Who should stand next to the suspect? Problems in the assessment of lineup fairness
Article Abstract:
A common procedure for assessing the fairness of a lineup is to give a verbal description of the perpetrator to people who did not witness the incident and ask them to select the likely perpetrator from the lineup. If people who never saw the perpetrator nonetheless make the "right choice" significantly more often than chance, the implication is that the lineup is unfairly suggestive. Little is known, however, about the factors that might bias this mock witness procedure. Two such biasing factors were examined in this study: the arrangement of photos in the lineup and the diagnosticity of the description. The results suggest that placing the target between 2 low-similarity foils increased the likelihood that he would be chosen, but only when the verbal description contained few diagnostic features. Implications for applied researchers and the construction of the lineups are discussed. (Reprinted by permission of the publisher.)
Publication Name: Journal of Applied Psychology
Subject: Social sciences
ISSN: 0021-9010
Year: 1995
User Contributions:
Comment about this article or add new information about this topic:
Estimation accuracy of Thomas's likelihood-based procedure of meta-analysis: a Monte Carlo simulation
Article Abstract:
This Monte Carlo study tested the accuracy of the likelihood-based vaidity generalization procedure proposed by Thomas (1990) for estimating the number of population correlations and their respective sizes and proportions in a mixed population. The simulation results showed that the Thomas procedure was also accurate in estimating sizes and proportions of the population correlations when the correct number of population correlation was identified. However, its accuracy in estimating the actual number of population correlations decreased when they were close together and when the number of population correlations increased. The power of the procedure in identifying the existence of populations correlations of magnitude zero and its accuracy in estimating means and standard deviations of population correlations are discussed. (Reprinted by permission of the publisher.)
Publication Name: Journal of Applied Psychology
Subject: Social sciences
ISSN: 0021-9010
Year: 1992
User Contributions:
Comment about this article or add new information about this topic:
- Abstracts: Evaluating office decentralization of a financial center: the case of Hong Kong. Expansion of the frontier and city of freedom
- Abstracts: Education and equality: the battle for school funding reform. Unequal funding: teaching Cadillac kids with Chevy dollars
- Abstracts: Reengineering in practice: where are the people? Where is the learning? Challenging "resistance to change."
- Abstracts: Sexuality education and sexual rights. Sexual rights are human rights
- Abstracts: The life history of the emergent organization in times of disaster. Social organization, profit cycles, and statewide land use controls: welcome to Oregon - enjoy your visit