nor-delta 9-tetrahydrocannabinol-9-carboxylic acid (9-THC-COOH) concentration as determined by GC/MS. Correlation coefficients obtained based on 26 (out of 1359 total sample population) highly relevant samples, are 0.601 and 0.438 for Abuscreen and TDx. Correlation coefficients obtained from a parallel study on a different set of 47 (out of 5070 total sample population) highly relevant specimens are 0.658 and 0.575 for Abuscreen and Emit. The immunoassay concentration levels, that correspond to the commonly used 15 ng/ml GC/MS cutoff value for 9-THC-COOH, as calculated from the regression equations are 82 ng/ml and 75 ng/ml for TDx and EMIT and 120 ng/ml and 72 ng/ml for Abuscreen manufactured at two different time periods. The difference of these calculated corresponding concentrations provides quantitative evidence of the reagent specificity differences.