Experimenter-Controlled Decisions in the Design and Analysis of Psychological Research.
Abstract
This paper reviews certain statistical rationales and procedures that can be useful to research staff in designing and analyzing research. It consists of three parts. The first deals with the necessity of computing the number of subjects required by a given experiment and presents rationales and procedures. The second part explains the advantages of having equal numbers of subjects in experimental treatment 'conditions' or 'cells,' and shows to what extent bias may affect analysis of unequal cell frequencies. The third part outlines the sometimes drastic effect of making multiple comparisons on the same data and suggests alternate procedures. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1965
- Accession Number
- ADA090784
Entities
People
- Donald Reynolds
Organizations
- Human Resources Research Organization