Confusion of confidence intervals and credibility intervals in meta-analysis
Article Abstract:
A review of 30 meta-analyses that have been conducted in organizational behavior and human resource management using procedures described by Hunter, Schmidt, and Jackson (1982) suggests that there is confusion regarding the use and interpretation of confidence intervals and credibility intervals. This confusion can lead to conflicting conclusions about the relationships between variables. The most frequent mistake has been the attempt to address the accuracy of the estimate of the mean effect size using 'confidence intervals' based on the corrected standard deviation instead of on the standard error of the mean r or d. The corrected standard deviation should be used to generate a 'credibility interval' to assess the extent to which moderators might account for the unexplained variance in effect sizes. (Reprinted by permission of the publisher.)
Publication Name: Journal of Applied Psychology
Subject: Social sciences
ISSN: 0021-9010
Year: 1990
User Contributions:
Comment about this article or add new information about this topic:
Three Mode PARAFAC Factor Analysis
Article Abstract:
The multivariate nature of applied research milieu is clear. While research uses greater complexity and comprehension of data collection schemes, more advanced methods of analysis will clarify the applied context. This study begins with the elemental research in three mode factor analysis, and looks at definite statistical development in multi mode analysis, known as the parallel factors model (PARAFAC). Substantive PARAFAC analysis is included. The structure of affective reactions to physical activities such as trampolining is analyzed. Ten affective response scales are used and thirty-four young trampoline users were involved. A former assimilation resistance model is confirmed. Tables of orthogonally restrained weight data and graphs of affect rating scales are included.
Publication Name: Journal of Applied Psychology
Subject: Social sciences
ISSN: 0021-9010
Year: 1983
User Contributions:
Comment about this article or add new information about this topic:
Confidence intervals for a quantile contrast: application of the bootstrap
Article Abstract:
Most comparisons of groups studied in applied psychology focus on mean and average group responses; however, in certain situations, the between-group comparison would be more effective if the portions of the two groups compared were nearer one of the two ends of responses. For example, it could be more beneficial to cross-analyze the best or worst examples within two samples. A technique for making best and worst comparisons between two sample groups is developed and explained. The new between-group comparison technique is based on the relatively recent development of the bootstrap principle of nonparametric inference.
Publication Name: Journal of Applied Psychology
Subject: Social sciences
ISSN: 0021-9010
Year: 1986
User Contributions:
Comment about this article or add new information about this topic:
- Abstracts: Models of conflict, negotiation and third party intervention: a review and synthesis. Extending and testing a five factor model of ethical and unethical bargaining tactics: introducing the SINS scale
- Abstracts: Social activism and disability rates in American Indian tribes. Independent and interdependent conceptions of self: an investigation of age, gender, and culture differences in importance and satisfaction ratings
- Abstracts: Recognition of facial stimuli following an intervening task involving the Identi-kit. Verbal, visual, and voice identification of a rape suspect under different levels of illumination