DEVELOPMENT OF A GENERAL PREDICTION METHOD FOR TRANSCRIPTION ERROR RATE.

Abstract

This study develops a General Prediction Method (GPM) for estimating the human error rate in a data transcription system. It also indicates the nature of the relationships between error rate and the significant determinants, and it specifies the relative importance of these relationships. Correlation and stepwise regression analyses are performed on data gathered in previous laboratory experiments in obtaining the desired results. The following factors are investigated for their influence on human error rate: Operator Age, Education, and Occupation; Transcription Methods; Code Length, Content, Mix, and Repetition; Code Blocking and Order; Grouping within Codes; Response Format; and Duration of Work Period. The following factors have the greatest influence: Code Length, Content, Transcription Method, and Repetition. The following have less influence: Sex, Age, Occupation, Duration of Work Period, Response Format, Education, Grouping within Codes, Code Blocking and Order. It is recommended that the GPM be validated by application to operational data systems and refinement made in operational settings where those factors found relatively unimportant in this study would be eliminated enabling easier application of the GPM. (Author)

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 1967
Accession Number
AD0659449

Entities

People

  • Anne Melby
  • Bruce N. Mcarthur
  • Richard L. Hawley

Organizations

  • FMC Corporation

Tags

DTIC Thesaurus Topics

  • Computing-Related Activities
  • Data Science
  • Education
  • Information Science
  • Interdisciplinary Science
  • Least Squares Method
  • Mathematical Analysis
  • Mathematics
  • Regression Analysis
  • Statistical Analysis
  • Statistics

Readers

  • Instructional Design and Training Evaluation.
  • Materials Science
  • Theoretical Analysis.