Citation

Plotkin RC, Azar S, Twentyman CT, Perri MG. Child Abuse Negl. 1981; 5(4): 449-455.

Copyright

(Copyright © 1981, Elsevier Publishing)

DOI

unavailable

PMID

unavailable

Abstract

The present assessment of the literature pertaining to the causative factors of child abuse was conducted to determine the scientific merits of the existing literature in this field. The reviewing process entailed randomly assigning those articles that presented original data to four reviewers. Each reviewer then evaluated the studies independently using a standard evaluation procedure that addressed quality of methodological design, clarity of exposition, reliability and validity of data source and appropriateness of analysis. It was discovered that the quality of this scientific inquiry was variable and often not satisfying the basic requirements of sound experimental design. In fact, the majority of the papers reviewed presented no original data to support their positions. Those studies incorporating empirical evidence were frequently plagued with methodological and statistical problems. The specific problems discussed were the following: inappropriate comparison group, lack of a priori predictions, insufficient documentation pertaining to the reliability and validity of data sources, lack of adequate demographic descriptors to provide a context and enable future analyses to be performed, omission of informed consent, and illogically derived conclusions. Concrete recommendations were then addressed to each area.